[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: how many users is enough? (was Re: Debian Installer 7.0 Beta4 release)



On Wed, Nov 28, 2012 at 11:00:07AM -0500, Miles Fidelman wrote:
> Jon Dowland wrote:
> >On Wed, Nov 28, 2012 at 03:47:46PM +0100, Slavko wrote:
> >>Ubuntu leaves 93 % of packages untouched and changes/additions are done
> >>only to 7 % from them (statistic by some Ubuntu & Debian developer -
> >>sorry i have no link). Then Ubuntu has significantly less to do...
> >That's a flawed argument. It takes no notice of how big the changes in
> >those 7% of packages might be. And it does not consider packages which
> >do not exist in Debian.
> 
> That's an even more flawed argument.  We're talking about packaging,
> not development.

How do you draw a distinction? Lots of software in Debian and Ubuntu
carry lots and lots of patches on top of the code supplied by upstream.

Slavko's argument was that Ubuntu rides on the coattails of Debian,
but the cited statistic did not include enough information to draw
that conclusion.

> (Personally, I'm suspicious of software
> and changes that are distribution-specific.)

It's considered a good rule of thumb to deviate as little as possible and
submit deviations to upstream for inclusion where possible. But it's not
always possible.

> What that 7% statistic really suggests is some combination of:
> 
> a. Some of those packages are developed by folks who use Ubuntu, and
> don't get around to releasing a separate package for Debian (or
> nobody has stepped up to maintain a Debian package).  I expect this
> doesn't matter very much - just as most Debian packages work just
> fine under Ubuntu, I expect most Ubuntu packages would work just
> fine under Debian (haven't tried this, though).

Slavko phrased it as 'Ubuntu leaves 93 % of packages untouched', which
does not imply that packages only existing in Ubuntu and not Debian are
considered in that 7%.

> b. Some small percentage of Debian packages need to be "tweaked" to
> accommodate minor differences between the Ubuntu and Debian
> environments.

So 7% of packages are tweaked, but that says nothing about how big
the tweaks are, which was my point.

> Now where the number of users/contributors might really come into
> play is when it comes to maintaining/developing those aspects of
> Debian and Ubuntu that are unique to the respective distros (e.g.,
> their installers and package repositories).

One of the most important contributions one can make to Debian is to
find, diagnose, test and fix bugs. They can exist in any package in
the repository, not just those that are OS-specific, or particularly
heavily customized in each OS - and how well that process works is 
directly impacted by the number of users.


Reply to: