RE: "shouted down", "shot down", apologies

RE: "shouted down", "shot down", apologies

[John Wiersba]

OK, given the "new spirit of cooperation" expressed in cgf's email, I'd like
to be in a position to help, too, in the sense of "submit patches" rather
than "find bugs, suggest new features".  But there's primarily one thing
stopping me:  a test environment.  I'm currently ignorant of how to
effectively use cvs and I don't have much burning desire to add csv to my
personal toolkit right now.  

So, if I want be able to rebuild all the cygwin packages, can I do that from
source downloaded with setup.exe?  Can someone recommend a convenient way of
building a "test cygwin" from that source which can be switched to (via a
change to cygwin.bat) to try out changes?  Or do I really have to go the cvs
route and work with the latest/greatest bleeding-edge packages?  If csv is
the only/best way, does someone have a cookbook which will allow me to setup
a test environment, refresh the source, build everything, make a change,
test it out, submit a patch -- all while keeping a working cygwin
environment built off of setup.exe's download?

-- John Wiersba

> -----Original Message-----
> From: Christopher Faylor [ mailto:cgf@redhat.com ]
> Sent: Wednesday, June 27, 2001 4:17 PM
> To: cygwin@cygwin.com
> Subject: "shouted down", "shot down", apologies
> 
> 
> I have been concerned by two recent messages where people have felt
> that their ideas have been "shouted down" or "shot down".
> 
> That bothers me.  It bothers me because I assume that most, if not
> all of the negative perception undoubtedly came from me.
> 
> I do have some strong opinions on how some things should be done.  For
> instance, I think that overloading a FAQ with excessive information is
> counter productive.
> 
> I also feel that the majority of "newbie" requests here do 
> not come from
> people who have exhaustively studied available documentation.
> 
> So, filling the FAQ with non-frequently asked questions does 
> not seem like
> the way to go to me.  It seems like it will make the FAQ 
> harder to navigate
> and will make it easier for people to miss things.
> 
> Telling people that the way to use google is to type something like:
> 
> http://www.google.com/search?q=cygwin+ssh&btnG=Google+Search
> 
> does not make sense to me.
> 
> Updating the documentation *does* make sense to me.
> 
> Some recent email of mine may have made it sound like I am an 
> inflexible
> bastard.  I regret sending it.
> 
> I'm open to new ideas but I sometimes need to be convinced.  And, even
> when convinced, it does not necessarily follow that I will now make
> it my life's mission to carry out the new ideas.
> 
> I've said that repetition is important, so I'll repeat it one 
> more time:
> If you want to see something change, don't "suggest".  Don't "it seems
> to me".  Don't "It would be nice".
> 
> Please reorient your thinking from "This is what they should do" to
> "This is what I can do".
> 
> If I have dropped the ball on someone volunteering or if I have rudely
> shot down your offer to help then I sincerely apologize.  I know that
> my attempts at humor have sometimes been interpreted as rudeness.  I
> know that sometimes I get impatient with ignorance (you can ask my
> family about this trait), especially intractable ignorance.
> 
> Regardless, I have no real excuse.  I am sometimes 
> exasperated and mean.
> I hope that it is clear that I am doing what I'm doing 
> because I want to
> help.  In some cases, I'm even doing things that I come close to
> detesting, like maintaining gcc or make.  I do this because I 
> know that
> it is important to people even though it is really not my specialty.
> 
> And, I also enjoy running a project like Cygwin.  I think that the net
> release of Cygwin has improved dramatically in the last 
> couple of years.
> That is because I've lobbied for changes inside of Red Hat 
> and solicited
> active maintainers outside of Red Hat.  And, I've encouraged the
> development of the cygwin installer.
> 
> There is still lots and lots and lots^10 of room for improvement.  I
> would like to improve the documentation.  I would really like 
> to expand
> the cygwin test suite.  There are still problems with cygwin signals
> and the cygwin spawn command.  setup.exe could stand all sorts of
> improvement.
> 
> I actually have a tendency to just see all of the negatives in cygwin.
> I have to keep reminding myself that people are using it successfully
> every day.  Most of them don't care that zip stores full MS-DOS paths
> or that spawn(_P_NOWAIT) doesn't work on non-cygwin programs.
> 
> Anyway, if someone has volunteered and I have dropped the ball, please
> ping me again.  I'll try to rectify my mistake in not acting on your
> offer.
> 
> If someone has suggested an idea and didn't appreciate my response,
> then I also apologize.  I'll try to do better in the future.
> 
> (Although, I will probably still try to be "humorous" from time to
> time.  Be warned.)
> 
> cgf
> 
> --
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting:         http://cygwin.com/bugs.html
> Documentation:         http://cygwin.com/docs.html
> FAQ:                   http://cygwin.com/faq/
> 

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Wed, Jun 27, 2001 at 04:34:30PM -0400, John Wiersba wrote:
>OK, given the "new spirit of cooperation" expressed in cgf's email, I'd
>like to be in a position to help, too, in the sense of "submit patches"
>rather than "find bugs, suggest new features".  But there's primarily
>one thing stopping me: a test environment.  I'm currently ignorant of
>how to effectively use cvs and I don't have much burning desire to add
>csv to my personal toolkit right now.

Well, just to backslide a little: http://cygwin.com/contrib.html ,
http://cygwin.com/cvs.html .

Or, more generically http://cygwin.com/ .

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Norman Vine]

John Wiersba writes:

> If csv is the only/best way, does someone have a cookbook which will 
>allow me to setup a test environment, 

The way I do this < and I am sure it is not the best but it works for me >
is to first create a directory to work in I'll call it $CYGWIN_SRC

place this shell script in that directory

======= cut here =======
#! /bin/sh
echo "(Logging in to anoncvs@anoncvs.cygnus.com)"
echo "Checking out Cygwin"
echo "CVS password: anoncvs"
export CVSROOT=:pserver:anoncvs@anoncvs.cygnus.com:/cvs/src
cvs login
cvs -z3 checkout winsup 2>&1 | tee cvs_co.log
./MakeCygwin.py
====== cut here =====

and place this python script in this same directory
http://home.cape.com/nhv/files/cygwin/MakeCygwin.py

>refresh the source, build everything, 

execute the above shell script and a tarball of the
Cygwin CVS files will be placed in this same directory
also under this directory you will have a ./usr directory 
containing the files in this tarball

>make a change, test it out, submit a patch -- 

I now just install this tarball by 
% cd /
% tar -xzvf PATH_TO_FRESH_TAR_BALL

then closing all applications using the Cygwin DLL
and copying the new cygwin1.dll from $CYGWIN_SRC / usr / bin / 
to $CYG_DRIVE / bin using a WIN32 tool < explorer, copy ect >

start using your new Cygwin

>all while keeping a working cygwin
>environment built off of setup.exe's download?

If for some reason I do not want to keep this version of Cygwin  
I either reinstall an earlier version that I have made using the same
techniques.  OR use the cygwin setup.exe program to reinstall
the latest official Cygwin distribution.

This method assumes that you have installed Python

FWIW
It normally takes less then 20 minutes for me to update and 
rebuild Cygwin using this method on a PIII 733 over a dialup.
It will take longer the first time in that CVS has to get all the
files initially.

I am sure there are better ways to do this
but until I discover them this works for me :-))

Comments and/or  Suggestions on how to do this solicited.

Cheers

Norman Vine


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Brian Keener]

Christopher Faylor wrote:
> Well, just to backslide a little: http://cygwin.com/contrib.html ,
> http://cygwin.com/cvs.html .
> 
> Or, more generically http://cygwin.com/ .
>
Well,

I am probably going to jump into this now without intending to but this really 
does not answer his question - sorry Chris and I am not aiming this at you 
Chris - This is simply an observation. He asked:  

>> So, if I want be able to rebuild all the cygwin packages, can I do that from
>> source downloaded with setup.exe?  Can someone recommend a convenient way of
>> building a "test cygwin" from that source which can be switched to (via a
>> change to cygwin.bat) to try out changes?  Or do I really have to go the cvs
>> route and work with the latest/greatest bleeding-edge packages?  If csv is
>> the only/best way, does someone have a cookbook which will allow me to setup
>> a test environment, refresh the source, build everything, make a change,
>> test it out, submit a patch -- all while keeping a working cygwin
>> environment built off of setup.exe's download?

As someone who is trying to learn C++ and never really worked with it much and 
as a contributor to the Cygwin project by contributing to setup.exe these are 
questions that I and many others have asked before at various times.  I have 
read the above links before and probably missed where the answers to John's 
questions were located but as far as I can tell there isn't a lot to get 
someone going on contributing to Cygwin.  Or let me say there is a not a lot if 
you are inexperienced in cvs, cygwin, C++, gcc, make as they are used in this 
environment. I have essentially pieced together from various sources (you Chris 
and Earnie and Dj and whole lot of others) and reading the man pages and asking 
very pointed questions just enough to be able to make changes, compile and test 
them and then submit them.  

And while I have improved some of my techniques and even have some scripts to 
do some of the updates and such for me - there is no "cookbook" which is what 
he asked about.  I still cringe when I do an update from cvs and then have to 
do a configure and make because I usually run into a load of problems that 
takes me several days to get around.  Of course the fact I am on Win95 as many 
have pointed out and have no clue what I am doing does not help.  I also have 
never used the testsuite and have no idea how to. I have never installed a 
cygwin I built from sources and have no idea how to although I build it 
regularly since it is part of my build script when I make changes to setup.

While the web pages and the FAQ do tell you how to get the source and how to 
send patches it really does not provide a cookbook.  I realize that we all like 
self starters and the major contributors do not like to have to answer these 
simplistic questions --- but those of us who are trying to learn and to help 
some times need a boost and a few pointers down the right path. 

If I step on anyones toes here I am sorry and if I have overlooked things on 
the web pages I am doubly sorry - for me (cause I've been missing out) and for 
everyone else for having to put up with me asking.  Food for thought for all of 
us - for some of us it is just as hard to bring ourselves to ask the questions 
as it is for the rest to have to read them over and over, but sometimes if you 
do not know what it is you're looking for and/or how to ask for it - it's 
mighty tough to find in a web page or faq.

Gonna go hide now :-)

bk






--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Greg Smith]

This is what I do, especially when I'm behind a firewall and
can't get out on the cvs port.

(1) I get the latest snapshot via ftp from a cygwin mirror site
    and put it in the \cygwin directory
(2) I bring up a cygwin window and do:
    rm -Rf src
    rm -Rf obj
    mkdir src
    mkdir obj
    cd src
    tar jzvf ../cygwin-src-yyyymmdd.tar.bz2
(3) cd /obj
    /src/configure --prefix=/tmp/cygwin && make && make install
    and exit the cygwin window
(4) from a windows command prompt I then do
    copy \cygwin\tmp\cygwin\bin\* \cygwin\bin\*
(5) Fire up a new cygwin window and run `cygcheck -s' to verify
    the snapshot level and build date (which is near the bottom
    of the display).

If there are patches that I want, then I can usually save them
from an attachment in a cygwin or cygwin-patches post, usually
in my \cygwin directory.  Usually the patches hit src/winsup/cygwin.
So, before step (3) I do
    cd /src/winsup/cygwin
    patch < /whatever.patch

Greg

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Wed, Jun 27, 2001 at 05:39:13PM -0400, Brian Keener wrote:
>Christopher Faylor wrote:
>> Well, just to backslide a little: http://cygwin.com/contrib.html ,
>> http://cygwin.com/cvs.html .
>> 
>> Or, more generically http://cygwin.com/ .
>>
>Well,
>
>I am probably going to jump into this now without intending to but this really 
>does not answer his question - sorry Chris and I am not aiming this at you 
>Chris - This is simply an observation. He asked:  
>
>>> So, if I want be able to rebuild all the cygwin packages, can I do that from
>>> source downloaded with setup.exe?  Can someone recommend a convenient way of
>>> building a "test cygwin" from that source which can be switched to (via a
>>> change to cygwin.bat) to try out changes?  Or do I really have to go the cvs
>>> route and work with the latest/greatest bleeding-edge packages?

>>>If csv is the only/best way, does someone have a cookbook which will
>>>allow me to setup a test environment, refresh the source, build
>>>everything, make a change, test it out, submit a patch -- all while
>>>keeping a working cygwin environment built off of setup.exe's download?

I was responding to this part.  It is at least a start for doing things.
I'm sorry that this wasn't obvious.

However, if it is not adequate, then it should be improved, right?

This is similar to the "I already read the FAQ" post.  If this was
inadequate then we need to know how so that it can be improved.  We
don't know if the original poster already read this and found it
inadequate or if he was completely unaware of its existence.

I guess this wasn't obvious from my response.  I promised to try
to be more polite.  I didn't promise that I would not be terse.

>As someone who is trying to learn C++ and never really worked with it
>much and as a contributor to the Cygwin project by contributing to
>setup.exe these are questions that I and many others have asked before
>at various times.  I have read the above links before and probably
>missed where the answers to John's questions were located but as far as
>I can tell there isn't a lot to get someone going on contributing to
>Cygwin.

Ok.  Then this is a problem.  How would you like to fix it?  I did spend
some time a while ago trying to augment the information there but this
is obviously old hat to me and you are saying that it is inadequate.

(You did check the pages before responding, right?)

I would appreciate update to the web pages that you find confusing.

Checkout instructions are below.

The Catch 22 is, of course, that the web pages are available under CVS.

cvs -d :pserver:anoncvs@sourceware.cygnus.com:/cvs/cygwin login
cvs -d :pserver:anoncvs@sourceware.cygnus.com:/cvs/cygwin co htdocs

The directory "htdocs" will contain the web pages.  If you post updates
here, I'll see that they get added to the web site.

When you make a change, the command "cvs diff -p" will produce the
differences.

>Or let me say there is a not a lot if you are inexperienced in cvs,
>cygwin, C++, gcc, make as they are used in this environment.  I have
>essentially pieced together from various sources (you Chris and Earnie
>and Dj and whole lot of others) and reading the man pages and asking
>very pointed questions just enough to be able to make changes, compile
>and test them and then submit them.

I guess the problem that I have with all of this is you almost never see
anyone in the gcc, or gdb projects professing that cvs checkout and building
to be a major obstacle.  I don't know what is so different about cygwin.

This project seems to attract more "newbies".  Maybe the "oldbies" are all
using linux and disdaining Windows.

>And while I have improved some of my techniques and even have some scripts to 
>do some of the updates and such for me - there is no "cookbook" which is what 
>he asked about.

IMO, There shouldn't really need to be any cookbook.

It should be as simple as following the instructions followed on:

http://cygwin.com/cvs.html

to check things out and then following the instructions in the FAQ under:

How do I rebuild the tools on my NT box?

to do the build.  (I forgot to mention the FAQ previously)

That is what I and many others do every day.  Once you have the tools
checked out and configured, it should be this simple:

bash$ cd /src/winsup
bash$ cvs update
bash$ cd /bld/i686-pc-cygwin/winsup
bash$ make

Why do you need a build script for this?

>I still cringe when I do an update from cvs and then have to do a
>configure and make because I usually run into a load of problems that
>takes me several days to get around.

Brian, I am sorry that you are having problems but I don't appreciate
this mystification of the whole process.  It is not that hard.  There
are occasional problems that are caused by people (like me) making
changes to things but the configure/make system changes only very
rarely.  Most of the time a simple "make clean; make all" fixes
the problem.

Very occasionally, you may have to wipe out your build directory and
reconfigure (I've never had to do this but some people seem to think it
fixes things).

I can see that people could have problems setting up their original
environment.  The documentation could be improved there.  Once
the environment is set up, however, it should be pretty easy to keep
it going.

>Of course the fact I am on Win95 as many have pointed out and have no
>clue what I am doing does not help.

I guess it is possible that Win95 is the source of your problem.  I
do recall that you had strange problems building.  I hate to say this
but you seemed to be the only one who had these problems.

Some people have problems with initial setup but, once they are beyond
this, they usually are able to use "cvs update" and "make" from then on.

>I also have never used the testsuite and have no idea how to.  I have
>never installed a cygwin I built from sources and have no idea how to
>although I build it regularly since it is part of my build script when
>I make changes to setup.

Then perhaps someone will volunteer how to do this.

You seem to be mystifying the process, again, however.  Testing cygwin
should be accomplished by typing "make check".  Installing cygwin is
tricky because you can't overwrite a running process.

So, what do you do?  You use the MS-DOS copy command to install it
to your bin directory.  Or, you drag it there with the GUI.

The new cygwin dll is named new-cygwin1.dll because you don't want
to create a cygwin1.dll while you're using the cygwin1.dll.

And, then once you've figured all of this out, you can offer these
pearls of wisdom to the mailing list as a modification to the FAQ or the
user documentation or the web page.

>While the web pages and the FAQ do tell you how to get the source and how to 
>send patches it really does not provide a cookbook.  I realize that we all like 
>self starters and the major contributors do not like to have to answer these 
>simplistic questions --- but those of us who are trying to learn and to help 
>some times need a boost and a few pointers down the right path. 

The major contributors answer these questions all of the time.  That is
how you learned.

>If I step on anyones toes here I am sorry and if I have overlooked things on 
>the web pages I am doubly sorry - for me (cause I've been missing out) and for 
>everyone else for having to put up with me asking.  Food for thought for all of 
>us - for some of us it is just as hard to bring ourselves to ask the questions 
>as it is for the rest to have to read them over and over, but sometimes if you 
>do not know what it is you're looking for and/or how to ask for it - it's 
>mighty tough to find in a web page or faq.
>
>Gonna go hide now :-)

This is another variation of the "This is what should be done" post.

I am exhorting people to think about how they can improve things.  The
major contributors are professing burn-out.

If you think that the documentation is inadequate, then improve it.

Or, I guess, you can offer your observations here, and maybe someone
else will be willing to improve it.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Greg Smith]

Missed a command in step (2), let me try again

This is what I do, especially when I'm behind a firewall and
can't get out on the cvs port.

(1) I get the latest snapshot via ftp from a cygwin mirror site
    and put it in the \cygwin directory
(2) I bring up a cygwin window and do:
    cd /
    rm -Rf src
    rm -Rf obj
    mkdir src
    mkdir obj
    cd src
    tar jzvf ../cygwin-src-yyyymmdd.tar.bz2
(3) cd /obj
    /src/configure --prefix=/tmp/cygwin && make && make install
    and exit the cygwin window
(4) from a windows command prompt I then do
    copy \cygwin\tmp\cygwin\bin\* \cygwin\bin\*
(5) Fire up a new cygwin window and run `cygcheck -s' to verify
    the snapshot level and build date (which is near the bottom
    of the display).

If there are patches that I want, then I can usually save them
from an attachment in a cygwin or cygwin-patches post, usually
in my \cygwin directory.  Usually the patches hit src/winsup/cygwin.
So, before step (3) I do
    cd /src/winsup/cygwin
    patch < /whatever.patch

Greg

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

CVS helpers

[Michael A. Chase]

CVS does tend to be a bit intimidating.  I have created some bash functions
and a couple Perl scripts that help me manage updating and making the Cygwin
source tree.  I don't think they are as flexible as the more experienced
users would need, but they have made it easier for me to keep up with the
CVS changes.

