[sf-lug] Red Hat Linux source code brouhaha?

[Rick Moen]

Quoting aaronco36 (aaronco36 at sdf.org):

> And directed mostly to Rick, how might/will IBM-owned RH's source
> code "Business Model"-change affect the content of the pertinent
> sections of your/his excellent and comprehensive RedHat
> knowledgebase [07], e.g., the 'RHEL Forks' webpage[08], the 'RHEL
> ISOs' webpage[09], and the 'RHEL ISOs' webpage's 'Bottom-line
> summary for the impatient' at [10] ??

This is a very complicated topic, but basically I concur with Bradley M.
Kuhn's analysis.  (Disclaimer:  I have not yet had a chance to read Mr.
Mike McGrath's quasi-response.)

In reading Brad's comments, please be aware that when he speaks of what
"GPL" requires, he's knowingly skipping over a complication that
matters.  The nature and details of what GPLv3 requires in the way of
obligations by distributors of covered code is a bit different and more 
strenuous than are those of codebases under GPLv2.  This is, mostly, the
result of lessons learned and includes language closing off some
opportunities for gamesmanship.

Point is, RHEL includes many packages under GPLv2, and some packages
under GPLv3.  (It also includes packages covered by numerous other
copyleft and non-copyleft licences.)

To determine whether specific conduct results in a redistributor of
covered code complying or not complying with conditions depends on the
specific language of the particular licence.  (And, by the way, it's a
really good idea to read both GPL texts to better understand them.)  
If you do, note the language about an obligation to furnish matching
source code in "preferred form" to _any_ requestor who has in any way
come by a copy of the work.


In theory, a non-compliant redistributor of Joe Example's codebase can
have its licence rights revoked by Joe, and, if the distributor keeps on
distributing Joe's code (or derivatives of it) anyway, then Joe can haul
that party into civil court on copyright violation tort charges
(assuming Joe has registered his copyright with Library of Congress
Copyright Office).

Of course, suing IBM for anything is not a hobby taken on lightly.

As to the effect on Rocky Linux and similar, and the scope of
restriction of access to RHEL matching source code, perhaps Greg Kurtzer 
will speak to that?  He'd be expert and up-to-date.  My expertise and
knowledge is behind the times, and will have to catch up.

It should go without saying that what is on my old rhel-forks.html and 
rhel-isos.html pages reflect the _prior_ situation.

There's a lot more to be said, but I'm sorry to say I don't have time at
the moment.

Also, suggestion:  See what lwn.net has to say.