User: Password:
Subscribe / Log in / New account

Preferred form for whom?

Preferred form for whom?

Posted Mar 11, 2011 22:35 UTC (Fri) by PaXTeam (guest, #24616)
In reply to: Preferred form for whom? by rahvin
Parent article: Red Hat and the GPL

> If the licensor hasn't made his definition of preferred form (if it's different than the GPL definition) [...]

ok, i think i see where your thinking went astray. you keep mentioning this 'GPL definition' of 'preferred form'. thing is, it does *not* exist. really, look at the GPLv2 (the license of linux) and see for yourself. what the GPL does define there is the 'source code', in terms of 'preferred form for modification'. it does *not* further elaborate on just what this 'preferred form' is. this is the entire reason for these discussions here because different people have quite different (or even contradictory) ideas about 'preferred form'.

> The reference to preferred form, in my opinion[...]

see, that's your own opinion only, noone seems to share it. and for good reason, that sentence in section 3.b does not define 'preferred form for modification'. really, if the license wanted to further elaborate on that term, it should and would have done so explicitly, like it already does for other terms where the license authors saw it necessary to be precise (whether they achieved that or not is another question but let's not digress). as to what your highlighted sentence was about see this thread somewhere else where others explained it. it's most definitely not about the 'preferred form for modification' term.

> Although the language could be a little clearer I don't think there is
> any doubt that the preferred form is machine readable source code on a
> medium typically used for software interchange. How that source is
> arranged as long as it's not obfuscated is beyond the license agreement.

the GPL does *not* contain the word 'obfuscation'. where did you pull that requirement from? and what does it mean? what is considered 'obfuscation'?

> It would be absolutely silly to assume that whatever the licensee
> interprets to be their preferred form is the form the licensor has to
> comply with.

i've seen people argue the exact opposite view in defense of RH here, but let see, if i take your view then it's bad news for RH as their *own* 'preferred form' is clearly sending patches around among their own developers, not entire tarballs as in the RHEL6 srpm. you know, not unlike what they currently distribute in the RHEL5 kernel srpm. how do you explain that two very different forms can both be preferred at the same time?

> But the license only requires that you be provided the source code in a
> manner that is editable and machine readable.

somewhere above you said it must not be 'obfuscated'. now you say it does not matter as long as the 'source code' is 'editable' (another word not present in the GPL, i don't know where you're getting these from ;). what does that mean? does it have to be in ASCII? 'cos your editor may not do EBCDIC, or whatever i invent for my own 'preferred form'. what if i translate everything into some intermediary representation and do my modification on that, what am i supposed to distribute then? certainly my own tools are mine and not affected by the GPL, so what will you do with that form then? really, instead of inventing new terms not present in the GPL, you should stick to its language and acknowledge where it is ambiguous.

> As far as providing quotes of RMS's speeches I have better things to do with my time.

it's just that you seemed very confident that he had precisely explained 'preferred form'. but it's not important at all, what really matters is the copyright holders' opinion, not his.

> But I also believe it's a necessary move if RedHat has any intention of
> responding to Oracle's attempts to destroy them.

paying customers of RH have access to the patches in question (one wonders why they exist if they're not the preferred form for modification, but let's not digress) and so does Oracle (how do you think Sun servers are certified to run RHEL?).

> Just imagine for a moment that anyone that uses the software can
> redefine "preferred form" however they would like and what that would do
> to the community.

i don't need to imagine anything, i just need to look at the past decades and see what people preferred and did not complain about. i'm sure you know it as well and apparently it didn't scare anyone.

(Log in to post comments)

Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds