problem with form for applying for cvs write privileges

problem with form for applying for cvs write privileges

[Caroline Tice]

[-- Attachment #1: Type: text/plain, Size: 1158 bytes --]


Hello,

I am a new compiler developer at Apple, and I will be working on the 
Apple version of
gcc (I have already submitted one patch for approval) and will need cvs 
write access for
checking in my changes.

I went  to the form (http;//sources.redhat.com/cgi-bin/pdw/ps_form.cgi 
and tried to enter
the information.  It won't accept my .ssh key.  I believe the problem 
is that my .ssh
key is a protocol 2 key, not a protocol 1 key, and it contains slashes 
which the form is
interpreting as line-breaks rather than characters.

I spoke with Geoffrey Keating (geoffk@apple.com) about the problem and 
he suggested
that I send you email explaining the problem and giving you the 
necessary information,
and you could create the access for me by hand.   Please let me know if 
there is
anything else I should do.  Thank you.

Sincerely,
Caroline Tice
ctice@apple.com

First & Last Name:  Caroline Tice
Prefered Login:       ctice
Email Address:       ctice@apple.com
Project:                     gcc
Email address of person who approved request:  geoffk@apple.com

.ssh/id_dsa.pub:  I am attaching the file to prevent cut/paste errors 
in the email system.

[-- Attachment #2: id_dsa.pub --]
[-- Type: application/octet-stream, Size: 612 bytes --]

ssh-dss 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 ctice@athena.apple.com

Re: problem with form for applying for cvs write privileges

[Gerald Pfeifer]

On Wed, 29 Jan 2003, Caroline Tice wrote:
> I am a new compiler developer at Apple, and I will be working on the
> Apple version of gcc (I have already submitted one patch for approval)
> and will need cvs write access for checking in my changes.

Usually we only provide CVS write access to developers who have
contributed regularily, not to those who perhaps will contribute
regularily.

That has been our policy, as far as I understand (though I am aware
that, notably in the case of Red Hat folks. it hasn't always been
followed strictly).

Gerald
-- 
Gerald Pfeifer (Jerry)   gerald@pfeifer.com   http://www.pfeifer.com/gerald/

Re: problem with form for applying for cvs write privileges

[Christopher Faylor]

On Thu, Jan 30, 2003 at 10:35:51AM +0100, Gerald Pfeifer wrote:
>On Wed, 29 Jan 2003, Caroline Tice wrote:
>> I am a new compiler developer at Apple, and I will be working on the
>> Apple version of gcc (I have already submitted one patch for approval)
>> and will need cvs write access for checking in my changes.
>
>Usually we only provide CVS write access to developers who have
>contributed regularily, not to those who perhaps will contribute
>regularily.
>
>That has been our policy, as far as I understand (though I am aware
>that, notably in the case of Red Hat folks. it hasn't always been
>followed strictly).

In the case of Red Hat folks the policy *is* followed strictly.  When
sourceware was first set up ages ago, I believe every engineer in Cygnus
had an account on it.  Since that time we grant write access using the
same rules as everyone else.

However, the policy is only that a maintainer has to elect you before you
are granted access.  It seemed like Geoff Keating was doing this.  Was
that not the case?  If it wasn't, I'll revoke the account.

cgf

Re: problem with form for applying for cvs write privileges

[Gerald Pfeifer]

On Thu, 30 Jan 2003, Christopher Faylor wrote:
>> Usually we only provide CVS write access to developers who have
>> contributed regularily, not to those who perhaps will contribute
>> regularily.
>>
> >That has been our policy, as far as I understand [...]
> In the case of Red Hat folks the policy *is* followed strictly.

Sorry, that was ambiguous: from the overseers side, the policy definitely
has been followed strictly.

What I ment is that (apparently) in the case of some Red Hat employees
GCC maintainers sponsored accounts relatively liberaly. I just checked
our MAINTAINERS file, and if you do an

  egrep \
    'Tom Rix|Jeff Knaggs|Matthew Hiller|Chandra Chavva|Michael Tiemann' \
    ChangeLog* */ChangeLog* */*/ChangeLog*

you'll see what I mean (Michael Tiemann being a very special case,
though ;-) ).

> However, the policy is only that a maintainer has to elect you before
> you are granted access.  It seemed like Geoff Keating was doing this.
> Was that not the case?  If it wasn't, I'll revoke the account.

If Geoff okayed it, it's okay. My response was more like "Perhaps we
(GCC folks) should not hand out accounts too easily, given security
considerations and long term maintainance".

Gerald
-- 
Gerald Pfeifer (Jerry)   gerald@pfeifer.com   http://www.pfeifer.com/gerald/

Re: problem with form for applying for cvs write privileges

[Jonathan Larmour]

Christopher Faylor wrote:
> 
> In the case of Red Hat folks the policy *is* followed strictly.  When
> sourceware was first set up ages ago, I believe every engineer in Cygnus
> had an account on it. 

<nitpick>Toolchain engineers maybe, but not all engineers.</nitpick>

Jifl
-- 
--[ "You can complain because roses have thorns, or you ]--
--[  can rejoice because thorns have roses." -Lincoln   ]-- Opinions==mine

Re: problem with form for applying for cvs write privileges

[law]

In message <3E394421.6040103@jifvik.org>, Jonathan Larmour writes:
 >Christopher Faylor wrote:
 >> 
 >> In the case of Red Hat folks the policy *is* followed strictly.  When
 >> sourceware was first set up ages ago, I believe every engineer in Cygnus
 >> had an account on it. 
 >
 ><nitpick>Toolchain engineers maybe, but not all engineers.</nitpick>
Not even toolchain engineers -- we didn't give them accounts until
they were contributing on a semi-regular basis.

jeff

Re: problem with form for applying for cvs write privileges

[Andrew Cagney]

Just FYI, GDB's policy reads:

> 			Write After Approval
> 			   (alphabetic)
> 
> To get recommended for the Write After Approval list you need a valid
> FSF assignment and have submitted one good patch.

Andrew

Re: problem with form for applying for cvs write privileges

[Christopher Faylor]

On Thu, Jan 30, 2003 at 03:26:25PM +0000, Jonathan Larmour wrote:
>Christopher Faylor wrote:
>>In the case of Red Hat folks the policy *is* followed strictly.  When
>>sourceware was first set up ages ago, I believe every engineer in
>>Cygnus had an account on it.
>
><nitpick>Toolchain engineers maybe, but not all engineers.</nitpick>

Oh.  Didn't know that.  That's strange given some of the folks I saw in
/etc/passwd.

cgf

Re: problem with form for applying for cvs write privileges

[Christopher Faylor]

On Thu, Jan 30, 2003 at 08:35:47AM -0700, law@redhat.com wrote:
>In message <3E394421.6040103@jifvik.org>, Jonathan Larmour writes:
> >Christopher Faylor wrote:
> >> 
> >> In the case of Red Hat folks the policy *is* followed strictly.  When
> >> sourceware was first set up ages ago, I believe every engineer in Cygnus
> >> had an account on it. 
> >
> ><nitpick>Toolchain engineers maybe, but not all engineers.</nitpick>
>Not even toolchain engineers -- we didn't give them accounts until
>they were contributing on a semi-regular basis.

Ok.  I stand corrected.  Sorry.  I shouldn't have been making definitive statements
about something I wasn't responsible for.

cgf