Re: Elks Distribution

[Javier Sedano <jsedano@dit.upm.es>]

Riley Williams wrote:
> 
> Hi Javier.

	It seems that I only replied to you... I am mailing to the list now.

> 
> >> Ideally, it would be an extension of the elkscmd package that was
> >> implemented as a `make instimg` command that created the disk images
> >> in a similar way to how the `make full3` creates a disk image of a
> >> 1.44M 3.5" ELKS boot floppy. If Neil wishes to do that, I have no
> >> problem with the idea, and my recommended route would be to create
> >> an elkscmd/hdinst directory in the existing tree and develop this in
> >> there, then go from that.
> 
> > I don't see it as part of elskcmd package, since it _uses_ elkscmd
> > and others.
> 
> I suggest you think about what you've said there. The elkscmd package is
> NOT a single entity, but a large collection of related entities. One can
> easily make use of the various entities that make up the elkscmd package
> without making use of the elkscmd package itself - the `make comb_net`
> command in the elkscmd package already does just that.
> 
> To me, the `make instimg` command is basically a variant of the existing
> `make comb_net` command, as per my email on Linux-8086 about 10 minutes
> ago, and I see no conflict here as a result.
> 

	It is currently. But when a packaging system was developed (and I think
Neil is working on it), it may grow appart from the current elkscmd
directory, providing precompiled binaries. It is not my decission, since
I am not the distro developer, but only my idea, of course ;-)

> > Anyway, I was only wondering about it being linked in the web page,
> > not about the CVS.
> 
> The problem with linking it to the web page is one of politics. The
> SourceForge rules state that we can only put something on the web page
> that is directly related to the ELKS project as developed on the CVS, so
> if it doesn't go in the CVS somewhere, we start flirting with SourceForge
> itself, and that's a step I would be very loathe to take.
> 

	You're right. What about creating a new module in the sf CVS?

-- 
Primera paradoja de Murphy: si la ley de Murphy puede fallar, fallara
--------
Javier Sedano Jarillo      http://www.it.uc3m.es/~jsedano
 jsedano@dit.upm.es (*)
 jsedano@ieee.org