[Scribus] Fwd: Scribus Digest, Vol 25, Issue 34

[Craig Ringer]

On Mon, 2005-03-14 at 10:33 +0100, Martin Costabel wrote:
> Craig Ringer wrote:
> []
> > Fink and MacOS/X native are somewhat orthogonal. 
> 
> I don't agree. What Fink is orthogonal to is the culture of the "single 
> application computer", where you have a build system that works if you 
> *only* build scribus, or if you *only* build a fortran compiler, or if 
> you *only* build a TeX system, or if you *only* build your favorite DNA 
> visualization program. This corresponds to the shareware culture where 
> the guru dedicates his computer and his time to one piece of software 
> and everyone else uses the binaries he has created. Repeated reinvention 
> of wheels is needed for this.

I'm not sure I understand this. What I'm saying is that one create a
MacOS/X native Scribus build without requiring fink. I think it's
important for a Qt/Mac build to work well with fink, but for creating
standalone builds my feeling when I started on the MacOS/X stuff was
that fink adds another variable in the mix. I'm open to being convinced
it'll be quicker and easier using fink, but fink is not something I'm
really familiar with.

I don't have a problem with needing a somewhat tweaked environment to
build and release self-contained Scribus Qt/Mac builds if that's what it
takes. My understanding is that doing things that way potentially
simplifies things like making a static build, etc.

If using fink will make that process easier, and I won't have to worry
about any sort of conflicts, unexpected changes, etc then I'm willing to
look into doing it that way too. 

I still think Scribus needs to build OK standalone, however, and that we
need to know how to get a good build from upstream sources. I don't much
care when that gets done, though - for now the priority is making
something that works.

> > It's also possible to make a standalone .app bundle that can be
> > distributed as a .dmg . That's what I'm working on currently and what
> > Andreas Vox helped kick off. It's actually largely working now, though
> 
> As I have pointed out repeatedly, this can be done using Fink, too.

My apologies, then. I hadn't noticed you mention that - probably
indicating that I'm blind and can't read (it'd hardly be a surprise).

> > not spectacularly polished. I just need to fix up the some lingering
> > build issues and (most importantly) make plugins work properly and I
> > might be able to put up an early preview .dmg.
> 
> The "some lingering build issues" is what this always has been about.

> > I'm on IRC a lot, so it'd be really good if we (Andreas, Martin and I)
> > could get together at some point and discuss some of this, as I'm
> > concerned that we're duplicating a lot of work here.
> 
> I'll not be duplicating anything, I'll wait now until this stabilizes. 
> My contribution is on the table.

Perhaps I should restate that as "I was unaware that much of your work
was also useful for a MacOS/X Aqua native build, and would like to know
more about what you needed to do."

I'm pretty new to this game Martin, and not exactly an expert working on
MacOS/X either. I've only been looking at this for a few days, prior to
which I had essentially no interest in the OS/X side of things. I
haven't been involved in fink or the MacOS/X community much, and really
don't know much about your fink work on Scribus. I made the apparently
erroneous assumption that your fink changes would be very mac specific
and not suitable to be merged into mainline (and I also figured you
would've posted changes for inclusion in the main tree if they were
suitable for it). I'm now interested in working to get the required
changes integrated into the main codebase. I'll look over your work and
I'm sure that'll help a lot.

If you decide to more directly show me the error of my ways, I won't
complain, either.

-- 
Craig Ringer