Misunderstanding GPL's terms and conditions as restrictions (and an apology)

[Les Mikesell]

Alexandre Oliva wrote:

> I apologize to all Fedora users and contributors for my excess and for
> the harm I caused, and I thank my colleagues who approached me with a
> friendly tone and helped me see my error.

On the bright side, you've made the news: 
http://www.linux.com/feature/142772

>>>> No, RSAREF couldn't have been modified.  It had restricted
>>>> distribution and everyone had to get their own copy.
> 
>>> http://www.nic.funet.fi/index/crypt/cryptography/rpem/ripem/rsaref/
> 
>> And the point was, and is, that the GPL makes really free software
>> distribution difficult or impossible even when source is available
>> for everything.
> 
> Having source available is not enough for Software to be Free.  It
> might come as a surprise to some, but it's not even enough for it to
> be Open Source.

Perhaps you don't even realize that the word 'free' had a meaning before 
the FSF distorted it to mean restricted.  It still has this meaning for 
everyone who does not drink this cult kool-aid.

>> Note that it was Stallman himself leading the charge against this
>> free distribution,
> 
> /me stares at 'free distribution', then at the sentence containing
> 'restricted distribution' above, and pauses, wondering if it makes
> sense to even try to understand this stance, compared with the stance
> directed at the GPL.

You have to understand 'free' in the original sense of not adding 
unnecessary restrictions for it to make sense.  Perhaps you are 
brainwashed to the point that you aren't capable of understanding it.

>> Later the license on the gmp library was changed to lgpl.
> 
> AFAIK the reasoning is that, once there is a functionally-equivalent
> library under a more permissive license, the requirements of the GPL
> that are relaxed by the LGPL no longer work as an incentive for more
> software to be released in terms that both respect and defend users'
> freedoms, because anyone who'd rather not respect or defend them would
> just use the equivalent library.

It didn't, and doesn't work as such an incentive anyway.  It just 
prevents the work from being used at all in many situations and forces a 
duplication of effort to create and maintain the usable alternative. 
And even in the situations where it can be used, why not take the moral 
high ground and let the people contributing their own portions choose 
their own license terms instead of taking away that choice from them?

>  So we might as well use the LGPL
> which, should someone want to further improve the library or the
> software that uses it, ensures one or the other can be offered under
> the GPL.
> 
>> http://hackage.haskell.org/trac/ghc/wiki/ReplacingGMPNotes#ReasonsforReplacingGMPastheBignumlibrary
> 
> "Interesting" arguments there.  #1. is the result of misreading LGPL
> v2.1, missing its section 6.  I know because at some point I'd misread
> it that way myself, and asked authoritative sources about it :-)

The exceptions to the exceptions are all pretty confusing.  My reading 
says you can't distribute static-linked binaries for systems that don't 
include compilers and all the other needed libraries as part of the 
stock OS.  That excludes the vast majority of target users.

> #2. and #3. amount to "we'd rather rewrite from scratch than adapt GMP
> [under its current license] so that it does what we want", which
> probably only makes sense under the influence of mistake #1.  Or a
> fair aomunt of alcohol :-)

No, it means they want something they are permitted to distribute to 
users without restrictions whether it happens to suit RMS's whims or not.

-- 
   Les Mikesell
    lesmikesell at gmail.com