Would anyone be interested in them?
--
Mac :})
** I normally forward private questions to the appropriate mail list. **
Give a hobbit a fish and he eats fish for a day.
Give a hobbit a ring and he eats fish for an age.
----- Original Message -----
From: "Christopher Faylor" <cgf@redhat.com>
To: <cygwin@cygwin.com>
Sent: Wednesday, June 27, 2001 13:46
Subject: Re: "shouted down", "shot down", apologies


> On Wed, Jun 27, 2001 at 04:34:30PM -0400, John Wiersba wrote:
> >OK, given the "new spirit of cooperation" expressed in cgf's email, I'd
> >like to be in a position to help, too, in the sense of "submit patches"
> >rather than "find bugs, suggest new features".  But there's primarily
> >one thing stopping me: a test environment.  I'm currently ignorant of
> >how to effectively use cvs and I don't have much burning desire to add
> >csv to my personal toolkit right now.
>
> Well, just to backslide a little: http://cygwin.com/contrib.html ,
> http://cygwin.com/cvs.html .
>
> Or, more generically http://cygwin.com/ .



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

If this method works for everyone, could someone provide a nice
HTML version of these instructions?

Anyone want to help me out in maintaining the web pages?

I forget if someone volunteered to do this already.  Feel free to
(gasp!) send me private email if you want to volunteer for something.
Just this once!

cgf

On Wed, Jun 27, 2001 at 06:33:50PM -0400, Greg Smith wrote:
>Missed a command in step (2), let me try again
>
>This is what I do, especially when I'm behind a firewall and
>can't get out on the cvs port.
>
>(1) I get the latest snapshot via ftp from a cygwin mirror site
>    and put it in the \cygwin directory
>(2) I bring up a cygwin window and do:
>    cd /
>    rm -Rf src
>    rm -Rf obj
>    mkdir src
>    mkdir obj
>    cd src
>    tar jzvf ../cygwin-src-yyyymmdd.tar.bz2
>(3) cd /obj
>    /src/configure --prefix=/tmp/cygwin && make && make install
>    and exit the cygwin window
>(4) from a windows command prompt I then do
>    copy \cygwin\tmp\cygwin\bin\* \cygwin\bin\*
>(5) Fire up a new cygwin window and run `cygcheck -s' to verify
>    the snapshot level and build date (which is near the bottom
>    of the display).
>
>If there are patches that I want, then I can usually save them
>from an attachment in a cygwin or cygwin-patches post, usually
>in my \cygwin directory.  Usually the patches hit src/winsup/cygwin.
>So, before step (3) I do
>    cd /src/winsup/cygwin
>    patch < /whatever.patch
>
>Greg
>
>--
>Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting:         http://cygwin.com/bugs.html
>Documentation:         http://cygwin.com/docs.html
>FAQ:                   http://cygwin.com/faq/

-- 
cgf@cygnus.com                        Red Hat, Inc.
http://sources.redhat.com/            http://www.redhat.com/

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Greg Smith]

Sheesh, another typo, the `tar' in step two shoulda been

   tar xjvf ../cygwin-src-yyyymmdd.tar.bz2

Though I do sometimes claim that I deliberately introduce
errors so as not make things too easy  ;-)

Oh, and the `v' isn't really needed.  I just sorta like to see
what is going on... probably for the same reason I distrust
scripts and prefer to enter the commands myself.

So I'm an anachronism, wanna make something of it ;-P

Greg Smith wrote:
>
> Missed a command in step (2), let me try again
>
> This is what I do, especially when I'm behind a firewall and
> can't get out on the cvs port.
>
> (1) I get the latest snapshot via ftp from a cygwin mirror site
>     and put it in the \cygwin directory
> (2) I bring up a cygwin window and do:
>     cd /
>     rm -Rf src
>     rm -Rf obj
>     mkdir src
>     mkdir obj
>     cd src
>     tar jzvf ../cygwin-src-yyyymmdd.tar.bz2
> (3) cd /obj
>     /src/configure --prefix=/tmp/cygwin && make && make install
>     and exit the cygwin window
> (4) from a windows command prompt I then do
>     copy \cygwin\tmp\cygwin\bin\* \cygwin\bin\*
> (5) Fire up a new cygwin window and run `cygcheck -s' to verify
>     the snapshot level and build date (which is near the bottom
>     of the display).
>
> If there are patches that I want, then I can usually save them
> from an attachment in a cygwin or cygwin-patches post, usually
> in my \cygwin directory.  Usually the patches hit src/winsup/cygwin.
> So, before step (3) I do
>     cd /src/winsup/cygwin
>     patch < /whatever.patch
>
> Greg
> >
> > --
> > Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> > Bug reporting:         http://cygwin.com/bugs.html
> > Documentation:         http://cygwin.com/docs.html
> > FAQ:                   http://cygwin.com/faq/

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

building cygwin with python script

[Jason Dufair]

This is brilliant.  Nice cookbook.  Worked out of the box.  Exactly what 
is needed, IMO.  The calculated timings at the end are schweet.  I'd 
like to suggest to the FAQ maintainer that this replace the section in 
the FAQ on "How do I build on NT?".  FAQ maintainer?  If it's OK, I'll 
do the changes.  I'd need CVS access since I assume anoncvs can't commit 
changes.  Maybe we can put the Python script on the sources.redhat.com 
site somewhere?

On a side note, should my new cygwin1.dll be ~5MB instead of ~645KB?  Is 
this just a side effect of the current snapshot?

Norman Vine wrote:

>John Wiersba writes:
>
>>If csv is the only/best way, does someone have a cookbook which will 
>>allow me to setup a test environment, 
>>
>
>The way I do this < and I am sure it is not the best but it works for me >
>is to first create a directory to work in I'll call it $CYGWIN_SRC
>
>place this shell script in that directory
>
>======= cut here =======
>#! /bin/sh
>echo "(Logging in to anoncvs@anoncvs.cygnus.com)"
>echo "Checking out Cygwin"
>echo "CVS password: anoncvs"
>export CVSROOT=:pserver:anoncvs@anoncvs.cygnus.com:/cvs/src
>cvs login
>cvs -z3 checkout winsup 2>&1 | tee cvs_co.log
>./MakeCygwin.py
>====== cut here =====
>
>and place this python script in this same directory
> http://home.cape.com/nhv/files/cygwin/MakeCygwin.py
>
>>refresh the source, build everything, 
>>
>
>execute the above shell script and a tarball of the
>Cygwin CVS files will be placed in this same directory
>also under this directory you will have a ./usr directory 
>containing the files in this tarball
>
>>make a change, test it out, submit a patch -- 
>>
>
>I now just install this tarball by 
>% cd /
>% tar -xzvf PATH_TO_FRESH_TAR_BALL
>
>then closing all applications using the Cygwin DLL
>and copying the new cygwin1.dll from $CYGWIN_SRC / usr / bin / 
>to $CYG_DRIVE / bin using a WIN32 tool < explorer, copy ect >
>
>start using your new Cygwin
>
>>all while keeping a working cygwin
>>environment built off of setup.exe's download?
>>
>
>If for some reason I do not want to keep this version of Cygwin  
>I either reinstall an earlier version that I have made using the same
>techniques.  OR use the cygwin setup.exe program to reinstall
>the latest official Cygwin distribution.
>
>This method assumes that you have installed Python
>
>FWIW
>It normally takes less then 20 minutes for me to update and 
>rebuild Cygwin using this method on a PIII 733 over a dialup.
>It will take longer the first time in that CVS has to get all the
>files initially.
>
>I am sure there are better ways to do this
>but until I discover them this works for me :-))
>
>Comments and/or  Suggestions on how to do this solicited.
>
>Cheers
>
>Norman Vine
>
>
>--
>Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting:         http://cygwin.com/bugs.html
>Documentation:         http://cygwin.com/docs.html
>FAQ:                   http://cygwin.com/faq/
>

-- 
Jason Dufair - jase@dufair.org
http://www.dufair.org/
"Being in politics is like being a football coach.
You have to be smart enough to understand the game,
and dumb enough to think it's important."
-- Eugene McCarthy




--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: building cygwin with python script

[Larry Hall (RFK Partners, Inc)]

At 06:59 PM 6/27/2001, Jason Dufair wrote:
>On a side note, should my new cygwin1.dll be ~5MB instead of ~645KB?  Is this just a side effect of the current snapshot?

Yes, because it has debug info in it.  If you don't want this, strip it
or build a release version.


Larry Hall                              lhall@rfk.com
RFK Partners, Inc.                      http://www.rfk.com
118 Washington Street                   (508) 893-9779 - RFK Office
Holliston, MA 01746                     (508) 893-9889 - FAX



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: building cygwin with python script

[Robert Collins]

----- Original Message -----
From: "Jason Dufair" <jase@dufair.org>


> do the changes.  I'd need CVS access since I assume anoncvs can't commit
> changes.  Maybe we can put the Python script on the sources.redhat.com
> site somewhere?

The standard process is for a patch to be sent to cygwin-patches@cygwin.com,
and then Chris commits the code after review. Few of the net maintainers
commit code directly to CVS.

> On a side note, should my new cygwin1.dll be ~5MB instead of ~645KB?  Is
> this just a side effect of the current snapshot?

It's a debug version. strip the .dll and you'll have a ~645Kb file.

Rob


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Brian Keener]

Christopher Faylor wrote:
> >>>If csv is the only/best way, does someone have a cookbook which will
> >>>allow me to setup a test environment, refresh the source, build
> >>>everything, make a change, test it out, submit a patch -- all while
> >>>keeping a working cygwin environment built off of setup.exe's download?
> 
> I was responding to this part.  It is at least a start for doing things.
> I'm sorry that this wasn't obvious.
> 
> However, if it is not adequate, then it should be improved, right?
> 
> This is similar to the "I already read the FAQ" post.  If this was
> inadequate then we need to know how so that it can be improved.  We
> don't know if the original poster already read this and found it
> inadequate or if he was completely unaware of its existence.
> 
> I guess this wasn't obvious from my response.  I promised to try
> to be more polite.  I didn't promise that I would not be terse.

You are correct - I had not thought of your original response in that light.  But 
you are right - we really do not know if he read or even knew of the FAQ.  That 
goes towards what I said - part of my understanding of the whole process comes from 
asking very pointed questions and not being to generic (which I do have a tendency 
to do as you can tell).

> >As someone who is trying to learn C++ and never really worked with it
> >much and as a contributor to the Cygwin project by contributing to
> >setup.exe these are questions that I and many others have asked before
> >at various times.  I have read the above links before and probably
> >missed where the answers to John's questions were located but as far as
> >I can tell there isn't a lot to get someone going on contributing to
> >Cygwin.
> 
> Ok.  Then this is a problem.  How would you like to fix it?  I did spend
> some time a while ago trying to augment the information there but this
> is obviously old hat to me and you are saying that it is inadequate.

All I was really trying to say is there is some general get started information, 
but for a real neophyte (and that is really the key) they end up aksing themselves 
where do I go from here.  That was probably my biggest problem - I like to think 
you won't find many people who will download cygwin, decide they like it and can 
use it to teach themselves C++, and want to contribute - and try to do it all at 
the same time from the beginning.  That was my problem.

> I guess the problem that I have with all of this is you almost never see
> anyone in the gcc, or gdb projects professing that cvs checkout and building
> to be a major obstacle.  I don't know what is so different about cygwin.

See my previous response but I will add here that you are referring to folks 
writing compilers and debuggers - not people simply trying to add something to 
setup.exe and learn C++ at the same time.  I realize that learning and contributing 
at the same time is probably not the best way - but I don't like having to think up 
projects to help me learn - setup was a good place to learn.

> This project seems to attract more "newbies".  Maybe the "oldbies" are all
> using linux and disdaining Windows.

I think this is a lot of it - I know it is in my case.

> >And while I have improved some of my techniques and even have some scripts to 
> >do some of the updates and such for me - there is no "cookbook" which is what 
> >he asked about.
> 
> IMO, There shouldn't really need to be any cookbook.

Just some step by step instructions - sometimes it's better to be over detailed 
than to leave information out.  I can always skip what I already know - but if it's 
not there I have to ask someone.

> 
> It should be as simple as following the instructions followed on:
> 
> http://cygwin.com/cvs.html
> 
> to check things out and then following the instructions in the FAQ under:
> 
> How do I rebuild the tools on my NT box?
> 
> to do the build.  (I forgot to mention the FAQ previously)
> 
> That is what I and many others do every day.  Once you have the tools
> checked out and configured, it should be this simple:
> 
> bash$ cd /src/winsup
> bash$ cvs update
> bash$ cd /bld/i686-pc-cygwin/winsup
> bash$ make
> 
> Why do you need a build script for this?

Your right - it shouldn't be tougher than that and for the most part for me now - 
it isn't.  As to the script(s) and why? - memory - I tried to make sure I had the 
documentation tools so my cvs update script handles cygwin, docbook-tools, and 
texi2html and the build script did a configure and the makes for me.  When you are 
learning it is a lot to remember.

> Brian, I am sorry that you are having problems but I don't appreciate
> this mystification of the whole process.  It is not that hard.  There

You are right - that is an over mystification and it has gotten easier.

> Very occasionally, you may have to wipe out your build directory and
> reconfigure (I've never had to do this but some people seem to think it
> fixes things).

I was having to do it a lot but I forgot that Earnie helped me fgind this one - I 
was doing a configure everytime I updated from cvs and then would do a make (which 
did sometime foul up big time) and then I was doing a configure and a make clean 
and it would leave stuff hanging around and Earnie pointed out I should make clean 
before I did the configure if I was going to do a configure.  That seemed to help a 
lot of my problems - sorry.

> I can see that people could have problems setting up their original
> environment.  The documentation could be improved there.  Once
> the environment is set up, however, it should be pretty easy to keep
> it going.

I would agree whole-heartedly - as my system has moved more to proper configuration 
and my knowledge has gotten deeper my system has stabilized as far as making 
changes and building.  Setup is really a key issue.

> 
> >Of course the fact I am on Win95 as many have pointed out and have no
> >clue what I am doing does not help.
> 
> I guess it is possible that Win95 is the source of your problem.  I
> do recall that you had strange problems building.  I hate to say this
> but you seemed to be the only one who had these problems.

That is usually the case with me - if someone will have all the problems it will be 
me.  But it may also be the number of people that are actually using this on Win95 
as opposed to NT.  This also goes back to my nature - I want it all - if there are 
5 options on something I want all 5.  I wasn't happy just learning C++ or working 
with cygwin - I wanted to learn C++, work with cygwin and contribute to setup too.

> You seem to be mystifying the process, again, however.  Testing cygwin
> should be accomplished by typing "make check".  Installing cygwin is
> tricky because you can't overwrite a running process.
> 
> So, what do you do?  You use the MS-DOS copy command to install it
> to your bin directory.  Or, you drag it there with the GUI.
> 
> The new cygwin dll is named new-cygwin1.dll because you don't want
> to create a cygwin1.dll while you're using the cygwin1.dll.

Thank you.

> And, then once you've figured all of this out, you can offer these
> pearls of wisdom to the mailing list as a modification to the FAQ or the
> user documentation or the web page.

Never miss a chance do you :-)

> >While the web pages and the FAQ do tell you how to get the source and how to 
> >send patches it really does not provide a cookbook.  I realize that we all like 
> >self starters and the major contributors do not like to have to answer these 
> >simplistic questions --- but those of us who are trying to learn and to help 
> >some times need a boost and a few pointers down the right path. 
> 
> The major contributors answer these questions all of the time.  That is
> how you learned.

I know and I thank you and all the others - but admit it - it is sometimes with 
mumbling under your breath because to you "oldies" it is so second nature and you 
might wait for us to ask a second time before answering.  But that goes along as 
well with the discussion we had here a while back about you making your presence 
less known (which by the way - does not appear to be in your nature ;-)) Sometimes 
you have to force us to look elsewhere for the answers and/or to dig deeper by not 
responding and for the most part we all understand that as well.

> This is another variation of the "This is what should be done" post.
> 
> I am exhorting people to think about how they can improve things.  The
> major contributors are professing burn-out.
> 
> If you think that the documentation is inadequate, then improve it.
> 
> Or, I guess, you can offer your observations here, and maybe someone
> else will be willing to improve it.

Yeah I guess it is a "You guys need to change this post" and I am sorry for that. 
I thought I was only offering an observation as food for thought and ended up 
offering way to much criticism and mystification.  I apologize and will conclude by 
saying - I really like Cygwin, I enjoy working with it and on the setup program and 
learning C++ in the process and I appreciate all the help everyone has given me.  
If I stepped on toes or fell into the "me to" or "*you* need to" trap - again I am 
sorry. 



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: building cygwin with python script

[Christopher Faylor]

On Wed, Jun 27, 2001 at 05:59:47PM -0500, Jason Dufair wrote:
>This is brilliant.  Nice cookbook.  Worked out of the box.  Exactly
>what is needed, IMO.  The calculated timings at the end are schweet.
>I'd like to suggest to the FAQ maintainer that this replace the section
>in the FAQ on "How do I build on NT?".  FAQ maintainer?  If it's OK,
>I'll do the changes.  I'd need CVS access since I assume anoncvs can't
>commit changes.  Maybe we can put the Python script on the
>sources.redhat.com site somewhere?

Ok.  I don't mean to shoot you down but I think that the FAQ should
really concentrate on how things are done.  So, it should focus on the
steps involved in building cygwin.

I would rather that people were familiar with what they were doing
rather than blindly relying on a python script.

I guess it can't hurt to make the python script available but I want
to be 100% certain that someone is willing to maintain it because you
KNOW that we'll start getting bug reports and "would be nice" reports
for it.

So, I would prefer that the cygwin FAQ contain rebuild instructions with
maybe a link to the python script as a "use at your own risk".

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Wed, Jun 27, 2001 at 08:32:52PM -0400, Brian Keener wrote:
>Just some step by step instructions - sometimes it's better to be over
>detailed than to leave information out.  I can always skip what I
>already know - but if it's not there I have to ask someone.

Yes.  That's why I'm probably not the best person to be writing the
instructions.

>>Very occasionally, you may have to wipe out your build directory and
>>reconfigure (I've never had to do this but some people seem to think it
>>fixes things).
>
>I was having to do it a lot but I forgot that Earnie helped me fgind
>this one - I was doing a configure everytime I updated from cvs and
>then would do a make (which did sometime foul up big time) and then I
>was doing a configure and a make clean and it would leave stuff hanging
>around and Earnie pointed out I should make clean before I did the
>configure if I was going to do a configure.  That seemed to help a lot
>of my problems - sorry.

I remember that now.  We were all confused.  You weren't doing anything
wrong.  What you were doing should have worked.

One thing that happens is that occasionally the "powers that be" decide
to make a change in the top-level configure and make process and this
screws up cygwin.  I usually live only in the winsup directory and do
all of my builds there.  It is safer.

The sad fact is that the people developing gcc, gdb, and binutils who
make occasional changes to the very top level configure of the tree
which includes gcc, gdb, binutils, ld, and winsup do not always take
"winsup" into account.  And, stuff can be broken for a while until
someone figures out what is wrong.

So the morale to the story is don't build from the top-level.

cd /build/i686-pc-cygwin/winsup
make

not

cd /build
make

...despite what some web-based instructions may say.

This *is* confusing and I was wrong to act like it was simple.

>>I guess it is possible that Win95 is the source of your problem.  I do
>>recall that you had strange problems building.  I hate to say this but
>>you seemed to be the only one who had these problems.
>
>That is usually the case with me - if someone will have all the
>problems it will be me.

Sounds like you should be in QE.

>> And, then once you've figured all of this out, you can offer these
>> pearls of wisdom to the mailing list as a modification to the FAQ or the
>> user documentation or the web page.
>
>Never miss a chance do you :-)

repeti... snore...

wha?

Oh yeah, repetition is the key.

>I know and I thank you and all the others - but admit it - it is
>sometimes with mumbling under your breath because to you "oldies" it is
>so second nature and you might wait for us to ask a second time before
>answering.  But that goes along as well with the discussion we had here
>a while back about you making your presence less known (which by the
>way - does not appear to be in your nature ;-))

I wasn't very present last week, was I?  Of course, I was out of town,
but still...  :-)

I have been trying not to answer questions when I think that others will step
in.  When it comes to philosophy questions, though, then I feel that I have
to.  Or, maybe I don't.  Hmm.

>Sometimes you have to force us to look elsewhere for the answers and/or
>to dig deeper by not responding and for the most part we all understand
>that as well.

Ok.  Point taken.  If I go silent for a while I hope that people understand
that I'm not being rude.  :-)

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Wed, Jun 27, 2001 at 08:51:32PM -0400, Christopher Faylor wrote:
>So the morale to the story is don't build from the top-level.
        moral

Sigh.  Spelling abilities diminish the oldre I gte.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: building cygwin with python script

[Norman Vine]

Christopher Faylor writes:
>
>I guess it can't hurt to make the python script available but I want
>to be 100% certain that someone is willing to maintain it because you
>KNOW that we'll start getting bug reports and "would be nice" reports
>for it.
>
>So, I would prefer that the cygwin FAQ contain rebuild 
>instructions with
>maybe a link to the python script as a "use at your own risk".

I will try to keep the python script at its current location but
in general anything on my website is only transitory at best.

I am delighted if it helps anybody.

However it is very long winded and IMHO not the best
example of how this could be done or of Python use for 
that matter. 

I really have little interest in 'officially' maintaining it and
if it is used IMO should definitely have "use at yout own risk" 
stamped all over it.

If someone else wants to give the script a home and maintain it,  
please feel free to do so,  but as I said in my original post 
"I am sure there are better ways to do this"

FWIW
I originally wrote this just as an exercise in seeing how
to go about doing this kind of thing not ever intending
for it to be 'exposed' to public scrutiny but then when it 
seemed as if I might be able to help put and end to a VERY
LONG thread by proffering a 'cookbook' method to building
the cygwin dll  .............

Cheers

Norman


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Michael A. Chase]

----- Original Message -----
From: "Christopher Faylor" <cgf@redhat.com>
To: <cygwin@cygwin.com>
Sent: Wednesday, June 27, 2001 17:51
Subject: Re: "shouted down", "shot down", apologies


> On Wed, Jun 27, 2001 at 08:32:52PM -0400, Brian Keener wrote:
> >Just some step by step instructions - sometimes it's better to be over
> >detailed than to leave information out.  I can always skip what I
> >already know - but if it's not there I have to ask someone.
>
> Yes.  That's why I'm probably not the best person to be writing the
> instructions.

That's one of the areas I'd like to contribute in, once my company's
disclaimer shakes loose.  It will be quite a while before I'm comfortable
with gdb or the other troubleshooting tools.

> One thing that happens is that occasionally the "powers that be" decide
> to make a change in the top-level configure and make process and this
> screws up cygwin.  I usually live only in the winsup directory and do
> all of my builds there.  It is safer.

On the other hand, until some poor schmuck does build from the top level,
the problem goes undetected.  I've been building from the base directory
partly because I figure if _I_ can't compile it won't be a disaster for
anyone and it might give earlier warning of a problem.

> Ok.  Point taken.  If I go silent for a while I hope that people
understand
> that I'm not being rude.  :-)

That would be a bit curt.  Even for you.  :}b



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Wed, Jun 27, 2001 at 08:05:03PM -0700, Michael A. Chase wrote:
>----- Original Message -----
>From: "Christopher Faylor" <cgf@redhat.com>
>To: <cygwin@cygwin.com>
>Sent: Wednesday, June 27, 2001 17:51
>Subject: Re: "shouted down", "shot down", apologies
>
>
>> On Wed, Jun 27, 2001 at 08:32:52PM -0400, Brian Keener wrote:
>> >Just some step by step instructions - sometimes it's better to be over
>> >detailed than to leave information out.  I can always skip what I
>> >already know - but if it's not there I have to ask someone.
>>
>> Yes.  That's why I'm probably not the best person to be writing the
>> instructions.
>
>That's one of the areas I'd like to contribute in, once my company's
>disclaimer shakes loose.  It will be quite a while before I'm comfortable
>with gdb or the other troubleshooting tools.
>
>> One thing that happens is that occasionally the "powers that be" decide
>> to make a change in the top-level configure and make process and this
>> screws up cygwin.  I usually live only in the winsup directory and do
>> all of my builds there.  It is safer.
>
>On the other hand, until some poor schmuck does build from the top level,
>the problem goes undetected.

Not for long.  Someone from Red Hat will invariably find it.  I probably
build from the top level once a week or so, too.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Vince Rice]

Chris,

Somewhere in this thread you expressed some wonder/frustration at why the
gcc/gdb projects don't generate the kind of "newbie" requests Cygwin (I just
tried to find it and it's nowhere to be found; I tend to lose things at this
time of night).  I have seen this expressed in your messages countless times
over the three years I've been lurking here.  And I've always wondered at
your wonder <g>.

GCC and gdb are tools for developers; you don't use them unless you're
developing something.  However, Cygwin is a tool for users; it's just as
easy (easier IMHBAO) to *use* the Cygwin tools (defined as the GNU tools,
not gcc/etc.) as it is to develop on Cygwin.  One doesn't have to be a
developer to use Unix.  One can be a raw user who has been thrust into a
Unix environment for one of a thousand reasons, or a college student (or
mid-life crisis male who I bear absolutely no resemblance to) who wants to
learn how to *use* Unix, not how to *develop* in Unix.  Your outlook of
Cygwin is the ability to develop/port Unix programs to Windows.  That is
Cygwin's reason for existence, from your (and Redhat's) and much of the
list's perspective.  However, I believe many (most?) users of Cygwin could
not care less about that.  Their view of Cygwin is the ability to *run* Unix
in Windows.

As such, we have no skills to debug.  Our approach to this mailing list is
the same as to a Word newsgroup or a game newsgroup or whatever.  We want to
know how to *use* the product, and when something doesn't work we come ask
questions.  We don't always RTM, but that's what users do (or don't do as
the case may be).  We have no desire to use the source, we have no ability
to use the source, and many of us wouldn't know source if it hit them in the
head (which with you around is a distinct possibility <bg>).

The fact that Cygwin allows developers to port Unix programs to Windows is
immaterial.  My introduction to Cygwin was from looking for Unix utilities
to run on Windows.  I have stuck around for three years because the tools
continue to get better and better, and many times, even though I run from
4NT instead of bash, I can barely tell the difference between a Windows
prompt and a Unix prompt, because I can do durn near anything I can with
Cygwin on my PC that I can do on my clients' AIX boxes.

However, because I'm in the computer business, because I've written code in
my distant past (but not C/C++), I lurk rather than ask questions, partly
because of the attitude around here towards those who want to know how to
*use* the product and partly because I believe in your philosophy of fishing
for myself, and if I don't have time to get the pole I try not to make
someone else do it either.

However, although I've practiced that here, I'm not completely convinced
that that is always a good thing.  Community building involves interaction.
Sometimes the way to get started is by asking questions.  If someone asks me
a question that will take me five minutes to answer but would take them two
minutes to look up themselves, then that's a waste of my time and I'll tell
them so.  But if someone asks me a question that will take them three hours
to figure out themselves when I can answer it in thirty seconds, I think
it's a little extreme to tell them to "use the source" when I can save them
three hours by just answering the question.   Whether they can figure it out
for themselves is beside the point:  I'm helping them be more productive,
just as someone helped me be more productive when I started out.

I didn't really mean to get into that.  My real point was to address your
original question -- Cygwin attracts at least as many users as developers (I
suspect far more), so they ask user questions, not developer ones, and they
think like users (hey Joe, how do I change the font on a footnote?), not
developers (hey Joe, where's the manual for that graphics card, I want to
re-write the device driver).

Vince


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Michael Erdely]

Documentation is an area I'd like to contribute in.  I've seen people on
the list (especially Earnie) get up in arms when someone suggests a
non-RedHat site (except for Delorie) with documentation.

I've corresponded with several people that find some of the
documentation difficult to follow for a newbie.  As Vince writes in
another email ( http://cygwin.com/ml/cygwin/2001-06/msg01773.html ), not
everyone that uses Cygwin knows Unix or knows C++.  Sometimes they have
a need for a specific tool.  I have a friend that mainly uses bash and
textutils from Cygwin so that the logs of some scripts are accessible to
other Windows users without having to install Samba on his *nix machines
or having some fancy file transfer scripts.  I know many people (who hit
my web site) are just interested in using a free SSH server with
Windows.  I don't think having this need should require an understanding
of the code involved.  Especially since you don't have to know C++ in
order to get sshd running.

So, my contribution (and I state so at the bottom on my pages) is my
whole section of http://mike.erdelynet.com/cygwin.asp .  It's mainly
documentation to help a non-*nix person get Cygwin installed, configured
and getting sshd running.  I'd be more than happy to contribute to more
documentation or have my documentation taken by someone who can "polish"
it.

Anyway, let me know... ;-)

-ME

-----Original Message-----
From: cygwin-owner@sources.redhat.com
[ mailto:cygwin-owner@sources.redhat.com ] On Behalf Of Michael A. Chase
Sent: Wednesday, June 27, 2001 11:05 PM
To: cygwin@cygwin.com
Subject: Re: "shouted down", "shot down", apologies


----- Original Message -----
From: "Christopher Faylor" <cgf@redhat.com>
To: <cygwin@cygwin.com>
Sent: Wednesday, June 27, 2001 17:51
Subject: Re: "shouted down", "shot down", apologies


> On Wed, Jun 27, 2001 at 08:32:52PM -0400, Brian Keener wrote:
> >Just some step by step instructions - sometimes it's better to be
over
> >detailed than to leave information out.  I can always skip what I
> >already know - but if it's not there I have to ask someone.
>
> Yes.  That's why I'm probably not the best person to be writing the
> instructions.

That's one of the areas I'd like to contribute in, once my company's
disclaimer shakes loose.  It will be quite a while before I'm
comfortable
with gdb or the other troubleshooting tools.

<snip>


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Thu, Jun 28, 2001 at 02:31:36AM -0500, Vince Rice wrote:
>Chris,
>
>Somewhere in this thread you expressed some wonder/frustration at why the
>gcc/gdb projects don't generate the kind of "newbie" requests Cygwin (I just
>tried to find it and it's nowhere to be found; I tend to lose things at this
>time of night).  I have seen this expressed in your messages countless times
>over the three years I've been lurking here.  And I've always wondered at
>your wonder <g>.

I used gcc and gdb as examples.  I could easily have used things like:
"bttv", "ssh", or "zsh".

I follow, to some degree, the discussions in those projects.  There are
few complaints about how hard it is to check things out using cvs or
build the tools.

>GCC and gdb are tools for developers; you don't use them unless you're
>developing something.  However, Cygwin is a tool for users; it's just as
>easy (easier IMHBAO) to *use* the Cygwin tools (defined as the GNU tools,

Well, if you are interested in building Cygwin, you should be some level
of developer, right?  If not, you are really due for an exercise in
frustration?  "What are all these semicolons, anyway?" "Why do all of
the files have this funny .cc extension?"

I won't comment on the rest of your message except to say that I rarely,
if ever, say "use the source" when I know the answer to the question.

I say that when I don't know and when *I* would have to look things up.

You're welcome to continue to ask "newbie" questions as frequently as
you like.  That will not stop me (or others, I presume) from pointing
people to references or suggesting the source when I don't know the
answer myself.

I don't work for anyone here.  I am not obligated to look things up to
make your life easier.  It is that simple.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Larry Hall (RFK Partners, Inc)]

At 10:22 AM 6/28/2001, Christopher Faylor wrote:
>I won't comment on the rest of your message except to say that I rarely,
>if ever, say "use the source" when I know the answer to the question.
>
>I say that when I don't know and when *I* would have to look things up.
>
>You're welcome to continue to ask "newbie" questions as frequently as
>you like.  That will not stop me (or others, I presume) from pointing
>people to references or suggesting the source when I don't know the
>answer myself.
>
>I don't work for anyone here.  I am not obligated to look things up to
>make your life easier.  It is that simple.

[long - those uninterested in this thread want to hit delete now! :-) ]

This is exactly my approach as well.  I have to say that I'm a bit 
dismayed that folks contributing to this and the "blunt tools" thread
have mentioned dissatisfaction with what seemed to me to be such a 
straight-forward and logical approach.  When responding to queries on
this list, I've always followed these simple rules:

   1. If I know the question is an FAQ, I point to the entry there (*very*
      rarely do I just point at the FAQ without the exact entry).  
      Generally I feel there's little benefit to restating what's in the 
      FAQ.  It just doesn't seem to be a good use of my time.  If its 
      inadequate in some way, we'll hear about it and make the appropriate 
      change (which seems to me as it should be).

   2. If I kn
ow something specific about the subject, I respond with it.
      Sometimes this means I have to ask a question or two before I'm 
      sure what's been tried already and whether the poster is aware of 
      a previous discussion on the subject.  That all seems like part of
      the process to me and I don't begrudge people for it.

   3. If I know that this subject has come up before and has been 
      discussed but don't remember allot of details, I point to the 
      email archives.  In this case, I don't point to a specific message,
      although I do occasionally offer a search key that I think might
      help find the discussion I recall.  I don't spend my time looking 
      up the exact archive entry or entries that I'm recalling.  I don't
      even promise that the stuff I'm remembering is even helpful (though
      that's my intent and what I'm hoping for!)  I'm just providing 
      potential source of information that may prove useful.  It may not 
      too.  If it doesn't or its too hard to find, I expect the original 
      poster will query the list again with an update of the things tried 
      and the results.  If there's no success at this point, I sometimes
      see if there's something more specific I can find myself and post 
      that if so.

   4. If the question being answered is specific and detailed enough
      that an inspection of the source is likely to be the only path to
      a useful answer (barring someone else who has been in the source,
      knows the answer, and will subsequently offer it), I *suggest* 
      looking at the source.  I do this when its clear someone is a 
      developer or has mentioned they are working with some other source.
      I mention it if I'm not sure whether the person is a developer or 
      not, usually pointing out that it is an option if they're up to 
      it.  I tend not to mention it if the person states that they have
      no experience reading/writing code.  Generally, I don't feel 
      obli
gated to go inspect the source to answer someone else's question,
      although there are exceptions or times I do it anyway.

   5. If I know nothing about the subject, I keep my mouth shut.

I've used all five of these modes in the past on this list and seen them
work, at least on some occasions, exactly as I expected them to.  We've
heard back from people who've had a hard time with an FAQ entry.  We've
heard from people who say they've searched the archives but turned up
nothing.  We've heard back from people saying they're not capable of 
looking at the source for one reason or another.  To me, all of this 
seems reasonable dialog in the course of trying to help someone with a 
problem.  I've always felt that providing some information, be it direct
or a pointer to something which could be helpful is better than no answer
at all (indeed, this list has more than once in the past been berated
for *not* responding in some way to a post!)  However, it troubles me 
that some in the recent discussions have pointed to the replies with 
references to previous discussions and the FAQ as "non-answers" (I'm 
using this term generally now although I know it was a specific member 
of the previous discussions that first offered it up and it may have 
applied in that case to a problem with the specific set of tools in use 
at the time.  I think it categorizes a general sentiment I got from 
reading these threads though).  The impression I'm left with is that 
there is at least some people on this list that feel these "non-answers" 
are offered in spite.  I'm not sure how prevalent this view is or where
the feeling comes from.  It's certainly not my intent when I provide such
an answer, as I've clarified above.  I know I don't sit in my chair 
reading email, jealously holding onto all the answers, and responding
with pointers (or worse, some obtuse reference), just to throw someone
off the track or to keep them chasing an answer I know.  I provide the 
best answer I can at the time and I expec
t if it doesn't meet the need,
someone will speak up.  If the poster does follow-up, I or someone else 
may be able to help home in on the it a little more and provide a better 
solution or pointer.  Perhaps others have a different agenda when 
answering, although I've pretty much read every post on this list for the
last 5+ years and I've never been left with that impression.  YMMV.

So I guess what I'd like to say is, let's not throw around accusations 
of this sort.  If you receive a response to your query and its not what
you want, you're free to use it or not.  Query further if you like too.
Don't expect others have all the answers or be willing to look into the
details of all your problems.  I'm not saying that people won't fix your
problems or help you do so.  But they're going to do it their way, in 
their time, and at their option.  If that's not what you need or want, you 
can again query further but keep in mind that you're dealing with 
volunteers here.  Pushing may have the opposite reaction to your intended 
goal.  I actually think its a shame for people to be critical in the
face of someone's sincere intent to help the poster address their issue.
After all, the responder is only trying to provide useful information or
be truthful about their level of personal involvement in any 
implementation of a solution.  That all seems pretty reasonable and 
professional to me, even if the result is not something the poster wants
to hear.  However, the impression I'm getting from the discussion is that
unless someone is willing to provide any and all support for an issue,
in the form the poster wants it, then no response is preferable to some
response.  I guess I can live with that, if that's what the list in 
general wants but I personally feel it would make for a much less helpful
and active community.  Maybe I'm wrong.  Maybe its time for me personally
to adjust my level of participation in Cygwin, since I see my way of
contributing could be construed as fitting the pattern of "
discouragement"
as defined by others.  Hm, maybe.  I'll have to think a little more 
about that.  As is always the case, we can all use a little more free 
time! ;-)  Anyway, since we've all been sharing our thoughts on this 
matter I thought I'd offer mine, since its a slightly different than
some of the those posted earlier.  I'm really for the idea of having a 
Cygwin community.  So far, I believe its been a great success.  I hope it 
continues to be in some form! :-)  Actually, this is a good time for me
to say "thanks" to all those who work to provide and improve Cygwin and
its tools.  I don't do this enough.  This is really top-notch stuff! :-)




Larry Hall                              lhall@rfk.com
RFK Partners, Inc.                      http://www.rfk.com
118 Washington Street                   (508) 893-9779 - RFK Office
Holliston, MA 01746                     (508) 893-9889 - FAX



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Web page access for Michael Erdely

[Christopher Faylor]

Michael,
Please fill out the form at:

http://sources.redhat.com/cgi-bin/pdw/ps_form.cgi

The project is cygwin, the approver is cgf@redhat.com.
The email address is the address for any email sent to cygwin.com
or sources.redhat.com.

The ssh identity.pub file should be a v1 not a v2 key.

Once this is done, I will provide you with access to the cygwin web
pages.

Thanks,
cgf

On Thu, Jun 28, 2001 at 10:03:55AM -0400, Michael Erdely wrote:
>Documentation is an area I'd like to contribute in.  I've seen people on
>the list (especially Earnie) get up in arms when someone suggests a
>non-RedHat site (except for Delorie) with documentation.
>
>I've corresponded with several people that find some of the
>documentation difficult to follow for a newbie.  As Vince writes in
>another email ( http://cygwin.com/ml/cygwin/2001-06/msg01773.html ), not
>everyone that uses Cygwin knows Unix or knows C++.  Sometimes they have
>a need for a specific tool.  I have a friend that mainly uses bash and
>textutils from Cygwin so that the logs of some scripts are accessible to
>other Windows users without having to install Samba on his *nix machines
>or having some fancy file transfer scripts.  I know many people (who hit
>my web site) are just interested in using a free SSH server with
>Windows.  I don't think having this need should require an understanding
>of the code involved.  Especially since you don't have to know C++ in
>order to get sshd running.
>
>So, my contribution (and I state so at the bottom on my pages) is my
>whole section of http://mike.erdelynet.com/cygwin.asp .  It's mainly
>documentation to help a non-*nix person get Cygwin installed, configured
>and getting sshd running.  I'd be more than happy to contribute to more
>documentation or have my documentation taken by someone who can "polish"
>it.
>
>Anyway, let me know... ;-)
>
>-ME
>
>-----Original Message-----
>From: cygwin-owner@sources.redhat.com
>[ mailto:cygwin-owner@sources.redhat.com ] On Behalf Of Michael A. Chase
>Sent: Wednesday, June 27, 2001 11:05 PM
>To: cygwin@cygwin.com
>Subject: Re: "shouted down", "shot down", apologies
>
>
>----- Original Message -----
>From: "Christopher Faylor" <cgf@redhat.com>
>To: <cygwin@cygwin.com>
>Sent: Wednesday, June 27, 2001 17:51
>Subject: Re: "shouted down", "shot down", apologies
>
>
>> On Wed, Jun 27, 2001 at 08:32:52PM -0400, Brian Keener wrote:
>> >Just some step by step instructions - sometimes it's better to be
>over
>> >detailed than to leave information out.  I can always skip what I
>> >already know - but if it's not there I have to ask someone.
>>
>> Yes.  That's why I'm probably not the best person to be writing the
>> instructions.
>
>That's one of the areas I'd like to contribute in, once my company's
>disclaimer shakes loose.  It will be quite a while before I'm
>comfortable
>with gdb or the other troubleshooting tools.
>
><snip>
>
>
>--
>Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting:         http://cygwin.com/bugs.html
>Documentation:         http://cygwin.com/docs.html
>FAQ:                   http://cygwin.com/faq/

-- 
cgf@cygnus.com                        Red Hat, Inc.
http://sources.redhat.com/            http://www.redhat.com/

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Michael L. Smeby, Jr.]

I'm a complete newbie to Cygwin, but I use it here for various reasons, and
I think that between the great people on this list and the  documentation
and FAQs all over the web, I have yet to have a question that I couldn't
find the answer to in less than ten to thirty minutes (and I've had some
obscure ones). The volunteer work on the entire project is almost
overwhelming, and I haven't seen anyone just shrug off questions or cop an
attitude.

Just my two cents worth. You guys are appreciated .... I'll shut up now and
return to lurking. :-)

===
"Ahhh, the old take over the world ploy!" -- The Mummy Returns
Regards, Michael L. Smeby, Jr.
http://www.tampagov.com
----- Original Message -----
From: "Larry Hall (RFK Partners, Inc)" <lhall@rfk.com>
To: <cygwin@cygwin.com>
Sent: Thursday, June 28, 2001 12:28 PM
Subject: Re: "shouted down", "shot down", apologies


> At 10:22 AM 6/28/2001, Christopher Faylor wrote:
> >I won't comment on the rest of your message except to say that I rarely,
> >if ever, say "use the source" when I know the answer to the question.
> >
> >I say that when I don't know and when *I* would have to look things up.
> >
> >You're welcome to continue to ask "newbie" questions as frequently as
> >you like.  That will not stop me (or others, I presume) from pointing
> >people to references or suggesting the source when I don't know the
> >answer myself.
> >
> >I don't work for anyone here.  I am not obligated to look things up to
> >make your life easier.  It is that simple.
>
> [long - those uninterested in this thread want to hit delete now! :-) ]
>
> This is exactly my approach as well.  I have to say that I'm a bit
> dismayed that folks contributing to this and the "blunt tools" thread
> have mentioned dissatisfaction with what seemed to me to be such a
> straight-forward and logical approach.  When responding to queries on
> this list, I've always followed these simple rules:
>
>    1. If I know the question is an FAQ, I point to the entry there (*very*

>       rarely do I just point at the FAQ without the exact entry).
>       Generally I feel there's little benefit to restating what's in the
>       FAQ.  It just doesn't seem to be a good use of my time.  If its
>       inadequate in some way, we'll hear about it and make the appropriate
>       change (which seems to me as it should be).
>
>    2. If I kn
> ow something specific about the subject, I respond with it.
>       Sometimes this means I have to ask a question or two before I'm
>       sure what's been tried already and whether the poster is aware of
>       a previous discussion on the subject.  That all seems like part of
>       the process to me and I don't begrudge people for it.
>
>    3. If I know that this subject has come up before and has been
>       discussed but don't remember allot of details, I point to the
>       email archives.  In this case, I don't point to a specific message,
>       although I do occasionally offer a search key that I think might
>       help find the discussion I recall.  I don't spend my time looking
>       up the exact archive entry or entries that I'm recalling.  I don't
>       even promise that the stuff I'm remembering is even helpful (though
>       that's my intent and what I'm hoping for!)  I'm just providing
>       potential source of information that may prove useful.  It may not
>       too.  If it doesn't or its too hard to find, I expect the original
>       poster will query the list again with an update of the things tried
>       and the results.  If there's no success at this point, I sometimes
>       see if there's something more specific I can find myself and post
>       that if so.
>
>    4. If the question being answered is specific and detailed enough
>       that an inspection of the source is likely to be the only path to
>       a useful answer (barring someone else who has been in the source,
>       knows the answer, and will subsequently offer it), I *suggest*
>       looking at the source.  I do this when its clear someone is a
>       developer or has mentioned they are working with some other source.
>       I mention it if I'm not sure whether the person is a developer or
>       not, usually pointing out that it is an option if they're up to
>       it.  I tend not to mention it if the person states that they have
>       no experience reading/writing code.  Generally, I don't feel
>       obli
> gated to go inspect the source to answer someone else's question,
>       although there are exceptions or times I do it anyway.
>
>    5. If I know nothing about the subject, I keep my mouth shut.
>
> I've used all five of these modes in the past on this list and seen them
> work, at least on some occasions, exactly as I expected them to.  We've
> heard back from people who've had a hard time with an FAQ entry.  We've
> heard from people who say they've searched the archives but turned up
> nothing.  We've heard back from people saying they're not capable of
> looking at the source for one reason or another.  To me, all of this
> seems reasonable dialog in the course of trying to help someone with a
> problem.  I've always felt that providing some information, be it direct
> or a pointer to something which could be helpful is better than no answer
> at all (indeed, this list has more than once in the past been berated
> for *not* responding in some way to a post!)  However, it troubles me
> that some in the recent discussions have pointed to the replies with
> references to previous discussions and the FAQ as "non-answers" (I'm
> using this term generally now although I know it was a specific member
> of the previous discussions that first offered it up and it may have
> applied in that case to a problem with the specific set of tools in use
> at the time.  I think it categorizes a general sentiment I got from
> reading these threads though).  The impression I'm left with is that
> there is at least some people on this list that feel these "non-answers"
> are offered in spite.  I'm not sure how prevalent this view is or where
> the feeling comes from.  It's certainly not my intent when I provide such
> an answer, as I've clarified above.  I know I don't sit in my chair
> reading email, jealously holding onto all the answers, and responding
> with pointers (or worse, some obtuse reference), just to throw someone
> off the track or to keep them chasing an answer I know.  I provide the
> best answer I can at the time and I expec
> t if it doesn't meet the need,
> someone will speak up.  If the poster does follow-up, I or someone else
> may be able to help home in on the it a little more and provide a better
> solution or pointer.  Perhaps others have a different agenda when
> answering, although I've pretty much read every post on this list for the
> last 5+ years and I've never been left with that impression.  YMMV.
>
> So I guess what I'd like to say is, let's not throw around accusations
> of this sort.  If you receive a response to your query and its not what
> you want, you're free to use it or not.  Query further if you like too.
> Don't expect others have all the answers or be willing to look into the
> details of all your problems.  I'm not saying that people won't fix your
> problems or help you do so.  But they're going to do it their way, in
> their time, and at their option.  If that's not what you need or want, you
> can again query further but keep in mind that you're dealing with
> volunteers here.  Pushing may have the opposite reaction to your intended
> goal.  I actually think its a shame for people to be critical in the
> face of someone's sincere intent to help the poster address their issue.
> After all, the responder is only trying to provide useful information or
> be truthful about their level of personal involvement in any
> implementation of a solution.  That all seems pretty reasonable and
> professional to me, even if the result is not something the poster wants
> to hear.  However, the impression I'm getting from the discussion is that
> unless someone is willing to provide any and all support for an issue,
> in the form the poster wants it, then no response is preferable to some
> response.  I guess I can live with that, if that's what the list in
> general wants but I personally feel it would make for a much less helpful
> and active community.  Maybe I'm wrong.  Maybe its time for me personally
> to adjust my level of participation in Cygwin, since I see my way of
> contributing could be construed as fitting the pattern of "
> discouragement"
> as defined by others.  Hm, maybe.  I'll have to think a little more
> about that.  As is always the case, we can all use a little more free
> time! ;-)  Anyway, since we've all been sharing our thoughts on this
> matter I thought I'd offer mine, since its a slightly different than
> some of the those posted earlier.  I'm really for the idea of having a
> Cygwin community.  So far, I believe its been a great success.  I hope it
> continues to be in some form! :-)  Actually, this is a good time for me
> to say "thanks" to all those who work to provide and improve Cygwin and
> its tools.  I don't do this enough.  This is really top-notch stuff! :-)
>
>
>
>
> Larry Hall                              lhall@rfk.com
> RFK Partners, Inc.                      http://www.rfk.com
> 118 Washington Street                   (508) 893-9779 - RFK Office
> Holliston, MA 01746                     (508) 893-9889 - FAX
>
>
>
> --
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting:         http://cygwin.com/bugs.html
> Documentation:         http://cygwin.com/docs.html
> FAQ:                   http://cygwin.com/faq/
>


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Vince Rice]

I wasn't the one interested in building Cygwin; I think that was John.

You may not say "use the source" when you know the answer, but many here do.

I haven't asked any questions, newbie or otherwise, for reasons already
explained.  Nor have I asked you to make my life easier, or suggested that
you should.

Vince


-----Original Message-----
From: cygwin-owner@sources.redhat.com
[ mailto:cygwin-owner@sources.redhat.com]On Behalf Of Christopher Faylor
Sent: Thursday, June 28, 2001 9:23 AM
To: cygwin@cygwin.com
Subject: Re: "shouted down", "shot down", apologies


On Thu, Jun 28, 2001 at 02:31:36AM -0500, Vince Rice wrote:
>Chris,
>
>Somewhere in this thread you expressed some wonder/frustration at why the
>gcc/gdb projects don't generate the kind of "newbie" requests Cygwin (I
just
>tried to find it and it's nowhere to be found; I tend to lose things at
this
>time of night).  I have seen this expressed in your messages countless
times
>over the three years I've been lurking here.  And I've always wondered at
>your wonder <g>.

I used gcc and gdb as examples.  I could easily have used things like:
"bttv", "ssh", or "zsh".

I follow, to some degree, the discussions in those projects.  There are
few complaints about how hard it is to check things out using cvs or
build the tools.

>GCC and gdb are tools for developers; you don't use them unless you're
>developing something.  However, Cygwin is a tool for users; it's just as
>easy (easier IMHBAO) to *use* the Cygwin tools (defined as the GNU tools,

Well, if you are interested in building Cygwin, you should be some level
of developer, right?  If not, you are really due for an exercise in
frustration?  "What are all these semicolons, anyway?" "Why do all of
the files have this funny .cc extension?"

I won't comment on the rest of your message except to say that I rarely,
if ever, say "use the source" when I know the answer to the question.

I say that when I don't know and when *I* would have to look things up.

You're welcome to continue to ask "newbie" questions as frequently as
you like.  That will not stop me (or others, I presume) from pointing
people to references or suggesting the source when I don't know the
answer myself.

I don't work for anyone here.  I am not obligated to look things up to
make your life easier.  It is that simple.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Larry Hall (RFK Partners, Inc)]

At 09:20 AM 6/29/2001, Vince Rice wrote:
>You may not say "use the source" when you know the answer, but many here do.


I'm curious, how did you determine this?  (I'm serious - I'm not trying
to imply any "attitude" with this query)


Larry Hall                              lhall@rfk.com
RFK Partners, Inc.                      http://www.rfk.com
118 Washington Street                   (508) 893-9779 - RFK Office
Holliston, MA 01746                     (508) 893-9889 - FAX


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Fri, Jun 29, 2001 at 08:20:09AM -0500, Vince Rice wrote:
>I wasn't the one interested in building Cygwin; I think that was John.

You were apparently responding to my query as to why CVS checkout,
configuring, and building were an apparent big deal here but not in
other mailing lists.

I opined that we got more programming newbies but I didn't really
understand why I didn't see more newbie confusion in other mailing
lists.

I have expressed rhetorical "wonder" at the huge number of newbies that
we get here in the past but, in this particular case, I was being pretty
specific.  I was referring to people who were volunteering to help out
with development.

I think I'm clever enough to realize that a person with no C experience
will not know a lot about C development.

So, you seemed to be responding to a specific query and I was responding
to your response.

>You may not say "use the source" when you know the answer, but many
>here do.

Ok, but you seemed to be accusing me, specifically, of being unhelpful.

It seems to be the case that if someone asks a question and I point them
to the source (these days, I even use a specific file in the source, if
I can) then that is interpreted as a slap in the face.  Perhaps that is
a newbie attitude.  Perhaps a newbie expects that software that they
download for free should come with high-level technical support.  I
don't know.

When I point people at the source, I usually do so because I do not
think that anyone else will be answering their questions.  I could
wai" and see if that is true but I would prefer not to have to set up
some kind of crude bug tracking system which ages requests and then
notifies me when no one has responded.  I assume that Larry Hall feels
the same way about this when he answers generally, rather than
specifically.

It still boils down to how much work I am willing to put into this
project.  I think I have become convinced that time spent answering
questions here is, by and large, not appreciated.  You can see it in the
zip MS-DOS path thread.  I tried to respond quickly with the rationale
for the apparent bug but was quickly accused of "disdain" -- apparently
because I wasn't interested in spending my time fixing that particular
problem.

So, that's fine.  I will try to curb my response to email here and let
the community develop its own mechanisms for answering questions.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Brian Keener]

Brian Keener wrote:
> I am probably going to jump into this now without intending to but this really 
> does not answer his question - sorry Chris and I am not aiming this at you
>
Chris et al,

I did jump into this with both feet and possibly with my eyes a little closed. 
While I do feel some of my comments were truthful, I also realize that my 
frustration from time to time in wanting to help and learn and not having (to 
some degree) any clue what I was really doing lended to my own personal 
aggravation trying to get things to work properly for me.  Part of the problem is 
none of us know the level of ability we are dealing with at the other end of this 
list and sometimes that person needs that little extra nudge in the right 
direction - but then none of us know that do we until they come back the second 
or third time and then depending on where their at they're really frustrated. 

At any rate, as one of the first to comment in this thread and to point out that 
I didn't feel it answered John's question I apologize to all for (in my opinion) 
starting this attack thread.  I didn't feel that particular answer (depending on 
the persons level) was appropriate and I had been to the web pages to try to work 
out some of my problems with starting from scratch, knowing nothing, and trying 
to work with cygwin and cvs and maybe I'm thick but they just didn't help. But 
then as Chris pointed out - if it needs work - if it needs improvement - change 
it.  Good point - I can't argue.

I made my comments entirely to broad by drawing in some of my own problems (which 
again as Chris pointed out - I was the only one having and he is right) and then 
proceeded to mystify the whole thing giving John and uneasy feeling and he has 
experience.  

So in closing - To Chris and all - I do apologize for throwing my opinion in 
there and do really want to thank everyone for their help.  I have learned a more 
than I could imagine from everything I have done with cygwin and the setup mods 
and from everyones help and while there is always room for improvement in all of 
us and everything we do - this is a fine project and a fine community.

I'm reeeaaaaallllllyyyyyy ssssoooorrrrrrrryyyyy I said anything Chris.




--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Fri, Jun 29, 2001 at 05:24:00PM -0400, Brian Keener wrote:
>I'm reeeaaaaallllllyyyyyy ssssoooorrrrrrrryyyyy I said anything Chris.

No worries and no apologies necessary, Brian.

Take care,
cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Warren Young]

Christopher Faylor wrote:
> 
> I used gcc and gdb as examples.  I could easily have used things like:
> "bttv", "ssh", or "zsh".
> 
> I follow, to some degree, the discussions in those projects.  There are
> few complaints about how hard it is to check things out using cvs or
> build the tools.

To pick one of your examples, probably the reason there are more
questions about Cygwin than about zsh is that zsh users are mostly
experienced in the Unix way already, and they're building and using it
on a system that is stable and (largely) bug-free.  Also, the zsh
community is mostly saturated: it isn't going to grow at a rate much
faster than that of the world's population.

With Cygwin, a lot of people are being thrust into the Unix Advernture
game for the first time, and have found that there are a lot of grues
lurking about in the dark corners.  By Cygwin's very nature, the Cygwin
community will always be full of people finding mismatches at the
interface between two different worlds.

Until recently, it wasn't even clear what the overall design philosophy
of Cygwin was going to be.  Consider the whole //d/path/on/drive/d vs
/cygpath/d/ vs d:/path issue.  I seem to recall that back in the
b{18-21} days, the first two options didn't even exist.  The installer
didn't use the mount points feature, if it existed at all -- I used to
install directly into c:\ (e.g. c:\usr\bin for the binaries, c:\bin for
critical things like sh.exe, etc.) so that I could use 4NT for my shell
and still use Unix-like paths.  As time went on, we got setup.exe, which
put Cygwin off in its own directory and set up mount points that only
worked when using a Unix shell.  Cygwin made this and other 
distinctions about the world of Cygwin vs the Windows world, while still
trying to maintain two-way compatibility.

I don't intend to try and decide whether this is bad or good, just to
show that Cygwin's design is still evolving, and the location of the
boundary between Unix and Windows is still getting tweaked to and fro. 
Just like Linux's policy of changing internal kernel interfaces on a
whim to improve the overall design, Cygwin's evolving design sometimes
causes problems.  But the system always improves through this process.

Until the day several eons hence when the design freezes and the user
community expands to its saturation point and the bugs are all (mostly)
worked out, there will always be people getting cut on the bleeding
edge.  This is expected.  We'll need some good docs to help these poor
newbies get accultured.

-- 
= Warren -- Video articles: http://www.cyberport.com/~tangent/video/
= 
= ICBM Address: 36.8274040 N, 108.0204086 W, alt. 1714m

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Michael A. Chase]

----- Original Message ----- 
From: "Christopher Faylor" <cgf@redhat.com>
To: <cygwin@cygwin.com>
Sent: Friday, June 29, 2001 14:43
Subject: Re: "shouted down", "shot down", apologies


> On Fri, Jun 29, 2001 at 05:24:00PM -0400, Brian Keener wrote:
> >I'm reeeaaaaallllllyyyyyy ssssoooorrrrrrrryyyyy I said anything Chris.
> 
> No worries and no apologies necessary, Brian.

Group hug time?


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Vince Rice]

>>I wasn't the one interested in building Cygwin; I think that was John.

> You were apparently responding to my query as to why CVS checkout,
> configuring, and building were an apparent big deal here but not in
> other mailing lists.

The applicable quote was:
	I guess the problem that I have with all of this is you almost never
	see anyone in the gcc, or gdb projects professing that cvs checkout
	and building to be a major obstacle.  I don't know what is so different
	about cygwin.

	This project seems to attract more "newbies".  Maybe the "oldbies" are
	all using linux and disdaining Windows.

I said in my initial message that you had said "somewhere in the thread",
because I couldn't find the message when I went looking for it, even though
I had just read it a few minutes before (I found the quote above in the
archives.  I didn't look in the archives the other night because I had just
read the message in my email ten minutes before and thought I remembered it.
Goes to show what memory's good for at 2:00am).  So, what I remembered of
your question was the *last* paragraph above, not the first one.  Thus, I
was addressing the "more newbies" comment, not specifically the checking out
and building.  So, by relying on memory, and not having your full quote in
front of me, I seized on the wrong statement.  My apologies.  I was
responding to the wrong question.

>>You may not say "use the source" when you know the answer, but many
>>here do.

> Ok, but you seemed to be accusing me, specifically, of being unhelpful.

I have no idea how you came to that conclusion, but it is incorrect.  I have
just reviewed my entire message, and absolutely nowhere do I either
explicitly or implicitly accuse you of such, nor was that my intention, nor
was that my thought.  In fact, I said the opposite -- that I agree with your
philosophy.  I also said at the end that that one paragraph digression on
"community" wasn't even the reason I wrote, it was to respond to the
original question (which we've already established I mis-targeted).

Let me spell it out -- I appreciate you and the work you do a great deal.  I
appreciate D.J. and Ernie and Charles and all the rest of the developers
with the skills to improve Cygwin from a code perspective, who work for
nothing.  (I even agree with you *completely* about the zip thread <g>).  I
appreciate it enough I *don't* ask questions here.  I just lurk.  Or I did
before my original message.  I think I'll get back to that.  The lurking
that is.  My deepest sincere apologies to anyone else I've offended.

Vince


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[David A. Cobb]

I know: This is a week old.  I'm just playing catchup.

At 6/27/01 05:39 PM (Wednesday), Brian Keener wrote:
>I am probably going to jump into this now without intending to but this 
>really
>does not answer his question - sorry Chris and I am not aiming this at you
>Chris - This is simply an observation. He asked:
>
> >> So, if I want be able to rebuild all the cygwin packages, can I do 
> that from
> >> source downloaded with setup.exe?  Can someone recommend a convenient 
> way of
> >> building a "test cygwin" from that source which can be switched to (via a
> >> change to cygwin.bat) to try out changes?  Or do I really have to go 
> the cvs
> >> route and work with the latest/greatest bleeding-edge packages?  If csv is
> >> the only/best way, does someone have a cookbook which will allow me to 
> setup
> >> a test environment, refresh the source, build everything, make a change,
> >> test it out, submit a patch -- all while keeping a working cygwin
> >> environment built off of setup.exe's download?
>
>As someone who is trying to learn C++ and never really worked with it much 
>and
>as a contributor to the Cygwin project by contributing to setup.exe these are
>questions that I and many others have asked before at various times.  I have
>read the above links before and probably missed where the answers to John's
>questions were located but as far as I can tell there isn't a lot to get
>someone going on contributing to Cygwin.  Or let me say there is a not a 
>lot if
>you are inexperienced in cvs, cygwin, C++, gcc, make as they are used in this
>environment. I have essentially pieced together from various sources (you 
>Chris
>and Earnie and Dj and whole lot of others) and reading the man pages and 
>asking
>very pointed questions just enough to be able to make changes, compile and 
>test
>them and then submit them.

Having done all those things, pieced together the information, etc --- 
perhaps you would be so good as to write a wee bit of a document that could 
be referenced from the "Contributing" page?

It might not be a "cookbook," but friendly tips are almost never unwelcome.



David A. Cobb, Software Engineer, Public Access Advocate, All around nice guy.
Get my PGP key at
:< http://pgpkeys.mit.edu:11371/pks/lookup?op=get&search=superbiskit >
Fingerprint=0x{6E3E_DB8C_2E8C_4248_62B2_FE29_08EE_CF0A_3629_E954}
:< http://wwwkeys.pgp.net:11371/pks/lookup?op=get&search=superbiskit&fingerprint=on >
"By God's Grace I am a Christian man, by my actions a great sinner."
--The Way of a Pilgrim, R. M. French [tr.]
Potentially Viral Software is any software for which you are not allowed
to examine the source.  Do not buy or use Potentially Viral Software!
<---.----!----.----!----.----!----.----!----.----!----.----!----.---->


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Glen Coakley]

> Believe it or not, I forgot to add something to this thread.
> 
> For those that aren't familiar with cvs there is Windows 
> wrapper for it called 'Wincvs' that you may find easier to 
> acclimate to.
> 
> Perhaps it would be useful to add something like the ACE 
> project's Problem Report Form (PRF) 
> ( http://www.cs.wustl.edu/~schmidt/ACE_wrappers/PROBLEM-REPORT-FORM ) 
> that would channel new users to providing more appropriate information 
> when they have questions. I realize that cygcheck -(whatever those 
> options are) will do much of this work but the hallmarks of debugging 
> (clear problem explanations, what was tried and a simple test
case/example) 
> are not always to non-developers (as Dilbert likes to illustrate).

Lest I add to recent tensions, that statement was meant to be a question. 
Would it be useful...? Sorry, I only checked my message for spelling not
intent.

________________________________
Glen Coakley, Sr. Software Engineer
MQSoftware Inc., (763) 543-4845
"Tinkero ergo sum." -- Chuck Murcko

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Glen Coakley]

Believe it or not, I forgot to add something to this thread.

For those that aren't familiar with cvs there is Windows wrapper for it
called 'Wincvs' that you may find easier to acclimate to.

Perhaps it would be useful to add something like the ACE project's Problem
Report Form (PRF)
( http://www.cs.wustl.edu/~schmidt/ACE_wrappers/PROBLEM-REPORT-FORM ) that
would channel new users to providing more appropriate information when they
have questions. I realize that cygcheck -(whatever those options are) will
do much of this work but the hallmarks of debugging (clear problem
explanations, what was tried and a simple test case/example) are not always
to non-developers (as Dilbert likes to illustrate).

________________________________
Glen Coakley, Sr. Software Engineer
MQSoftware Inc., (763) 543-4845
"Tinkero ergo sum." -- Chuck Murcko

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Robert Collins]

----- Original Message -----
From: "Warren Young" <warren@etr-usa.com>
<snip>

> ...and you now allow "contrib" packages.  For the longest time, Cygwin
> was just a GNU environment for Windows.  These days, I think of Cygwin
> as "Linux for Windows", what with all the new packages attached to it.
>
> Which, by the way, leads me back to another proposal I've made that
> never went anywhere: a setup.exe option allowing a minimal Cygwin
> install.  This one requires code to be convincing, I know.

My turn to be unpopular. Here's a hypothetical question: If development to
allow such a setup.exe was in progress, where would you see it (assuming
nothing has hit CVS yet)? cygwin-developers@cygwin.com maybe? Or perhaps
cygwin-patches@cygwin.com?

Well both those lists have seen significant activity towards such a
setup.exe in the last fortnight. We've coded per package dependencies and
package cetgorisation, with only certaion categories defaulting to install.

The point? This is all publicly available information, in the "expected
place" that you could have looked at before making "your point". A point
that was agreed with by Chris when you made it ~2-3 months ago (and the
answer then was "code it".

This gets me fairly annoyed. You make a valid point, are told "put up or
shut up" (a fair comment IMO), don't do anything for x months, and then make
the same point again, when some of us folk have being doing something about
it.

If you're going to throw suggestions, at least spend the ~30 seconds it
would have taken to search for "setup" on the cygwin-patches archive.

Rob


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Warren Young]

Christopher Faylor wrote:
> 
> I have been concerned by two recent messages where people have felt
> that their ideas have been "shouted down" or "shot down".
> 
> That bothers me.  It bothers me because I assume that most, if not
> all of the negative perception undoubtedly came from me.

You have to realize that no-one can feel shot down by mere words if they
don't respect the person doing the shooting.  You are arguably Cygwin's
Linus -- Linus doesn't "make" Linux any more than you "make" Cygwin, yet
your central role in maintaining the core gives your opinions a certain
weight.  I and others respect that, so that if you think a thing should
not be done, we are likely to go along with your view.  We might not be
happy with it, but we are more interested in keeping you happy than in
upsetting the apple cart to our own ends.  This is one of the forces
that binds Open Source projects together, and we shouldn't try to
second-guess it; we just need to understand and wield this force to good
ends.

On the FAQ issue, Larry Hall and I have had a private email exchange
that may bring some resolution to this issue.  Watch the list for
details in the next few days.

Just so you know, I'm not all that upset or bitter, just disappointed
that I was unable to change any minds.  After deciding that I was just
going to let the status remain quo (so to speak :) ) I only piped back
up on this issue because I thought I knew what the answer was going to
be.  If I was wrong, that's fine, too, because I do think something
should be done about the current way of distributing answers to newbies.

> So, filling the FAQ with non-frequently asked questions does not seem like
> the way to go to me.  It seems like it will make the FAQ harder to navigate
> and will make it easier for people to miss things.

Consider navigating the world: one does not try to bring along a single
map to do this.

If you were going to visit a friend in Jakarta, you might first find it
on a globe, then set out in that general direction.  Once you got in the
vicinity, you might pick up a country map, to show you major roads you
could take to get to the city.  Then when you reach the city, you might
pick up a street map to find your friend's house.

My point is that easy navigation is about hierarchical organization. 
The world is a mondo-huge place, but it's still possible to find
relatively small things like particular houses.

The moral: large Q&A sources are fine, if organized properly.

> Updating the documentation *does* make sense to me.

That's all we're really disagreeing on: the organization of the "help me
fix Cygwin" document, and what it shall be called.  I started off
calling it a FAQ, but if that term is distasteful, that's fine.  Let's
just _do_ something and let the name follow.

> Please reorient your thinking from "This is what they should do" to
> "This is what I can do".

Let us call my proposal of a few months ago the "bloated FAQ" idea. 
Would you care to speculate about what would have happened if I had
decided to go ahead and make my own bloated FAQ?  Let's say it had a lot
of good material, that it was easy to navigate, etc., but that it was
hosted on my site, not an official Cygwin publication.  I am curious
what the reaction would have been.

> And, I also enjoy running a project like Cygwin.  I think that the net
> release of Cygwin has improved dramatically in the last couple of years.

Without a doubt.  There will always be something that isn't "right" yet,
and as Cygwin grows there will be more and more newbies.  This is why it
makes sense to start making more newbie documentation at this time. 
Obviously you're a coder and with respect to Cygwin I'm mostly a
technical writer, so maybe I should have just started the new "help me
fix Cygwin" document instead of just proposing it.

> That is because I've lobbied for changes inside of Red Hat and solicited
> active maintainers outside of Red Hat.  And, I've encouraged the
> development of the cygwin installer.

...and you now allow "contrib" packages.  For the longest time, Cygwin
was just a GNU environment for Windows.  These days, I think of Cygwin
as "Linux for Windows", what with all the new packages attached to it.

Which, by the way, leads me back to another proposal I've made that
never went anywhere: a setup.exe option allowing a minimal Cygwin
install.  This one requires code to be convincing, I know.

> I actually have a tendency to just see all of the negatives in cygwin.

I agree, Cygwin is a pretty crappy Linux distribution.  Still useful,
tho'.  :)

> (Although, I will probably still try to be "humorous" from time to
> time.  Be warned.)

Me, too.  See above.  :)

-- 
= Warren -- ICBM Address: 36.8274040 N, 108.0204086 W, alt. 1714m

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Fred T. Hamster]

Christopher Faylor wrote:

    It still boils down to how much work I am willing to put into this
    project. I think I have become convinced that time spent answering
    questions here is, by and large, not appreciated. You can see it in
    the zip MS-DOS path thread. I tried to respond quickly with the
    rationale for the apparent bug but was quickly accused of "disdain"
    -- apparently because I wasn't interested in spending my time fixing
    that particular problem.

uh, here's the text being referenced:

Fred opined:
in my opinion, one cannot simultaneously disdain the win32 path conventions and yet also promote a product that is intended 
exclusively for win32 without resorting to some form of schizophrenia.

note that the text is not referring to any particular person as 
disdaining.  it is a general statement of opinion about the state of 
mind potentially involved in such a standpoint.  it could just as easily 
have applied to myself as to anyone else on the list.
  again, i admit that my choice of words may have been unfortunately 
outrageous.  i apologize for that again.
  i really appreciated your explaining your viewpoint in the first 
posting on this thread, chris.  i apologize for my role in all of this 
if your openness gave people some feeling that bashing you or the list 
or the product was what you were inviting.
  i still think that several people took comments of mine personally, 
when they were meant to be applied to a potential attitude (which no one 
may even have).  i suck.  i'm sorry.
  however, please don't warp what i said into something it was not.  i 
did appreciate your comments explaining the rationale, but i had (and 
have) a basic logical disconnect with them and with how i was hoping to 
use the product.  i didn't think you were disdaining me, but i thought 
that in general there was some disdain (exhibited by multiple list 
posters) for win32 pathnames.  i have that disdain too, but i often have 
to use win32 paths, regardless of how i feel about them.
  anyway, i really am hoping that people can increase the amount of 
respect they exhibit towards others on this list (and within all 
internet mailing lists and newsgroups).  i've seen newbie posts that are 
just as corrosive as some of the comments that annoyed experts have 
made.  to quote rodney king (my favorite example of a man who has been 
terribly beaten but still wants to recover and move past it): "why can't 
we all just get along?".
thanks,
fred.

-- 
_____ chosen by the Nechung Oracle Program [ http://www.gruntose.com/ ] _____

It's too bad that whole families have been torn apart by something as
simple as wild dogs.
  -- Deep Thoughts by Jack Handey

_____________ not necessarily my opinions, not necessarily not. _____________



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[John Wiersba]

> -----Original Message-----
> From: Christopher Faylor [ mailto:cgf@redhat.com ]
> Sent: Friday, June 29, 2001 11:06 AM
> To: cygwin@cygwin.com
> Subject: Re: "shouted down", "shot down", apologies
>
> It still boils down to how much work I am willing to put into this
> project.  I think I have become convinced that time spent answering
> questions here is, by and large, not appreciated.  You can 
> see it in the
> zip MS-DOS path thread.  I tried to respond quickly with the rationale
> for the apparent bug but was quickly accused of "disdain" -- 
> apparently
> because I wasn't interested in spending my time fixing that particular
> problem.
> 
> So, that's fine.  I will try to curb my response to email here and let
> the community develop its own mechanisms for answering questions.

Chris, let me say that I for one *do* appreciate all the work that you and
others (Larry, Charles, etc) put in answering questions, stupid or
otherwise.  When I ask a question which is appropriately answered by a
pointer to a FAQ or other web page, I do *not* take offense at that.  It's a
perfectly appropriate response.  I also agree with Larry's approach to
answering questions which he detailed in a previous post to this thread.
So, thank all of you for your answers, short or otherwise.

Regarding the zip MS-DOS path thread, I felt obligated to reply to that
thread in an effort to try to help clarify your position (which I also held)
when it didn't seem to be understood by the poster.  After spending your
time replying to someone who doesn't seem to be understanding what you're
saying, I could understand that you might be getting frustrated, so I even
took your "hasty" comments, which you later apologized for, in that light
and just glossed over them.

People (myself included) may at times ask questions which they can answer
for themselves with a little nudge in the right direction.  So, a (very)
brief nudge is appropriate and sufficient.  No offense need be taken by
either side.  I especially like Larry's mention in his post of giving URLs
to subsections inside a web page or FAQ page.  That is especially helpful
and a lot more friendly than just saying "read the FAQ", because the FAQ and
Users Guide are not tiny documents.  Hopefully, once someone gets a response
like this they will start becoming more familiar with the documentation
available and will be less likely to ask questions like this.

-- John Wiersba

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Larry Hall (RFK Partners, Inc)]

David,

Did you read (5)?  How is that different than what you're suggesting?
I don't see the difference between this professed practice of mine and
your suggestion.  What did I miss?

Larry


At 07:58 AM 6/29/2001, Robinow, David wrote:

>Larry, have you considered just shutting up when you don't know the answer?
> > From: Larry Hall (RFK Partners, Inc) [ mailto:lhall@rfk.com ]
> > Sent: Thursday, June 28, 2001 12:28 PM
> > To: cygwin@cygwin.com
> > Subject: Re: "shouted down", "shot down", apologies
> > This is exactly my approach as well.  I have to say that I'm a bit 
> > dismayed that folks contributing to this and the "blunt tools" thread
> > have mentioned dissatisfaction with what seemed to me to be such a 
> > straight-forward and logical approach.  When responding to queries on
> > this list, I've always followed these simple rules:
> > 
> >    1. If I know the question is an FAQ, I point to the entry 
> > there (*very*
> >       rarely do I just point at the FAQ without the exact entry).  
> >       Generally I feel there's little benefit to restating 
> > what's in the 
> >       FAQ.  It just doesn't seem to be a good use of my time.  If its 
> >       inadequate in some way, we'll hear about it and make 
> > the appropriate 
> >       change (which seems to me as it should be).
> > 
> >    2. If I kn
> > ow something specific about the subject, I respond with it.
> >       Sometimes this means I have to ask a question or two before I'm 
> >       sure what's been tried already and whether the poster 
> > is aware of 
> >       a previous discussion on the subject.  That all seems 
> > like part of
> >       the process to me and I don't begrudge people for it.
> > 
> >    3. If I know that this subject has come up before and has been 
> >       discussed but don't remember allot of details, I point to the 
> >       email archives.  In this case, I don't point to a 
> > specific message,
> >       although I do occasionally offer a search key that I think might
> >       help find the discussion I recall.  I don't spend my 
> > time looking 
> >       up the exact archive entry or entries that I'm 
> > recalling.  I don't
> >       even promise that the stuff I'm remembering is even 
> > helpful (though
> >       that's my intent and what I'm hoping for!)  I'm just providing 
> >       potential source of information that may prove useful.  
> > It may not 
> >       too.  If it doesn't or its too hard to find, I expect 
> > the original 
> >       poster will query the list again with an update of the 
> > things tried 
> >       and the results.  If there's no success at this point, 
> > I sometimes
> >       see if there's something more specific I can find 
> > myself and post 
> >       that if so.
> > 
> >    4. If the question being answered is specific and detailed enough
> >       that an inspection of the source is likely to be the 
> > only path to
> >       a useful answer (barring someone else who has been in 
> > the source,
> >       knows the answer, and will subsequently offer it), I *suggest* 
> >       looking at the source.  I do this when its clear someone is a 
> >       developer or has mentioned they are working with some 
> > other source.
> >       I mention it if I'm not sure whether the person is a 
> > developer or 
> >       not, usually pointing out that it is an option if they're up to 
> >       it.  I tend not to mention it if the person states that 
> > they have
> >       no experience reading/writing code.  Generally, I don't feel 
> >       obli
> > gated to go inspect the source to answer someone else's question,
> >       although there are exceptions or times I do it anyway.
> > 
> >    5. If I know nothing about the subject, I keep my mouth shut.
> > 
> > I've used all five of these modes in the past on this list 
> > and seen them
> > work, at least on some occasions, exactly as I expected them 
> > to.  We've
> > heard back from people who've had a hard time with an FAQ 
> > entry.  We've
> > heard from people who say they've searched the archives but turned up
> > nothing.  We've heard back from people saying they're not capable of 
> > looking at the source for one reason or another.  To me, all of this 
> > seems reasonable dialog in the course of trying to help 
> > someone with a 
> > problem.  I've always felt that providing some information, 
> > be it direct
> > or a pointer to something which could be helpful is better 
> > than no answer
> > at all (indeed, this list has more than once in the past been berated
> > for *not* responding in some way to a post!)  However, it troubles me 
> > that some in the recent discussions have pointed to the replies with 
> > references to previous discussions and the FAQ as "non-answers" (I'm 
> > using this term generally now although I know it was a 
> > specific member 
> > of the previous discussions that first offered it up and it may have 
> > applied in that case to a problem with the specific set of 
> > tools in use 
> > at the time.  I think it categorizes a general sentiment I got from 
> > reading these threads though).  The impression I'm left with is that 
> > there is at least some people on this list that feel these 
> > "non-answers" 
> > are offered in spite.  I'm not sure how prevalent this view 
> > is or where
> > the feeling comes from.  It's certainly not my intent when I 
> > provide such
> > an answer, as I've clarified above.  I know I don't sit in my chair 
> > reading email, jealously holding onto all the answers, and responding
> > with pointers (or worse, some obtuse reference), just to throw someone
> > off the track or to keep them chasing an answer I know.  I 
> > provide the 
> > best answer I can at the time and I expec
> > t if it doesn't meet the need,
> > someone will speak up.  If the poster does follow-up, I or 
> > someone else 
> > may be able to help home in on the it a little more and 
> > provide a better 
> > solution or pointer.  Perhaps others have a different agenda when 
> > answering, although I've pretty much read every post on this 
> > list for the
> > last 5+ years and I've never been left with that impression.  YMMV.
> > 
> > So I guess what I'd like to say is, let's not throw around 
> > accusations 
> > of this sort.  If you receive a response to your query and 
> > its not what
> > you want, you're free to use it or not.  Query further if you 
> > like too.
> > Don't expect others have all the answers or be willing to 
> > look into the
> > details of all your problems.  I'm not saying that people 
> > won't fix your
> > problems or help you do so.  But they're going to do it their way, in 
> > their time, and at their option.  If that's not what you need 
> > or want, you 
> > can again query further but keep in mind that you're dealing with 
> > volunteers here.  Pushing may have the opposite reaction to 
> > your intended 
> > goal.  I actually think its a shame for people to be critical in the
> > face of someone's sincere intent to help the poster address 
> > their issue.
> > After all, the responder is only trying to provide useful 
> > information or
> > be truthful about their level of personal involvement in any 
> > implementation of a solution.  That all seems pretty reasonable and 
> > professional to me, even if the result is not something the 
> > poster wants
> > to hear.  However, the impression I'm getting from the 
> > discussion is that
> > unless someone is willing to provide any and all support for an issue,
> > in the form the poster wants it, then no response is 
> > preferable to some
> > response.  I guess I can live with that, if that's what the list in 
> > general wants but I personally feel it would make for a much 
> > less helpful
> > and active community.  Maybe I'm wrong.  Maybe its time for 
> > me personally
> > to adjust my level of participation in Cygwin, since I see my way of
> > contributing could be construed as fitting the pattern of "
> > discouragement"
> > as defined by others.  Hm, maybe.  I'll have to think a little more 
> > about that.  As is always the case, we can all use a little more free 
> > time! ;-)  Anyway, since we've all been sharing our thoughts on this 
> > matter I thought I'd offer mine, since its a slightly different than
> > some of the those posted earlier.  I'm really for the idea of 
> > having a 
> > Cygwin community.  So far, I believe its been a great 
> > success.  I hope it 
> > continues to be in some form! :-)  Actually, this is a good 
> > time for me
> > to say "thanks" to all those who work to provide and improve 
> > Cygwin and
> > its tools.  I don't do this enough.  This is really top-notch 
> > stuff! :-)
>
>--
>Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
>Bug reporting:         http://cygwin.com/bugs.html
>Documentation:         http://cygwin.com/docs.html
>FAQ:                   http://cygwin.com/faq/


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Robinow, David]

> -----Original Message-----
> From: Robert Collins [ mailto:robert.collins@itdomain.com.au ]
> Sent: Friday, June 29, 2001 9:16 AM
> To: Robinow, David; cygwin@cygwin.com
 
> David, Have you considered only quoting the relevant portion 
> of the email
> you are referring to?
 Yes. Failure to do that is one of my pet peeves.  I actually did cut some
of  it but the long-winded message seemed appropriate to the point I was
trying to make.  Perhaps I erred?

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Robert Collins]

----- Original Message -----
From: "Robinow, David" <drobinow@dayton.adroit.com>
To: <cygwin@cygwin.com>
Sent: Friday, June 29, 2001 9:58 PM
Subject: RE: "shouted down", "shot down", apologies


>
> Larry, have you considered just shutting up when you don't know the
answer?
> > From: Larry Hall (RFK Partners, Inc) [ mailto:lhall@rfk.com ]


David, Have you considered only quoting the relevant portion of the email
you are referring to?

Rob


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Robinow, David]

Larry, have you considered just shutting up when you don't know the answer?
> From: Larry Hall (RFK Partners, Inc) [ mailto:lhall@rfk.com ]
> Sent: Thursday, June 28, 2001 12:28 PM
> To: cygwin@cygwin.com
> Subject: Re: "shouted down", "shot down", apologies
> This is exactly my approach as well.  I have to say that I'm a bit 
> dismayed that folks contributing to this and the "blunt tools" thread
> have mentioned dissatisfaction with what seemed to me to be such a 
> straight-forward and logical approach.  When responding to queries on
> this list, I've always followed these simple rules:
> 
>    1. If I know the question is an FAQ, I point to the entry 
> there (*very*
>       rarely do I just point at the FAQ without the exact entry).  
>       Generally I feel there's little benefit to restating 
> what's in the 
>       FAQ.  It just doesn't seem to be a good use of my time.  If its 
>       inadequate in some way, we'll hear about it and make 
> the appropriate 
>       change (which seems to me as it should be).
> 
>    2. If I kn
> ow something specific about the subject, I respond with it.
>       Sometimes this means I have to ask a question or two before I'm 
>       sure what's been tried already and whether the poster 
> is aware of 
>       a previous discussion on the subject.  That all seems 
> like part of
>       the process to me and I don't begrudge people for it.
> 
>    3. If I know that this subject has come up before and has been 
>       discussed but don't remember allot of details, I point to the 
>       email archives.  In this case, I don't point to a 
> specific message,
>       although I do occasionally offer a search key that I think might
>       help find the discussion I recall.  I don't spend my 
> time looking 
>       up the exact archive entry or entries that I'm 
> recalling.  I don't
>       even promise that the stuff I'm remembering is even 
> helpful (though
>       that's my intent and what I'm hoping for!)  I'm just providing 
>       potential source of information that may prove useful.  
> It may not 
>       too.  If it doesn't or its too hard to find, I expect 
> the original 
>       poster will query the list again with an update of the 
> things tried 
>       and the results.  If there's no success at this point, 
> I sometimes
>       see if there's something more specific I can find 
> myself and post 
>       that if so.
> 
>    4. If the question being answered is specific and detailed enough
>       that an inspection of the source is likely to be the 
> only path to
>       a useful answer (barring someone else who has been in 
> the source,
>       knows the answer, and will subsequently offer it), I *suggest* 
>       looking at the source.  I do this when its clear someone is a 
>       developer or has mentioned they are working with some 
> other source.
>       I mention it if I'm not sure whether the person is a 
> developer or 
>       not, usually pointing out that it is an option if they're up to 
>       it.  I tend not to mention it if the person states that 
> they have
>       no experience reading/writing code.  Generally, I don't feel 
>       obli
> gated to go inspect the source to answer someone else's question,
>       although there are exceptions or times I do it anyway.
> 
>    5. If I know nothing about the subject, I keep my mouth shut.
> 
> I've used all five of these modes in the past on this list 
> and seen them
> work, at least on some occasions, exactly as I expected them 
> to.  We've
> heard back from people who've had a hard time with an FAQ 
> entry.  We've
> heard from people who say they've searched the archives but turned up
> nothing.  We've heard back from people saying they're not capable of 
> looking at the source for one reason or another.  To me, all of this 
> seems reasonable dialog in the course of trying to help 
> someone with a 
> problem.  I've always felt that providing some information, 
> be it direct
> or a pointer to something which could be helpful is better 
> than no answer
> at all (indeed, this list has more than once in the past been berated
> for *not* responding in some way to a post!)  However, it troubles me 
> that some in the recent discussions have pointed to the replies with 
> references to previous discussions and the FAQ as "non-answers" (I'm 
> using this term generally now although I know it was a 
> specific member 
> of the previous discussions that first offered it up and it may have 
> applied in that case to a problem with the specific set of 
> tools in use 
> at the time.  I think it categorizes a general sentiment I got from 
> reading these threads though).  The impression I'm left with is that 
> there is at least some people on this list that feel these 
> "non-answers" 
> are offered in spite.  I'm not sure how prevalent this view 
> is or where
> the feeling comes from.  It's certainly not my intent when I 
> provide such
> an answer, as I've clarified above.  I know I don't sit in my chair 
> reading email, jealously holding onto all the answers, and responding
> with pointers (or worse, some obtuse reference), just to throw someone
> off the track or to keep them chasing an answer I know.  I 
> provide the 
> best answer I can at the time and I expec
> t if it doesn't meet the need,
> someone will speak up.  If the poster does follow-up, I or 
> someone else 
> may be able to help home in on the it a little more and 
> provide a better 
> solution or pointer.  Perhaps others have a different agenda when 
> answering, although I've pretty much read every post on this 
> list for the
> last 5+ years and I've never been left with that impression.  YMMV.
> 
> So I guess what I'd like to say is, let's not throw around 
> accusations 
> of this sort.  If you receive a response to your query and 
> its not what
> you want, you're free to use it or not.  Query further if you 
> like too.
> Don't expect others have all the answers or be willing to 
> look into the
> details of all your problems.  I'm not saying that people 
> won't fix your
> problems or help you do so.  But they're going to do it their way, in 
> their time, and at their option.  If that's not what you need 
> or want, you 
> can again query further but keep in mind that you're dealing with 
> volunteers here.  Pushing may have the opposite reaction to 
> your intended 
> goal.  I actually think its a shame for people to be critical in the
> face of someone's sincere intent to help the poster address 
> their issue.
> After all, the responder is only trying to provide useful 
> information or
> be truthful about their level of personal involvement in any 
> implementation of a solution.  That all seems pretty reasonable and 
> professional to me, even if the result is not something the 
> poster wants
> to hear.  However, the impression I'm getting from the 
> discussion is that
> unless someone is willing to provide any and all support for an issue,
> in the form the poster wants it, then no response is 
> preferable to some
> response.  I guess I can live with that, if that's what the list in 
> general wants but I personally feel it would make for a much 
> less helpful
> and active community.  Maybe I'm wrong.  Maybe its time for 
> me personally
> to adjust my level of participation in Cygwin, since I see my way of
> contributing could be construed as fitting the pattern of "
> discouragement"
> as defined by others.  Hm, maybe.  I'll have to think a little more 
> about that.  As is always the case, we can all use a little more free 
> time! ;-)  Anyway, since we've all been sharing our thoughts on this 
> matter I thought I'd offer mine, since its a slightly different than
> some of the those posted earlier.  I'm really for the idea of 
> having a 
> Cygwin community.  So far, I believe its been a great 
> success.  I hope it 
> continues to be in some form! :-)  Actually, this is a good 
> time for me
> to say "thanks" to all those who work to provide and improve 
> Cygwin and
> its tools.  I don't do this enough.  This is really top-notch 
> stuff! :-)

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Steve Jorgensen]

I couldn't be left out.  It's amazing that I've had the great and timely 
help I have had from this mailing list in the mere 3 days I've been here, 
and I've asked some pretty dumb questions and given some pretty dumb 
reponses to perfectlyt good help.

A good community makes a good product, and Cygwin would be good even 
without it.  The combination is unbeatable.  Try getting this kind of 
hand-holding on the Debian newsgroup.

-----Original Message-----
From:	Michael L. Smeby, Jr. [SMTP:msmeby@nicusa.com]
Sent:	Friday, June 29, 2001 3:49 AM
To:	cygwin@cygwin.com; Larry Hall (RFK Partners, Inc)
Subject:	Re: "shouted down", "shot down", apologies

I'm a complete newbie to Cygwin, but I use it here for various reasons, and
I think that between the great people on this list and the  documentation
and FAQs all over the web, I have yet to have a question that I couldn't
find the answer to in less than ten to thirty minutes (and I've had some
obscure ones). The volunteer work on the entire project is almost
overwhelming, and I haven't seen anyone just shrug off questions or cop an
attitude.

Just my two cents worth. You guys are appreciated .... I'll shut up now and
return to lurking. :-)

===
"Ahhh, the old take over the world ploy!" -- The Mummy Returns
Regards, Michael L. Smeby, Jr.
http://www.tampagov.com
----- Original Message -----
From: "Larry Hall (RFK Partners, Inc)" <lhall@rfk.com>
To: <cygwin@cygwin.com>
Sent: Thursday, June 28, 2001 12:28 PM
Subject: Re: "shouted down", "shot down", apologies


> At 10:22 AM 6/28/2001, Christopher Faylor wrote:
> >I won't comment on the rest of your message except to say that I rarely,
> >if ever, say "use the source" when I know the answer to the question.
> >
> >I say that when I don't know and when *I* would have to look things up.
> >
> >You're welcome to continue to ask "newbie" questions as frequently as
> >you like.  That will not stop me (or others, I presume) from pointing
> >people to references or suggesting the source when I don't know the
> >answer myself.
> >
> >I don't work for anyone here.  I am not obligated to look things up to
> >make your life easier.  It is that simple.
>
> [long - those uninterested in this thread want to hit delete now! :-) ]
>
> This is exactly my approach as well.  I have to say that I'm a bit
> dismayed that folks contributing to this and the "blunt tools" thread
> have mentioned dissatisfaction with what seemed to me to be such a
> straight-forward and logical approach.  When responding to queries on
> this list, I've always followed these simple rules:
>
>    1. If I know the question is an FAQ, I point to the entry there 
(*very*

>       rarely do I just point at the FAQ without the exact entry).
>       Generally I feel there's little benefit to restating what's in the
>       FAQ.  It just doesn't seem to be a good use of my time.  If its
>       inadequate in some way, we'll hear about it and make the 
appropriate
>       change (which seems to me as it should be).
>
>    2. If I kn
> ow something specific about the subject, I respond with it.
>       Sometimes this means I have to ask a question or two before I'm
>       sure what's been tried already and whether the poster is aware of
>       a previous discussion on the subject.  That all seems like part of
>       the process to me and I don't begrudge people for it.
>
>    3. If I know that this subject has come up before and has been
>       discussed but don't remember allot of details, I point to the
>       email archives.  In this case, I don't point to a specific message,
>       although I do occasionally offer a search key that I think might
>       help find the discussion I recall.  I don't spend my time looking
>       up the exact archive entry or entries that I'm recalling.  I don't
>       even promise that the stuff I'm remembering is even helpful (though
>       that's my intent and what I'm hoping for!)  I'm just providing
>       potential source of information that may prove useful.  It may not
>       too.  If it doesn't or its too hard to find, I expect the original
>       poster will query the list again with an update of the things tried
>       and the results.  If there's no success at this point, I sometimes
>       see if there's something more specific I can find myself and post
>       that if so.
>
>    4. If the question being answered is specific and detailed enough
>       that an inspection of the source is likely to be the only path to
>       a useful answer (barring someone else who has been in the source,
>       knows the answer, and will subsequently offer it), I *suggest*
>       looking at the source.  I do this when its clear someone is a
>       developer or has mentioned they are working with some other source.
>       I mention it if I'm not sure whether the person is a developer or
>       not, usually pointing out that it is an option if they're up to
>       it.  I tend not to mention it if the person states that they have
>       no experience reading/writing code.  Generally, I don't feel
>       obli
> gated to go inspect the source to answer someone else's question,
>       although there are exceptions or times I do it anyway.
>
>    5. If I know nothing about the subject, I keep my mouth shut.
>
> I've used all five of these modes in the past on this list and seen them
> work, at least on some occasions, exactly as I expected them to.  We've
> heard back from people who've had a hard time with an FAQ entry.  We've
> heard from people who say they've searched the archives but turned up
> nothing.  We've heard back from people saying they're not capable of
> looking at the source for one reason or another.  To me, all of this
> seems reasonable dialog in the course of trying to help someone with a
> problem.  I've always felt that providing some information, be it direct
> or a pointer to something which could be helpful is better than no answer
> at all (indeed, this list has more than once in the past been berated
> for *not* responding in some way to a post!)  However, it troubles me
> that some in the recent discussions have pointed to the replies with
> references to previous discussions and the FAQ as "non-answers" (I'm
> using this term generally now although I know it was a specific member
> of the previous discussions that first offered it up and it may have
> applied in that case to a problem with the specific set of tools in use
> at the time.  I think it categorizes a general sentiment I got from
> reading these threads though).  The impression I'm left with is that
> there is at least some people on this list that feel these "non-answers"
> are offered in spite.  I'm not sure how prevalent this view is or where
> the feeling comes from.  It's certainly not my intent when I provide such
> an answer, as I've clarified above.  I know I don't sit in my chair
> reading email, jealously holding onto all the answers, and responding
> with pointers (or worse, some obtuse reference), just to throw someone
> off the track or to keep them chasing an answer I know.  I provide the
> best answer I can at the time and I expec
> t if it doesn't meet the need,
> someone will speak up.  If the poster does follow-up, I or someone else
> may be able to help home in on the it a little more and provide a better
> solution or pointer.  Perhaps others have a different agenda when
> answering, although I've pretty much read every post on this list for the
> last 5+ years and I've never been left with that impression.  YMMV.
>
> So I guess what I'd like to say is, let's not throw around accusations
> of this sort.  If you receive a response to your query and its not what
> you want, you're free to use it or not.  Query further if you like too.
> Don't expect others have all the answers or be willing to look into the
> details of all your problems.  I'm not saying that people won't fix your
> problems or help you do so.  But they're going to do it their way, in
> their time, and at their option.  If that's not what you need or want, 
you
> can again query further but keep in mind that you're dealing with
> volunteers here.  Pushing may have the opposite reaction to your intended
> goal.  I actually think its a shame for people to be critical in the
> face of someone's sincere intent to help the poster address their issue.
> After all, the responder is only trying to provide useful information or
> be truthful about their level of personal involvement in any
> implementation of a solution.  That all seems pretty reasonable and
> professional to me, even if the result is not something the poster wants
> to hear.  However, the impression I'm getting from the discussion is that
> unless someone is willing to provide any and all support for an issue,
> in the form the poster wants it, then no response is preferable to some
> response.  I guess I can live with that, if that's what the list in
> general wants but I personally feel it would make for a much less helpful
> and active community.  Maybe I'm wrong.  Maybe its time for me personally
> to adjust my level of participation in Cygwin, since I see my way of
> contributing could be construed as fitting the pattern of "
> discouragement"
> as defined by others.  Hm, maybe.  I'll have to think a little more
> about that.  As is always the case, we can all use a little more free
> time! ;-)  Anyway, since we've all been sharing our thoughts on this
> matter I thought I'd offer mine, since its a slightly different than
> some of the those posted earlier.  I'm really for the idea of having a
> Cygwin community.  So far, I believe its been a great success.  I hope it
> continues to be in some form! :-)  Actually, this is a good time for me
> to say "thanks" to all those who work to provide and improve Cygwin and
> its tools.  I don't do this enough.  This is really top-notch stuff! :-)
>
>
>
>
> Larry Hall                              lhall@rfk.com
> RFK Partners, Inc.                      http://www.rfk.com
> 118 Washington Street                   (508) 893-9779 - RFK Office
> Holliston, MA 01746                     (508) 893-9889 - FAX
>
>
>
> --
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting:         http://cygwin.com/bugs.html
> Documentation:         http://cygwin.com/docs.html
> FAQ:                   http://cygwin.com/faq/
>


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Heribert Dahms]

Hi Charles,

tar c instead of t?

Bye, Heribert (heribert_dahms@icon-gmbh.de)

> -----Original Message-----
> From:	Charles S. Wilson [SMTP:cwilson@ece.gatech.edu]
> Sent:	Thursday, June 28, 2001 01:20
> To:	John Wiersba
> Cc:	'cygwin@cygwin.com'
> Subject:	Re: "shouted down", "shot down", apologies
> 
	[Heribert]  [snip]

> For instance (readline):
>    CFLAGS="-I/usr/include/ncurses" \
>      ./configure --prefix=/usr --with-curses
>    make
>    make shared
>    make examples
>    make test
>    make install prefix=/tmp/usr
>    make install-shared prefix=/tmp/usr
>    make install-examples prefix=/tmp/usr
>    strip /tmp/usr/bin/cygreadline5.dll
>    strip /tmp/usr/bin/cyghistory5.dll
>    (cd /tmp ; tar tvjf readline-X.Y-Z.tar.bz2 usr/)
> 
> and now you have a tarball that *should* be the same as the 'official' 
> one.  In most cases, the maintainers document their build procedures or 
> special config options in /usr/doc/Cygwin/<package>.README
> 
	[Heribert]  [snip]



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[John Wiersba]

Vince,

Very well said.  I believe you're right -- there is a large cygwin
contingent which are not cygwin developers, but simply cygwin users.  In
fact I *am* a developer, as I suspect many/most cygwin users are.  But, I'm
not a *cygwin* developer, using cygwin to port anything to Windows.  I'm
using cygwin because the company I work for puts a Windows box on my
desktop.  I coined the phrase, "Cygwin...don't leave Unix without it" a
while back -- I truly feel that cygwin is a lifesaver for many of us (a BIG
thanks to all the cygwin developers and porters!).  

-- John Wiersba

> -----Original Message-----
> From: Vince Rice [ mailto:vrice@solidrocksystems.com ]
> Sent: Thursday, June 28, 2001 3:32 AM
> To: cygwin@cygwin.com
> Subject: RE: "shouted down", "shot down", apologies
> 
> 
> Chris,
> 
> Somewhere in this thread you expressed some 
> wonder/frustration at why the
> gcc/gdb projects don't generate the kind of "newbie" requests 
> Cygwin (I just
> tried to find it and it's nowhere to be found; I tend to lose 
> things at this
> time of night).  I have seen this expressed in your messages 
> countless times
> over the three years I've been lurking here.  And I've always 
> wondered at
> your wonder <g>.
> 
> GCC and gdb are tools for developers; you don't use them unless you're
> developing something.  However, Cygwin is a tool for users; 
> it's just as
> easy (easier IMHBAO) to *use* the Cygwin tools (defined as 
> the GNU tools,
> not gcc/etc.) as it is to develop on Cygwin.  One doesn't have to be a
> developer to use Unix.  One can be a raw user who has been 
> thrust into a
> Unix environment for one of a thousand reasons, or a college 
> student (or
> mid-life crisis male who I bear absolutely no resemblance to) 
> who wants to
> learn how to *use* Unix, not how to *develop* in Unix.  Your 
> outlook of
> Cygwin is the ability to develop/port Unix programs to 
> Windows.  That is
> Cygwin's reason for existence, from your (and Redhat's) and 
> much of the
> list's perspective.  However, I believe many (most?) users of 
> Cygwin could
> not care less about that.  Their view of Cygwin is the 
> ability to *run* Unix
> in Windows.
> 
> As such, we have no skills to debug.  Our approach to this 
> mailing list is
> the same as to a Word newsgroup or a game newsgroup or 
> whatever.  We want to
> know how to *use* the product, and when something doesn't 
> work we come ask
> questions.  We don't always RTM, but that's what users do (or 
> don't do as
> the case may be).  We have no desire to use the source, we 
> have no ability
> to use the source, and many of us wouldn't know source if it 
> hit them in the
> head (which with you around is a distinct possibility <bg>).
> 
> The fact that Cygwin allows developers to port Unix programs 
> to Windows is
> immaterial.  My introduction to Cygwin was from looking for 
> Unix utilities
> to run on Windows.  I have stuck around for three years 
> because the tools
> continue to get better and better, and many times, even 
> though I run from
> 4NT instead of bash, I can barely tell the difference between 
> a Windows
> prompt and a Unix prompt, because I can do durn near anything 
> I can with
> Cygwin on my PC that I can do on my clients' AIX boxes.
> 
> However, because I'm in the computer business, because I've 
> written code in
> my distant past (but not C/C++), I lurk rather than ask 
> questions, partly
> because of the attitude around here towards those who want to 
> know how to
> *use* the product and partly because I believe in your 
> philosophy of fishing
> for myself, and if I don't have time to get the pole I try not to make
> someone else do it either.
> 
> However, although I've practiced that here, I'm not 
> completely convinced
> that that is always a good thing.  Community building 
> involves interaction.
> Sometimes the way to get started is by asking questions.  If 
> someone asks me
> a question that will take me five minutes to answer but would 
> take them two
> minutes to look up themselves, then that's a waste of my time 
> and I'll tell
> them so.  But if someone asks me a question that will take 
> them three hours
> to figure out themselves when I can answer it in thirty 
> seconds, I think
> it's a little extreme to tell them to "use the source" when I 
> can save them
> three hours by just answering the question.   Whether they 
> can figure it out
> for themselves is beside the point:  I'm helping them be more 
> productive,
> just as someone helped me be more productive when I started out.
> 
> I didn't really mean to get into that.  My real point was to 
> address your
> original question -- Cygwin attracts at least as many users 
> as developers (I
> suspect far more), so they ask user questions, not developer 
> ones, and they
> think like users (hey Joe, how do I change the font on a 
> footnote?), not
> developers (hey Joe, where's the manual for that graphics 
> card, I want to
> re-write the device driver).
> 
> Vince
> 
> 
> --
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting:         http://cygwin.com/bugs.html
> Documentation:         http://cygwin.com/docs.html
> FAQ:                   http://cygwin.com/faq/
> 

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Brian Keener]

Christopher Faylor wrote:
> setup.exe so I'm very happy that you signed on.

I didn't want to let this go unanswered.  I just wanted to say thanks - I am 
too. I have learned a lot despite my sometimes frustration and my mystification 
with some of this.  I plan to continue contributing to Setup as soon as the 
blast of changes from you and Robert dies down a bit.

bk



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[John Wiersba]

Thanks!  Yes that answers my questions about how to use the source tarballs.
I was in the process of downloading all the source tarballs as this thread
was going on.  Everyone and the documentation seemed to be talking about
needing to use cvs, so my original question was about how to set up a test
environment using the source tarballs available through setup.exe.

-- John Wiersba

> -----Original Message-----
> From: Robert Collins [ mailto:robert.collins@itdomain.com.au ]
> Sent: Wednesday, June 27, 2001 6:58 PM
> To: John Wiersba; cygwin@cygwin.com
> Subject: Re: "shouted down", "shot down", apologies
> 
> 
> 
> ----- Original Message -----
> From: "John Wiersba" <John.Wiersba@medstat.com>
> > Well, what I'd like to do, and what should be standard for 
> any package
> > accepted into the cygwin archive (available to be downloaded with
> > setup.exe), is that each package source tarball should build
> out-of-the-box
> > under cygwin with a simple make configure; make; make test 
> (check?); make
> > install, including the most recent "stable" cygwin source 
> tarball.  With a
> 
> For everything but cygwin1.dll, that _should_ work. Us maintainers put
> pre-patched source (if cygwin patches are needed) into the 
> src tarball. Have
> you tried? If you've tried and a package didn't work, report 
> a bug to the
> maintainer - they will take it seriously. If you're asking a 
> hyopthetical
> "have the cygwin package maintainers done the obvious thing" 
> then I think
> you're wasting my time.
> 
> Some packages such as cygwin1.dll and gcc _WILL NOT_ build in the same
> directory as the source. For them you need
> mkdir pkgdir
> cd pkgdir
> tar xzf /path/to/tarball
> mkdir obj
> cd obj
> ../extractedpackagesrc/configure
> make
> 
> Hope that answers the question.
> 
> Rob
> 

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Wed, Jun 27, 2001 at 08:32:52PM -0400, Brian Keener wrote:
>John Wiersba wrote:
>>As Brian Keener implied in his email, I don't really want to spend a
>>huge amount of time just getting setup to successfully build the
>>source, before I can even start to contribute.  I routinely build perl
>>and quite a few GNU packages on AIX, SunOS, HP/UX and linux, but all of
>>them build relatively cleanly out-of-the-box with only minor tweaking.
>
>I think part of this is as Chris stated - maybe I over mystified some
>of the problems I had/have and the step involved in being a contributor
>or maybe it was/is my total lack of C++ programming on any platform and
>cygwin knowledge before getting started.  As a total newcomer to C++
>that found Cygwin very appealing and not having worked with anything
>using a make or a makefile since a little bit of C work - getting
>started on Cygwin was a challenge.  I guess I just bit off too much to
>start with

Well, as a total newcomer, you made some amazing contributions to
setup.exe so I'm very happy that you signed on.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Brian Keener]

John Wiersba wrote:
> As Brian Keener implied in his email, I don't really want to spend a huge
> amount of time just getting setup to successfully build the source, before I
> can even start to contribute.  I routinely build perl and quite a few GNU
> packages on AIX, SunOS, HP/UX and linux, but all of them build relatively
> cleanly out-of-the-box with only minor tweaking.

I think part of this is as Chris stated - maybe I over mystified some of the 
problems I had/have and the step involved in being a contributor or maybe it 
was/is my total lack of C++ programming on any platform and cygwin knowledge 
before getting started.  As a total newcomer to C++ that found Cygwin very 
appealing and not having worked with anything using a make or a makefile since 
a little bit of C work - getting started on Cygwin was a challenge.  I guess I 
just bit off too much to start with - I was interested and I looked at the code 
and thought I could work with it and make some contributions - so I did, but 
coming from nothing it was tough and in some case I do but I still do not 
understand.  At any rate - from what it sounds like your knowledge level is - 
you will have no problems.  

Also as Chris pointed out in his response to me - once it is configured 
properly then there is really no need for the gyrations and the reconfiguring I 
seem to go through and as well as we all say - being on Win95 for me does not 
help.  I don't think I am there yet as far as setup properly but I do get 
closer all the time.  

> Version of cygwin:  
>    1) Download all source for all cygwin packages using setup.exe
>    2) Install all source
>    3) Run some master script to loop through all the source packages and
> build them binary-image packages
>    4) Install the binary-image packages into a test directory
>    5) Exit all cygwin processes
>    6) Run a version of cygwin.bat which points to the test directory to test
> it out.
You really need a step (.5) here - in order to build the cygwin and the other 
sources you have to have installed the binaries for cygwin and some of the 
other packages first.  This is probably understood but I thought I would point 
it out.

bk



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Charles S. Wilson]

Oh yeah -- one other thing.  Most maintainers use separate build and src 
trees.  I don't -- I build "my" packages right in the source directory. 
  However, you *MUST* use separate build and src trees for gcc, 
binutils, and cygwin -- or it won't work.  The "directions" below for 
readline assume the build-in-source-tree model.

I wonder if the following is a useful thing to have in the 
documentation: at set of build instructions (and nothin' but build 
instructions) for each package.  Say, a web page with a bunch of links 
to 10 line txt files (bash scripts?) that each build a given package.

I'm willing to provide that for "my" packages, if somebody else will 
handle the logistics -- and folks think it is a good idea.  (I shudder 
to consider the binutils, gcc, and cygwin scripts -- that python monster 
someone posted earlier today scared even me...)

--Chuck

Charles S. Wilson wrote:

> John Wiersba wrote:
> 
>> So, is the "download source" option for setup.exe useful for 
>> anything?  Can
>> you rebuild (all of) cygwin from it?  What I mean is:  there are these 
>> 80+
>> packages available with setup.exe.  If I download the source for all of
>> them, install the source somewhere, run some build process against it, 
>> will
>> I end up with a working cygwin which is functionally equivalent to the
>> binary packages I downloaded using setup.exe?
>> 
>> -- John Wiersba
> 
> 
> Yes, you can build the tools from the source packages provided as part 
> of the downloads.  However, there is no 'make world' procedure that 
> builds EVERYTHING all in one step.  You have to go into each package's 
> source directory, and do the typical './configure ; make ; make install' 
> procedure.  However, to get EXACTLY the same binaries, you'll need to 
> use the same configuration options that the maintainer used --
> 
> For instance (readline):
>   CFLAGS="-I/usr/include/ncurses" \
>     ./configure --prefix=/usr --with-curses
>   make
>   make shared
>   make examples
>   make test
>   make install prefix=/tmp/usr
>   make install-shared prefix=/tmp/usr
>   make install-examples prefix=/tmp/usr
>   strip /tmp/usr/bin/cygreadline5.dll
>   strip /tmp/usr/bin/cyghistory5.dll
>   (cd /tmp ; tar tvjf readline-X.Y-Z.tar.bz2 usr/)
> 
> and now you have a tarball that *should* be the same as the 'official' 
> one.  In most cases, the maintainers document their build procedures or 
> special config options in /usr/doc/Cygwin/<package>.README
> 
> The exception to this is possibly binutils, gcc, cygwin, mingw.  These 
> packages are kindof intermingled and cgf uses special buildscripts to 
> generate the separate binary tarballs.  I think.
> 
> --Chuck


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Brian Keener]

John Wiersba wrote:
> So, is the "download source" option for setup.exe useful for anything?  

Without a doubt - that is the source code for the packaged version you 
selected.

> Can
> you rebuild (all of) cygwin from it?  What I mean is:  there are these 80+
> packages available with setup.exe.  If I download the source for all of
> them, install the source somewhere, run some build process against it, will
> I end up with a working cygwin which is functionally equivalent to the
> binary packages I downloaded using setup.exe?
>
With my limited knowledge - I would say the answer is yes.  With the 
appropriate Configure command and make commands you would end up with the same 
thing you would get from setup if you downloaded the binary packages.  This 
would be the current packaged software for whatever version(s) you selected.  

If you wanted changes that might not be in these packages then you would need 
to go to cvs to get the more recent changes that may not have been packaged 
yet.  And if you wanted even more recent changes you would want to get a 
snapshot.  Or at least that's my interpretation.

I only work with the cvs and have never downloaded a source tarball or any 
snapshots - dunna know how.  I believe (through the help of others) I actually 
have a script that does accomplish for me updating from cvs to my copy of the 
source and also a second script(s) which accomplish a make clean, configuring, 
building and one that makes an install directory. I have never actually used 
the install directory to override my cygwin and the like but I do occasionally 
run it just to see them in the install directory.

As I said - what I have is for the way I work with the cvs - if you would like 
a copy let me know.  I have never tried using just the source tarballs.  My 
disk, drive probably wouldn't be big enough.

Bk



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Charles S. Wilson]

John Wiersba wrote:

> So, is the "download source" option for setup.exe useful for anything?  Can
> you rebuild (all of) cygwin from it?  What I mean is:  there are these 80+
> packages available with setup.exe.  If I download the source for all of
> them, install the source somewhere, run some build process against it, will
> I end up with a working cygwin which is functionally equivalent to the
> binary packages I downloaded using setup.exe?
> 
> -- John Wiersba

Yes, you can build the tools from the source packages provided as part 
of the downloads.  However, there is no 'make world' procedure that 
builds EVERYTHING all in one step.  You have to go into each package's 
source directory, and do the typical './configure ; make ; make install' 
procedure.  However, to get EXACTLY the same binaries, you'll need to 
use the same configuration options that the maintainer used --

For instance (readline):
   CFLAGS="-I/usr/include/ncurses" \
     ./configure --prefix=/usr --with-curses
   make
   make shared
   make examples
   make test
   make install prefix=/tmp/usr
   make install-shared prefix=/tmp/usr
   make install-examples prefix=/tmp/usr
   strip /tmp/usr/bin/cygreadline5.dll
   strip /tmp/usr/bin/cyghistory5.dll
   (cd /tmp ; tar tvjf readline-X.Y-Z.tar.bz2 usr/)

and now you have a tarball that *should* be the same as the 'official' 
one.  In most cases, the maintainers document their build procedures or 
special config options in /usr/doc/Cygwin/<package>.README

The exception to this is possibly binutils, gcc, cygwin, mingw.  These 
packages are kindof intermingled and cgf uses special buildscripts to 
generate the separate binary tarballs.  I think.

