Re: [RFC] syntax highlighting

[Steven Johnson <sjohnson@sakuraindustries.com>]

Keith Seitz wrote:

> Jon Beniston wrote:
>
>>> Sorry but I do not understand where I've done something that 
>>> violates the existing copyright.
>>
>>  
>> You haven't.
>
>
> Correct -- you haven't done anything wrong in any way even remotely 
> imaginable.
>
[Snip]

>
>>
>> The only reason I can see why you need to assign copyright is so that 
>> Red
>> Hat can license Insight under a different (non-free) license. But 
>> maybe I'm
>> being a tad cynical.
>
>
> Well, I can't speak for that, because I'm pretty sure Red Hat is 
> finished with Insight (so much so that I can't even get 10 minutes 
> from an attorney to sign paperwork assigning Red Hat's copyrights to 
> the FSF).
>
> I am simply following the FSF's rules for development. If you make a 
> non-trivial change to the code (and I'm usually quite generous with 
> what I consider trivial), we need to have you (and your employer, if 
> necessary) assign copyright to us. [For a more official explanation of 
> why this is done, see 
> http://www.fsf.org/licensing/licenses/why-assign.html/view?searchterm=assignment 
>  .]
>
> In Insight's case, the compelling reason AFAIAC is specifically for 
> the dream of one day assigning all of insight over to the FSF. If the 
> FSF suspects for a moment that some chunk of code doesn't have 
> sufficient legal proof that we own the copyright, it will be rejected.

Which would support an on going position of all major changes be in
their own files, that are copyright assigned to the FSF, just like
normal GDB.  Then only trivial changes (hooks and the like) are in the
Redhat code.  Then (in the "dream of one day") that code wouldnt be a
problem for an assignment.  It would also be possible, that a specific
comment could be put in such a file stating that the hooks to this code
have also been assigned to the FSF, so that there is even less of a problem.

Also, it would mean if one day someone took the view or re-implementing
all the Redhat copyright code, they wouldnt have to re-implement the
contributed stuff.

(Digressing from this point)

Redhat might think Insight is dead, but in the last couple of days, ive
seen a large number of embedded tools vendors web sites, peddling GNU
Tools where the included debugger is Insight/GDB.  So for some people,
they just dont see it as dead, or see any viable alternatives.  Maybe
what should happen is these companies that make money off of Insight
should pool their resources, and get together with Redhat, and pay them
to release the code to the FSF, for the good of all. I suspect those
vendors would be subscribed to this list, so yes, im talking to you.

It also seems, if Redhat arent interested enough to fill out a standard
FSF Assignment document, then they probably wont be interested in
providing an assignment document for Insight to major contributors
either (seems about the same amount of work).  Which means maybe a fork
is the only way forward for those who want to continue to see Insight
prosper.  Then the fork can work on the job of removing the Redhat
code.  For a dream of one day brining the sanitized fork back into the
mainline.  But weve had similar discussions like this before.

Steven

PS.  Nothing in this is intended as a flame or other derogatory
statement of anybody or company, i dont want to see this turn into a
flame war, as that is hardly productive for anybody.  Im just proposing
alternatives, and im not particularly wed to any proposition ive made,