From: Steven D'Aprano on
On Thu, 13 May 2010 06:24:04 -0700, Ed Keith wrote:

> --- On Thu, 5/13/10, Lawrence D'Oliveiro
> <ldo(a)geek-central.gen.new_zealand> wrote:
>>
>> What have you got against LGPL for this purpose? --
>>
> Most of my clients would not know how to relink a program if their life
> depended on it. And I do not want to put then in DLL hell. So I avoid
> the LGPL.


Are you implying that by distributing your libraries under the MIT or
Apache licence, no linking is required? That's a cool trick, can you
explain how it works please?



--
Steven
From: Carl Banks on
On May 13, 4:30 pm, Brendan Abel <007bren...(a)gmail.com> wrote:
> While I think most of the disagreement in this long thread results
> from different beliefs in what "freedom" means, I wanted to add, that
> most of the responses that argue that the MIT license permits the user
> more freedom than the GPL, suffer from the broken window fallacy.
> This fallacy results from the short-sided-ness of the user base, as it
> is only considering the first generation of derivative works.
>
> I agree, that under an MIT license, the first generation of derivative
> works have more freedom.  But any extra freedom gained here comes at
> the direct expense of all future generations of derivative software.
>
> Under a GPL license, it is true that the first generation will have
> less freedom to distribute their software as they would like.  But it
> also ensures that all subsequent generations of derivative works have
> the freedom to access all previous derivative works.

I believe the you have the fallacy backwards.

The thing you GPL fanbois refuse to understand or accept is that, in
the real world, a person or company who doesn't want to open source
their "derivative work" will only rarely be forced to by the GPL.
They'll work around it instead, vast majority of the time. They
could:

1. Derive their work from a project with a license that grants the
user more freedom
2. Reimplment the functionality seperately (*cough* PySide)
3. Ignore the license

And no, a small number of anecdotal counterexamples is not any strong
evidence to the contrary.

On the other hand, those who intended to release their work as open
source are going to do it even if the license is permissive. The way
some of you GPL fanbois talk you'd think the MIT license prohibitied
open source derivatives.

So, you see, the rights given to users of first generation works (as
you say) are far more important than requiring people who create
"derivatives" to also offer those rights. Those who intend to do that
will anyways, those who don't intend to will find a way not to.


Carl Banks

PS The word "derivative" is quoted throught because the GPL's
definition of "derivative" is ludicrous and not in accordance any
common defintion of the word.
From: Aahz on
In article <2d625c61-7a94-4c71-8953-69c3b3c76d07(a)k29g2000yqh.googlegroups.com>,
Paul Boddie <paul(a)boddie.org.uk> wrote:
>
>All my position has ever been is this:
>
>A copyrighted work denies recipients virtually all rights to do stuff
>with that work, such as modify and redistribute it. Copyleft licences
>grant some privileges and uphold some obligations in order to ensure
>that these privileges are universally maintained in all forms and
>extensions of the work. Permissive licences grant more privileges to
>immediate recipients but do not uphold as many obligations.
>
>You rejected the suggestion that people using permissive licences
>afford users fewer privileges than those using copyleft licences, yet
>on balance when considering all forms and extensions of the work, they
>do. And the only such privilege that the copyleft licences withholds
>from recipients is that of withholding any other privilege from others
>who receive the work.

IMO this only makes sense if one agrees that people should not be allowed
to sell software for money. Absent that agreement, your argument about
freedom seems rather limited.
--
Aahz (aahz(a)pythoncraft.com) <*> http://www.pythoncraft.com/

f u cn rd ths, u cn gt a gd jb n nx prgrmmng.
From: Patrick Maupin on
On May 13, 6:30 pm, Brendan Abel <007bren...(a)gmail.com> wrote:
> While I think most of the disagreement in this long thread results
> from different beliefs in what "freedom" means, I wanted to add, that
> most of the responses that argue that the MIT license permits the user
> more freedom than the GPL, suffer from the broken window fallacy.
> This fallacy results from the short-sided-ness of the user base, as it
> is only considering the first generation of derivative works.

The broken window fallacy is about labor that could have been spent
elsewhere if someone else had done something differently. The only
time that comes into play in my programming life is when I have to
recode something that is nominally available under the GPL, so I'm not
sure this is really making the point you think it is.

Regards,
Pat
From: Patrick Maupin on
On May 13, 6:39 pm, Steven D'Aprano
<ste...(a)REMOVE.THIS.cybersource.com.au> wrote:
> On Thu, 13 May 2010 08:06:52 -0700, Patrick Maupin wrote:
> > If I download an Ubuntu
> > ISO, burn it and give it away (let's say I give away 100 copies, just to
> > remove the fair use defense), then I have violated the GPL. I provided
> > chapter and verse on this; go look it up.
>
> I'm sorry, I can't see where you have provided "chapter and verse", or
> even a URL.

In the original message, I wrote "In the case of GPL v3, for example,
Ubuntu lets me download code under 6d, so if I download it and burn
it, I would have to use 6a or 6b; if I had actually received a CD from
Ubuntu, I might be able to use 6c, but not if I downloaded it." I
thought it was clear those were references to the license clauses.

> >> If you compiled the CD yourself, and failed to provide a written offer
> >> on the CD, then yes absolutely you would be in violation of the licence
> >> terms, and shame on you.
>
> > Not relevant.
>
> You didn't specify whether the "Linux CD" you were distributing was a
> mere copy of an existing CD , or one you created yourself, so you will
> pardon me for covering both possibilities.

Well, in the section I just quoted, I did mention "Ubuntu"...

> >> The GPL doesn't require you to force source code on those who don't
> >> want it, but it does require you to make it available if they ask, and
> >> for you to notify them appropriately of this fact. You don't even have
> >> to explicitly tell your friend he can have the source code. You just
> >> have to make sure that the written offer is available on the disk you
> >> give him.
>
> > There is no written offer on the disk, because I burned it from Ubuntu's
> > repository. It really is that simple -- if I give away copies I've made
> > of Ubuntu, I've violated the GPL.
>
> No, I think this use-case would count as "propagation without conveying",
> since you are merely acting as a mechanical proxy between your friend(s)
> and Ubuntu.

No, I'm actually creating a copy and distributing it (in GPL v3
terminology, conveying it), especially since my friends don't
specifically ask for Ubuntu, and I'm foisting it off on them
(especially if I burn 10 CDs at a time so I have one if I need it).
GPL v2 also has similar rules, and there's lots of v2 licensed
software on the Ubuntu CD.

> I will admit that the GPL FAQs are not as clear about this matter as they
> should be.

I think it's quite clear, although a bit of a tedious slog. But since
you want a URL, try this: http://www.gnu.org/licenses/gpl-faq.html#UnchangedJustBinary

> > Unless you can cite some authority
> > that tells me I'm wrong and gives real reasons. I actually quoted
> > chapter and verse from the license, but you chose to ignore that and
> > make unsubstantiated claims.
>
> I'm sorry, I can't find where you have quoted "chapter and verse" from
> the licence, so I can't comment.

Well, I just re-copied what I posted, and added a URL from the FAQ.

Regards,
Pat