--Chuck



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Robert Collins]

----- Original Message -----
From: "John Wiersba" <John.Wiersba@medstat.com>
> Well, what I'd like to do, and what should be standard for any package
> accepted into the cygwin archive (available to be downloaded with
> setup.exe), is that each package source tarball should build
out-of-the-box
> under cygwin with a simple make configure; make; make test (check?); make
> install, including the most recent "stable" cygwin source tarball.  With a

For everything but cygwin1.dll, that _should_ work. Us maintainers put
pre-patched source (if cygwin patches are needed) into the src tarball. Have
you tried? If you've tried and a package didn't work, report a bug to the
maintainer - they will take it seriously. If you're asking a hyopthetical
"have the cygwin package maintainers done the obvious thing" then I think
you're wasting my time.

Some packages such as cygwin1.dll and gcc _WILL NOT_ build in the same
directory as the source. For them you need
mkdir pkgdir
cd pkgdir
tar xzf /path/to/tarball
mkdir obj
cd obj
../extractedpackagesrc/configure
make

Hope that answers the question.

Rob


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[John Wiersba]

> -----Original Message-----
> From: Greg Smith [ mailto:gsmith@nc.rr.com ]
> Sent: Wednesday, June 27, 2001 6:35 PM
> To: cygwin@cygwin.com
> Subject: Re: "shouted down", "shot down", apologies
> 
> 
> John Wiersba wrote:
>
> > Back to my original question (building entire cygwin suite 
> > from source):
> 
> Umm, what we're talking here is not building the packages 
> that are provided
> with Cygwin, but rather the support programs that enable 
> Cygwin to do its thing.
> For example, cygwin1.dll
> 
> To compile/build the other stuff, you'll need to download the 
> package sources and
> follow the build instructions for each package.

Well, what I'd like to do, and what should be standard for any package
accepted into the cygwin archive (available to be downloaded with
setup.exe), is that each package source tarball should build out-of-the-box
under cygwin with a simple make configure; make; make test (check?); make
install, including the most recent "stable" cygwin source tarball.  With a
little extra work (getting them to install to a directory hierarchy not
rooted at /) I should be able to rebuild each of the binary tarballs
downloadable with setup.exe.

If I want to build the bleeding edge cygwin package, I'll have to use cvs to
get it.  But right now, all I want to do is rebuild the "stable" packages.

-- John Wiersba

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Christopher Faylor]

On Wed, Jun 27, 2001 at 06:15:00PM -0400, John Wiersba wrote:
>The FAQ mentions that
>
>"Normally, this will also attempt to build the documentation, which
>additionally requires db2html, texi2html and possibly others. These tools
>are not included in
>the Cygwin distribution, but are readily obtainable"
>
>Is this still true?  texi2html is provided via setup.exe, but db2html is
>not.  Can it really be that each contributor has downloaded the source for
>docbook-tools and built it before building the cygwin source?

I build things on linux so I have downloaded this via RPM.  I did a google
search and found several references to db2html.

Maybe someone will post a definitive source.  You don't need to build the
documentation, though.  The makefile will skip this step if it can't build
the docs.

>Back to my original question (building entire cygwin suite from
>source):
>
>As Brian Keener implied in his email, I don't really want to spend a
>huge amount of time just getting setup to successfully build the
>source, before I can even start to contribute.  I routinely build perl
>and quite a few GNU packages on AIX, SunOS, HP/UX and linux, but all of
>them build relatively cleanly out-of-the-box with only minor tweaking.
>Since cygwin can be used to build cygwin, what I'm hoping for is that
>the source for cygwin available from setup.exe will also build cleanly
>out-of-the-box.

There is a certain amount of setup available but I really don't advocate
using the cygwin sources from setup.exe for building anything.  The
sources change rapidly in CVS so you probably will be automatically out
of date if you use the sources for cygwin 1.3.2.

If you can't use cvs, then the snapshot sources should be adequate.  If
you extract the cygwin-src-yyymmdd.tar.bz2 file you should be able to
build.

Other individual packages may be buildable with some tweaking.

>It's certainly should be possible.  I won't fault anyone if it's not
>currently in that state, since it may take a reasonable amount of work
>to get it there.  But if it's not there yet, then that would be a nice
>goal which would certainly lower the barrier for new contributors.  If
>it's already in that state, then it should be a relatively painless
>process to get setup to build a test

I am not sure how useful it is to be able to build the entire cygwin
release from scratch.  That sort of goes counter to the new philsophy of
updating individual packages on the fly.  I can't see why anyone would
want to build perl and postgres together.  I think the chances of
accomplishing this are low.

Although, now that I've said that, I will point out that I have a
/netrel/src directory that looks like this:

binutils-20010425-2  config.sub      cygwin-1.3.1-1      gcc-2.95.2-9    inetutils-1.3.2   mpw-install     tcltk
bison-1.28-1         configure       cygwin-1.3.2-1      gcc-2.95.3-6    install-sh        ocygwin-1.1.4   tcltk-20010307-1
bzip2-20001120-1     configure.in    cygwin-1.3.3-1      gdb-20010428-1  less-358-3        ogcc            termcap-20001216-1
common               cygwin-1.1.5-7  dejagnu-20010117-1  grep-2.4.2-1    make-3.79.1-4     sh-utils-2.0-2  texinfo-4.0-4
config-ml.in         cygwin-1.1.6-1  expect-20010117-1   groff-1.16.1-1  mfput.log         t.c             textutils-2.0-2
config.guess         cygwin-1.1.7-1  fileutils-4.1-2     gzip-1.3-1      mingw-20010424-1  tar-1.13.18-3   w32api-20010520-1
config.if            cygwin-1.1.8-2  flex-2.5.4-1        include.gcc     mkinstalldirs     tar-1.13.19-1   zsh-4.1.0-dev-0

I also have a 'mknetrel' program that allows you to say:

mknetrel flex

to build flex-2.54-1.

I plan on checking this into a cygwin-apps repository on sources.redhat.com
soon.

This is not a generic "build everything from one makefile" solution but
it does provide a measure of consistency in building packages.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Greg Smith]

John Wiersba wrote:
>
> The FAQ mentions that
>
> "Normally, this will also attempt to build the documentation, which
> additionally requires db2html, texi2html and possibly others. These tools
> are not included in
> the Cygwin distribution, but are readily obtainable"
>
> Is this still true?  texi2html is provided via setup.exe, but db2html is
> not.  Can it really be that each contributor has downloaded the source for
> docbook-tools and built it before building the cygwin source?

As I tell my programmers, `I know one way how you can find out' ;-)

> Back to my original question (building entire cygwin suite from source):

Umm, what we're talking here is not building the packages that are provided
with Cygwin, but rather the support programs that enable Cygwin to do its thing.
For example, cygwin1.dll

To compile/build the other stuff, you'll need to download the package sources and
follow the build instructions for each package.

Greg

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[John Wiersba]

The FAQ mentions that

"Normally, this will also attempt to build the documentation, which
additionally requires db2html, texi2html and possibly others. These tools
are not included in
the Cygwin distribution, but are readily obtainable"

Is this still true?  texi2html is provided via setup.exe, but db2html is
not.  Can it really be that each contributor has downloaded the source for
docbook-tools and built it before building the cygwin source?

Back to my original question (building entire cygwin suite from source):

As Brian Keener implied in his email, I don't really want to spend a huge
amount of time just getting setup to successfully build the source, before I
can even start to contribute.  I routinely build perl and quite a few GNU
packages on AIX, SunOS, HP/UX and linux, but all of them build relatively
cleanly out-of-the-box with only minor tweaking.  Since cygwin can be used
to build cygwin, what I'm hoping for is that the source for cygwin available
from setup.exe will also build cleanly out-of-the-box.  It's certainly
should be possible.  I won't fault anyone if it's not currently in that
state, since it may take a reasonable amount of work to get it there.  But
if it's not there yet, then that would be a nice goal which would certainly
lower the barrier for new contributors.  If it's already in that state, then
it should be a relatively painless process to get setup to build a test
version of cygwin:  
   1) Download all source for all cygwin packages using setup.exe
   2) Install all source
   3) Run some master script to loop through all the source packages and
build them binary-image packages
   4) Install the binary-image packages into a test directory
   5) Exit all cygwin processes
   6) Run a version of cygwin.bat which points to the test directory to test
it out.

-- John Wiersba

> -----Original Message-----
> From: Christopher Faylor [ mailto:cgf@redhat.com ]
> Sent: Wednesday, June 27, 2001 4:47 PM
> To: cygwin@cygwin.com
> Subject: Re: "shouted down", "shot down", apologies
> 
> 
> On Wed, Jun 27, 2001 at 04:34:30PM -0400, John Wiersba wrote:
> >OK, given the "new spirit of cooperation" expressed in cgf's 
> email, I'd
> >like to be in a position to help, too, in the sense of 
> "submit patches"
> >rather than "find bugs, suggest new features".  But there's primarily
> >one thing stopping me: a test environment.  I'm currently ignorant of
> >how to effectively use cvs and I don't have much burning 
> desire to add
> >csv to my personal toolkit right now.
> 
> Well, just to backslide a little: http://cygwin.com/contrib.html ,
> http://cygwin.com/cvs.html .
> 
> Or, more generically http://cygwin.com/ .
> 
> cgf
> 
> --
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting:         http://cygwin.com/bugs.html
> Documentation:         http://cygwin.com/docs.html
> FAQ:                   http://cygwin.com/faq/
> 

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[Robert McNulty Junior]

Yes. It is available though setup.exe 2.57.
I use it all the time. I have a CD here where I can get the sources any time
I want. And I keep it updated constantly.


-----Original Message-----
From: cygwin-owner@sources.redhat.com
[ mailto:cygwin-owner@sources.redhat.com]On Behalf Of John Wiersba
Sent: Wednesday, June 27, 2001 4:50 PM
To: 'cygwin@cygwin.com'
Subject: RE: "shouted down", "shot down", apologies

So, is the "download source" option for setup.exe useful for anything?  Can
you rebuild (all of) cygwin from it?  What I mean is:  there are these 80+
packages available with setup.exe.  If I download the source for all of
them, install the source somewhere, run some build process against it, will
I end up with a working cygwin which is functionally equivalent to the
binary packages I downloaded using setup.exe?

-- John Wiersba

> -----Original Message-----
> From: Christopher Faylor [ mailto:cgf@redhat.com ]
> Sent: Wednesday, June 27, 2001 4:47 PM
> To: cygwin@cygwin.com
> Subject: Re: "shouted down", "shot down", apologies
>
>
> On Wed, Jun 27, 2001 at 04:34:30PM -0400, John Wiersba wrote:
> >OK, given the "new spirit of cooperation" expressed in cgf's
> email, I'd
> >like to be in a position to help, too, in the sense of
> "submit patches"
> >rather than "find bugs, suggest new features".  But there's primarily
> >one thing stopping me: a test environment.  I'm currently ignorant of
> >how to effectively use cvs and I don't have much burning
> desire to add
> >csv to my personal toolkit right now.
>
> Well, just to backslide a little: http://cygwin.com/contrib.html ,
> http://cygwin.com/cvs.html .
>
> Or, more generically http://cygwin.com/ .
>
> cgf
>
> --
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting:         http://cygwin.com/bugs.html
> Documentation:         http://cygwin.com/docs.html
> FAQ:                   http://cygwin.com/faq/
>

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/




--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: "shouted down", "shot down", apologies

[John Wiersba]

So, is the "download source" option for setup.exe useful for anything?  Can
you rebuild (all of) cygwin from it?  What I mean is:  there are these 80+
packages available with setup.exe.  If I download the source for all of
them, install the source somewhere, run some build process against it, will
I end up with a working cygwin which is functionally equivalent to the
binary packages I downloaded using setup.exe?

-- John Wiersba

> -----Original Message-----
> From: Christopher Faylor [ mailto:cgf@redhat.com ]
> Sent: Wednesday, June 27, 2001 4:47 PM
> To: cygwin@cygwin.com
> Subject: Re: "shouted down", "shot down", apologies
> 
> 
> On Wed, Jun 27, 2001 at 04:34:30PM -0400, John Wiersba wrote:
> >OK, given the "new spirit of cooperation" expressed in cgf's 
> email, I'd
> >like to be in a position to help, too, in the sense of 
> "submit patches"
> >rather than "find bugs, suggest new features".  But there's primarily
> >one thing stopping me: a test environment.  I'm currently ignorant of
> >how to effectively use cvs and I don't have much burning 
> desire to add
> >csv to my personal toolkit right now.
> 
> Well, just to backslide a little: http://cygwin.com/contrib.html ,
> http://cygwin.com/cvs.html .
> 
> Or, more generically http://cygwin.com/ .
> 
> cgf
> 
> --
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting:         http://cygwin.com/bugs.html
> Documentation:         http://cygwin.com/docs.html
> FAQ:                   http://cygwin.com/faq/
> 

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: "shouted down", "shot down", apologies

[Tak Ota]

Applause...

I've been quietly observing the arguments with some nervous feeling.
I now feel much better as being a part of this list.

-Tak

On Wed, 27 Jun 2001 16:16:30 -0400, Christopher Faylor <cgf@redhat.com> wrote:

> I have been concerned by two recent messages where people have felt
> that their ideas have been "shouted down" or "shot down".
> 
> That bothers me.  It bothers me because I assume that most, if not
> all of the negative perception undoubtedly came from me.
> 
> I do have some strong opinions on how some things should be done.  For
> instance, I think that overloading a FAQ with excessive information is
> counter productive.
> 
> I also feel that the majority of "newbie" requests here do not come from
> people who have exhaustively studied available documentation.
> 
> So, filling the FAQ with non-frequently asked questions does not seem like
> the way to go to me.  It seems like it will make the FAQ harder to navigate
> and will make it easier for people to miss things.
> 
> Telling people that the way to use google is to type something like:
> 
> http://www.google.com/search?q=cygwin+ssh&btnG=Google+Search
> 
> does not make sense to me.
> 
> Updating the documentation *does* make sense to me.
> 
> Some recent email of mine may have made it sound like I am an inflexible
> bastard.  I regret sending it.
> 
> I'm open to new ideas but I sometimes need to be convinced.  And, even
> when convinced, it does not necessarily follow that I will now make
> it my life's mission to carry out the new ideas.
> 
> I've said that repetition is important, so I'll repeat it one more time:
> If you want to see something change, don't "suggest".  Don't "it seems
> to me".  Don't "It would be nice".
> 
> Please reorient your thinking from "This is what they should do" to
> "This is what I can do".
> 
> If I have dropped the ball on someone volunteering or if I have rudely
> shot down your offer to help then I sincerely apologize.  I know that
> my attempts at humor have sometimes been interpreted as rudeness.  I
> know that sometimes I get impatient with ignorance (you can ask my
> family about this trait), especially intractable ignorance.
> 
> Regardless, I have no real excuse.  I am sometimes exasperated and mean.
> I hope that it is clear that I am doing what I'm doing because I want to
> help.  In some cases, I'm even doing things that I come close to
> detesting, like maintaining gcc or make.  I do this because I know that
> it is important to people even though it is really not my specialty.
> 
> And, I also enjoy running a project like Cygwin.  I think that the net
> release of Cygwin has improved dramatically in the last couple of years.
> That is because I've lobbied for changes inside of Red Hat and solicited
> active maintainers outside of Red Hat.  And, I've encouraged the
> development of the cygwin installer.
> 
> There is still lots and lots and lots^10 of room for improvement.  I
> would like to improve the documentation.  I would really like to expand
> the cygwin test suite.  There are still problems with cygwin signals
> and the cygwin spawn command.  setup.exe could stand all sorts of
> improvement.
> 
> I actually have a tendency to just see all of the negatives in cygwin.
> I have to keep reminding myself that people are using it successfully
> every day.  Most of them don't care that zip stores full MS-DOS paths
> or that spawn(_P_NOWAIT) doesn't work on non-cygwin programs.
> 
> Anyway, if someone has volunteered and I have dropped the ball, please
> ping me again.  I'll try to rectify my mistake in not acting on your
> offer.
> 
> If someone has suggested an idea and didn't appreciate my response,
> then I also apologize.  I'll try to do better in the future.
> 
> (Although, I will probably still try to be "humorous" from time to
> time.  Be warned.)
> 
> cgf
> 
> --
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting:         http://cygwin.com/bugs.html
> Documentation:         http://cygwin.com/docs.html
> FAQ:                   http://cygwin.com/faq/

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

"shouted down", "shot down", apologies

[Christopher Faylor]

I have been concerned by two recent messages where people have felt
that their ideas have been "shouted down" or "shot down".

That bothers me.  It bothers me because I assume that most, if not
all of the negative perception undoubtedly came from me.

I do have some strong opinions on how some things should be done.  For
instance, I think that overloading a FAQ with excessive information is
counter productive.

I also feel that the majority of "newbie" requests here do not come from
people who have exhaustively studied available documentation.

So, filling the FAQ with non-frequently asked questions does not seem like
the way to go to me.  It seems like it will make the FAQ harder to navigate
and will make it easier for people to miss things.

Telling people that the way to use google is to type something like:

http://www.google.com/search?q=cygwin+ssh&btnG=Google+Search

does not make sense to me.

Updating the documentation *does* make sense to me.

Some recent email of mine may have made it sound like I am an inflexible
bastard.  I regret sending it.

I'm open to new ideas but I sometimes need to be convinced.  And, even
when convinced, it does not necessarily follow that I will now make
it my life's mission to carry out the new ideas.

I've said that repetition is important, so I'll repeat it one more time:
If you want to see something change, don't "suggest".  Don't "it seems
to me".  Don't "It would be nice".

Please reorient your thinking from "This is what they should do" to
"This is what I can do".

If I have dropped the ball on someone volunteering or if I have rudely
shot down your offer to help then I sincerely apologize.  I know that
my attempts at humor have sometimes been interpreted as rudeness.  I
know that sometimes I get impatient with ignorance (you can ask my
family about this trait), especially intractable ignorance.

Regardless, I have no real excuse.  I am sometimes exasperated and mean.
I hope that it is clear that I am doing what I'm doing because I want to
help.  In some cases, I'm even doing things that I come close to
detesting, like maintaining gcc or make.  I do this because I know that
it is important to people even though it is really not my specialty.

And, I also enjoy running a project like Cygwin.  I think that the net
release of Cygwin has improved dramatically in the last couple of years.
That is because I've lobbied for changes inside of Red Hat and solicited
active maintainers outside of Red Hat.  And, I've encouraged the
development of the cygwin installer.

There is still lots and lots and lots^10 of room for improvement.  I
would like to improve the documentation.  I would really like to expand
the cygwin test suite.  There are still problems with cygwin signals
and the cygwin spawn command.  setup.exe could stand all sorts of
improvement.

I actually have a tendency to just see all of the negatives in cygwin.
I have to keep reminding myself that people are using it successfully
every day.  Most of them don't care that zip stores full MS-DOS paths
or that spawn(_P_NOWAIT) doesn't work on non-cygwin programs.

Anyway, if someone has volunteered and I have dropped the ball, please
ping me again.  I'll try to rectify my mistake in not acting on your
offer.

If someone has suggested an idea and didn't appreciate my response,
then I also apologize.  I'll try to do better in the future.

(Although, I will probably still try to be "humorous" from time to
time.  Be warned.)

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/