[Haskell-cafe] Haskell-friendly Linux Distribution

[Erik de Castro Lopo]

Mathijs Kwik wrote:

> As a developer in 3 languages (ruby & java professionally, haskell as
> hobby) I must say I really prefer just managing this manually,
> separate from the package manager.
> 
> I'm running ubuntu LTS (8.04) on production servers.
>
> But this would mean that an environment for a language would also be
> somewhat frozen for at least 2 years, which isn't very useful. When
> 10.04 gets out with ghc 6.12.1, it will still mean that's the only
> thing available until 2012, or I need to upgrade the entire OS every 6
> months.

Not necessarily.

I am faced with a similar problem, having over 700 production
client machines (administered remotely) in the field running
the 8.04 LTS release.

Because we use Debian packaging as the only sane way to manage
binary distribution to that number of machines, we manage our
own repository which is basically a validated version of 8.04,
plus validated backports, plus our own packages.

Furthermore, all of our own packages are built on an autobuilder
to ensure that what is in revision control will actually build
from source. The autobuilders all start off with a nearly bare
install in a chroot, then install the build dependencies and
finally build the package.

In order for this to work for our one haskell package, I backported
ghc-6.10.4 and a bunch of haskell libraries from Debian Testing so
that this one package can be built in the autobuilder.

I have also been hearing rumours that the next LTS release 10.04
will be more of a rolling release, where more recent versions of
things will be available by enabling backports.

<snip>

> This might not be a solution for you, it really depends on your needs,
> but for me, I found it's often useful to control the exact environment
> an application needs and it gives developers the freedom to run
> whatever OS they like, which is a huge benefit if you use contractors
> or if devs want to work from home.

I think I found a solution with the same goals as your's, but with
a different implementation. Since your machine count is smaller than
mine, your scheme probably works better for your situation. For my
larger machine count, I would not be happy to trade my scheme for
yours :-).

Cheers,
Erik
-- 
----------------------------------------------------------------------
Erik de Castro Lopo
http://www.mega-nerd.com/