public inbox for
 help / color / mirror / Atom feed
* "Unusual" contribution licensing problems
@ 2005-10-20 22:06 Érsek László
  2005-10-20 22:43 ` Brian Dessent
  0 siblings, 1 reply; 2+ messages in thread
From: Érsek László @ 2005-10-20 22:06 UTC (permalink / raw)
  To: cygwin-licensing

Valued Cygwin Team,

after grepping the cygwin mailing list and my up-to-date cygwin
installation for "nftw" and "fts_open", I thought that it could make sense
(and fun) to implement nftw().

It works for me, so I'd like to give it to cygwin. Of course your
acceptance depends on at least two factors: licensing and technical merit.
I'd like to ask about licensing first because if it's legally impossible
for me to contribute this code to cygwin, I'm not interested in its
further development (I mean that of my code, not of cygwin). And your
answers could be of value for others as well, even if my code proves naive
or if nftw() is already in cygwin and it's just me who didn't look hard
enough. (In the latter case, I apologize for the noise.)

I've read


some time ago.

Let me list some facts and guesses:

1 "Not very interesing"

1.1 I developed the code on my home machine, GNU/Linux installation, in my
free time, alone. The cygwin installation I used for testing is on a
laptop owned by my employer. (I used the laptop for this purpose only in
my free time at home, alone. Such use of the laptop is not prohibited.)

1.2 I'm employed as a developer. I think I can get a written disclaimer
from my employer for the code.

1.3 As an alumnus of a university, I still have (legal) accounts on some
servers of the university. I used those accounts to compile and test my
code (besides the cygwin installation mentioned above, of course). I don't
know whether I need some written permission from the university.

1.4 I didn't look at other nftw() sources. However, I did write a small
test program and (legally) observed the behavior of other nftw() (and
ftw()) implementations. I also strace'd some of the executables.

1.5 As specification, I used the SUSv[123] of The Open Group (which I
think I'm entitled to download for reading on my own monitor) and various
info and man pages.

2 "More interesing"

2.1 I want to give this code (as a gift) to a company which develops
proprietary software _after_ (if at all) cygwin has accepted my
contribution. I wouldn't mind if they sold a (-n improved) version of my
code without even mentioning my name. (I don't intend to make money with
this code myself.) I'd simply like to have nftw() functionality in that
libc, even if they wouldn't allow me to publicly brag about my code in
their libc (if that nftw() was my code at all). I didn't contact them yet;
cygwin and free availability comes first.

2.2 That's why I don't think about assigning copyright to anybody but
placing my nftw() code into the public domain.

3 "Problematic"

3.1 In my country, Hungary (member of the EU since May 1, 2004, if that
matters), copyright is known and regulated. Copyright can be divided into
two groups of "subrights": the one of "financial" or proprietary rights,
and the one of personal rights.

3.2 In my country, everything one creates becomes automatically
copyrighted by him/her, and it can _not_ be placed into the public domain
at once.

3.3 One can trade or give away its proprietary copyright "subrights" like
any other property.

3.4 One can _not_ alienate one's own personal copyright "subrights".

3.5 Copyright expires after 70 years after creation (or first known
publication, if creation time is unknown).

3.5 I might try to sign a copyright assignment contract, but through that
I would perhaps break the law; or the contract to sign would be void in
Hungary right from conception. This depends on whether the contract in
question assigns proprietary or personal rights.

3.6 An example. Formally, the GPL is an invalid contract, AFAIK, for
Hungarian authors. (Possible) reasons: (i) it is only available in
electronic form (when slapped into a package under the name COPYING), (ii)
it is not a contract between concrete parties, (iii) it tries to alienate
non-alienateable rights, (iv) it tries to do so in an irrevokable way.

3.7 Since employed software developers are generally compensated for their
employment work with their salary and then their work fully belongs
(whatever that means) to their employers, it should be possible to "get
rid" of every relevant copyright "subright". Maybe I should "sell" this
code to somebody for 1 cent who promises to place it into the public
domain in some country where that's possible. (Maybe I should have done
that myself through some fake nick and/or IP address, but what I _don't_
want to lose if my code qualifies is a "thanks" e-mail with my name

3.8 IANALATEIHSMBB. IHNTTTTAL. IHNMTHAL. (OK, for sake of unambiguousness:
I Am Not A Lawyer, Therefore Everything I Have Said May Be Bogus. I Have
No Time To Talk To A Lawyer. I Have No Money To Hire A Lawyer.)

What should I do?

Thank you, and I'm sorry for the long post.

Laszlo Ersek

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: "Unusual" contribution licensing problems
  2005-10-20 22:06 "Unusual" contribution licensing problems Érsek László
@ 2005-10-20 22:43 ` Brian Dessent
  0 siblings, 0 replies; 2+ messages in thread
From: Brian Dessent @ 2005-10-20 22:43 UTC (permalink / raw)
  To: cygwin-licensing

Érsek László wrote:

> after grepping the cygwin mailing list and my up-to-date cygwin
> installation for "nftw" and "fts_open", I thought that it could make sense
> (and fun) to implement nftw().

Unfortunately (or perhaps fortunately) the whole discussion is moot
because these functions were added to Cygwin several months ago by
Corinna: <>.

> Let me list some facts and guesses:

As for the rest of the question, I'm not a lawyer so I'll shut up.  But
I do believe that one way or another, whether you are assigning the
copyright to Redhat, or abandoning all copyrights, you still need
something signed by both you and your company on file with Redhat before
they can accept patches.


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2005-10-20 22:43 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2005-10-20 22:06 "Unusual" contribution licensing problems Érsek László
2005-10-20 22:43 ` Brian Dessent

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).