[CLUE-Tech] GPL loopholes

[Dave Hahn]

[snip]
> What about when they have "Red Hat" embedded into some of their GPL
> software?  Doesn't this nullify or make it more difficult to defend the
> trademark?

There is a legal aspect here as well.  Red Hat has their name/logo
trademarked.  U.S. trademark law stipulates that if you *do not* make
efforts to defend and protect your trademark (i.e. defending against
dilution, theft, inappropriate use, use by other parties with proper
attribution, etc into a large legal mire), you *can not* have the privileges
of a trademark.  So, they have to go down this road if they want legal
protection from someone else calling their distro "RedHat".  May not seem
nice, but, that's the way the law reads. IANAL, but, my best recollection.

> I don't think you should assume that all I want is a free ride.  The
> spirit of GPL and Linux has always been sharing.

No, the spirit of the GPL was to insure that users of code had the
availbility to modify the code to meet their needs.  Hence, you can build
code, GPL it, and *only* sell it to people that you want to sell it to -
free speech - not free beer.  The Linux community has had a history of
sharing freely, but not the requirement, necessarily, to do so free of
charge.  So, the community may be acting in the spirit of the "law", the GPL
(and the other licenses used) are the letter of the "law".

  The fact that Red Hat
> does share their source code could be even stronger grounds for
> participating in sharing.  I mean after all, how much of Libranet uses
> Red Hat's code?  Maybe none.  But either way, distros use code which
> they didn't write.  From just a moral standpoint, if I used code that I
> didn't write, I'd certainly at least give ANOTHER site my source code to
> allow downloads.  You say "it costs them money to provide downloads" but
> there are plenty of sites out there that are willing to provide space
> for your code for free!

Sure, within reason and within their restrictions.  The argument still
stands that bandwidth, server space, etc, costs *someone* money.  There is
not way of getting around that.  Money has to move in some fashion to pay
for these things.  Some are supported by advertising, some by product sales,
some by corporate supporters, etc.  Sourceforge, I'm assuming, is one of the
sites to which you are refering.  Remember, in order to make that space and
bandwidth avaialble, they *sell* their software.

So even though legally it may be more sound to
> say "at a reasonable price" the reality is people can often distribute
> GPL code at no cost to themselves.  (e.g. Libranet can provide
> linuxiso.org with images and not have to pay for the bandwidth it takes
> for you to download them)

Still a head in the sand argument on the over all sense, linuxiso.org still
pays money for their bandwidth.  Their way of getting that money is
different.  However, a single model doesn't work for everyone.

> I guess I just don't like to see people out there being "stingy" with
> their code when in order to write their software they had to use someone
> elses.

"There is nothing new under the Sun" - What about authors of articles,
books, etc.  Should an author that downloads software, reads the manuals and
becomes familiar with the product, who then publishes a book, give it away
for free because everything upon which they based their work came freely
from someone else?  If so, I think you'd have a bigger gripe with O'Reilly
then the Libranet guys.

> But most of all, my point is I don't want this software for
> myself, I  just like to see people share code which they have borrowed
> from others in the first place.  So please, don't say I'm just looking
> for a free ride.

"Infomation yearns to be free" - true, but, freedom costs money to someone.
I'm sure the Libranet guys will share the code they used - however, it is
completely unfair to cause that be a financial strain in them.  Think about
it, they downloaded that code in the first place, using bandwidth that they
paid for - saved it to a hard drive that they paid for - burned/mastered it
to a CD they paid for - and you want to force them to go through all this
expanse, however small, and be forced to give it to you free?  Flatly, the
economics don't work here at all.  If that was a requirement of the GPL, no
one would bother.  The statement that the code must be provided "at a
reasonable cost" is an acceptance by the architects of the GPL that this
movement of data costs money.

Many people in our community do not like to face the fact that the things we
do, the software we enjoy, the movement forward of ideals we support, costs
time and money.  Kernel developers who works on they kernel because they
love doing so - no arguement.  However, these people still need to eat, pay
rent, buy gas, etc.  In order for kernel advancement to be a reality, money
has to move.  Linus went to work for Transmeta. Transmets gave him a lot of
freedom to work on the kernel while he was there, however, don't assume for
a minute that they didn't make money while having Linus' skill and name
associated with the company.  Last I recall, Alan Cox is paid by Red Hat.
Red Hat has to have money coming in for money to be able to go out.  And,
mind you, they have quite a few products that are pay for play only.  Even
independant projects like Samba claim big money supporters. Paul "Rusty"
Russell, the main IPTables developer, states he counldn't have contributed
what he did without the support of comapnies that benenfit financially from
Linux. Again, money has to move in order for us to enjoy what we enjoy from
Linux.  These are but a few examples that show this fact.  So, if you want
to continue to enjoy the benenfits of the software and the community do not
begrudge those that are doing what must be done in order to keep the wheels
moving.

It may not be pretty, it may not be utopian, but it's the truth.

>
> Angelo
>
> _______________________________________________
> CLUE-Tech mailing list
> Post messages to: CLUE-Tech at clue.denver.co.us
> Unsubscribe or manage your options:
http://clue.denver.co.us/mailman/listinfo/clue-tech
>
>