[clc-devel] CRUXCon 2004 results
ncrfgs at tin.it
Wed Dec 22 19:40:46 UTC 2004
On Wed, Dec 22, 2004 at 02:34:22PM +0100, Johannes Winkelmann wrote:
> On Wed, Dec 22, 2004 at 13:18:04 +0100, ncrfgs wrote:
> > On Wed, Dec 22, 2004 at 10:14:05AM +0100, Johannes Winkelmann wrote:
> > > Please take the time to read it and ask, since the document is a really
> > > short summary only. Personally, I think that if we manage to follow this
> > > plan, 2005 will be a step forward for both CLC and CRUX. I hope you see
> > > it the same way!
> > - "Look into distributed revision control to enable non
> > 686 architectures"
> > What do you mean with `distributed control version
> > system'?
> Revision control systems which allow local replication
> of the repository. We've discussed those some time ago,
> typical examples are bitkeeper, svk, monotone, darcs,
> arch gnuarch/tla/arx.
> > - What about the `nvhelper.sh file' issue we discussed
> > some time ago?
> We didn't talk about this at CRUXCon.
BTW, what do you think about it after some time?
> > - Is the `Depends on' tag the definitive solution to the
> > dependencies issue?
> For the time being, yes. That said, there's probably
> going to be a change in the Pkgfile structure when
> "attributes" (see my other post regarding aliases) are
> introduced, but the format is yet to be decided.
> For the time being, the only tools supporting
> dependencies only understand the "Depends on" tag, so
> there's no good reason to change
> that now.
> > Will the `Depends on' tag still include `Nice to have'
> > packages instead of only the real dependecies?
> They're all real dependencies WRT the footprint. That's
> a selection the package makes.
> > - Will the, let's say, new `entity' born from the fusion
> > of CRUX and the CLC, share the "keep it simple" and
> > "what can't be done in a simple and nice way shouldn't
> > be done at all" philosophy some of us love so much?
I hope so and I hope that Per will be able to keep the
control over the way things are done in CRUX.
> > - Now that the CruxCON2004 is far away I'd like to ask you
> > a question...
> > Wouldn't have been nice to invite CRUX PPC developers,
> > too? Actually, giulivo has been invited but (I quote Per
> > Liden) "only as a person, not as a developer".
> > Why?
> Well, according to your webpage, that's half of your
> team, right? So you might as ask "why didn't you invite
> me?", since that's really what it is.
No. My question was clear, I thought.
Actually, giulivo has been invited but (I quote Per
Liden) "only as a person, not as a developer". Why?
> We wanted to focus on those technical issues outlined in
> the agenda. In the past, there were some rather
> unfriendly discussions here between members from CLC and
> CRUX PPC (you and me included...), and it was not
> the intention to continue this nor to resolve our
> differences at this meeting.
> I also get the impression that all CRUX-IT projects are
> very independent, and you aren't afraid to adjust
> certain things your way (think: GNU/, ilenia, categories
> in cvsup, Evolution)
We aren't afraid of adjust things? And what about you?
What about prt-get, what about Depends on, what about
README and pre-install and post-install scripts?
> and critizise our views (be it Per's or CLC's).
What's bad in critizising?
BTW, the crux-it one is another issue. We were talking
about CRUX PPC, weren't we?
And, personally speaking, even if it may looks like the
contrary am I big Per's "way" fan.
> At CLC, we have a different approach, since we tried to
> stay as close to Per's philosophy as possible, discuss
> changes with him and have as much integrated in CRUX
> mainline as possible (e.g. the 'ports' backend).
I don't think you've ever used CRUX PPC. If you did, you
would know it is made just with this spirit.
And this doesn't depend on whether you call it GNU/linux
(as it should be called) or just linux.
Oh, and since we are talking about phylosophy...
> > - "accept non 686 subprojects if they are CRUX
> > compatible"
> > Could you please define `CRUX compatible'?
> Share the philosophy and vision we have.
Funny. I would have expected a less "philosophycal" answer
from one who doesn't care a lot about "phylosophy". =)
What about the technical requirements to be defined CRUX
> and have [them?] as much integrated in CRUX
> mainline as possible (e.g. the 'ports' backend).
We asked him, too if was it possible to integrate somehow
the two CVS repos, but we were answered that, because of
logistic problems, it wasn't possible to.
Since you are going to merge, it looks like, just for you,
these "problems" have been "solved".
> addition, we tried hard to maintain a positive climate
> in communication related to CRUX (also when it's PPC
> related, e.g. see the xemacs issue some time ago).
> In the lights of the above, the selection included those
> people we expected to create a productive team for
> CRUXCon, to come up with solutions in this very short
> initial meeting.
What were we expected to create instead? We would have
loved to discuss about those topic. But maybe
narrowmindness prevent somebody from viewing it.
> If you decide to drop some of the independency and
> accept certain decisions you would take differently, I'm
> sure there's no problem to become a CRUX subproject.
Some of the independency? What do you mean? In which way
is the CRUX PPC independent from CRUX (the examples you
provided were about crux-it)?
> Independent of that, I'd assume that we can organize a
> meeting which is open to CRUX PPC and others to discuss
> integration issues between CRUX (the core project(s))
> and others, but you have to understand that for CRUXCon
> 2004, the goals were completely different.
> > - In the end... Has it been the CruxCON2004 or the
> > CLCCON2004?
> You can call it either way. I call it CRUXCon 2004.
They could seem the same but I think there is a wide
Thank in advance.
Value your freedom, or you will lose it, teaches history.
``Don't bother us with politics,'' respond those who don't
want to learn.
-- Richard M. Stallman
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 189 bytes
Desc: not available
More information about the crux-devel