Re: About the recent DD retirements
On Wed, Jan 21, 2015 at 06:46:16PM -0800, Russ Allbery wrote:
> Gunnar Wolf <gwolf@gwolf.org> writes:
> > ... And yes, maybe Debian is less attractive in general ...
> One side effect of this is that packaging feels like a solved problem.
If so (and assuming that's all Debian's about), then that'd be a good
reason to consider Debian a mature project and the only thing that's
really needed is enough people to keep the lights on.
I think that's a meritorious [0] argument.
Personally, I'd argue the other side though. Packaging works fine within
Debian -- which is certainly an improvement on how things were in the
90s (or even ten yars ago); but there are a bunch of unsolved problems
in the area. For instance:
- there are "archive networks" for most programming languages these days:
CPAN, CRAN, Hackage, PyPI, RubyGems, NPM, CCAN, etc. Installing
software from these sources is often necessary for Debian users, but
doesn't mesh well with packaged software (unlss you're a DD and can
package it yourself). Since it's all free software, I don't really
see why Debian doesn't have a set of automatic tools to repackage
all that software, so it's all just an "apt-get" away.
- perhaps it's all been fixed since I last looked, but "web apps" still
don't seem to be a "solved problem" to me. If you install, say,
libreoffice, you run apt-get (or whatever), then you run libreoffice,
and you're done. But if you want to install wordpress, you have a whole
bunch of additional steps to go through [1].
- application isolation is a popular issue now, often solved by blatting
whole OS images around (VMs, docker, ...). That brings up a whole host
of packaging issues (how do you create the image, how do you keep it up
to date, how do you distribute the image)
Other folks are answering those questions -- eg in order to run my own
instance of etherpad [2], I found that it needs some nodejs modules that
aren't packaged, so rather than jump through the hoops to Debianise it,
I created a docker instance for it, and followed the upstream instructions
to install it, and put a port redirect in place via systemd. Of course,
that system is out of date (because I forget to login to run apt-get on
it, and who knows what the npm packages are up to).
And, of course, I run Linux systems that aren't Debian these days --
I have both a phone and a tablet. In some sense you could say they're
kind-of open hardware running free software, but both running my own
software on them or rebuilding what's already on there aren't things I
can actually do. That's not because they're .apk's rather than .deb's,
but more because (as far as I've looked anyway) the android ecosystem
doesn't have the same build-package-distribute infrastructure that distros
like Debian and Fedora have built and now take for granted.
Oh, for heaven's sake, how could I almost forget to mention the
reproducible builds effort? I don't know how the laws of reality changed
to make that possible let alone feasible, but wow. And that /is/ even
being addressed in Debian [3].
> But some of it is also inevitable. A lot of smart, talented people want
> to work on really hard problems that are currently major pain points,
> since that's where they can have the most leverage. Packaging just
> *isn't* that any more, ...
So I'd say there are plenty of major pain points in this space that are
hard and interesting and attract smart, talented people. Heck, Docker
got a $.4B valuation [4] a few months ago just a few months ago.
Why aren't those people doing things in Debian? I claim two reasons:
1) because they can make (more/any) money doing it elsewhere; and 2)
because Debian folks are more interested in the solved problems of the
90s than the problems of today.
Cheers,
aj
[0] *Not* meretricious as I was about to write. Thankyou dict.
[1] https://wiki.debian.org/WordPress#Basic_Installation_guide_for_Wheezy
[2] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=576998
[3] https://people.debian.org/~lunar/blog/posts/eighty_percent/
(This should totally be posted to d-d-a)
[4] http://www.bloomberg.com/news/2014-09-16/docker-said-to-be-valued-at-400-million-in-funding-round.html
Reply to: