Snaps: a failed experiment

[Paul Smith]

On Tue, 2025-09-23 at 02:32 +0800, Bret Busby wrote:
> > Unfortunately the advantages are almost all gained by the people
> > maintaining the packages, while the downsides are almost all felt
> > by the people using the packages.  That's probably why maintainers
> > generally like snaps, and users are more likely to not like snaps
> > :).
> 
> Unfortunately, it is the direction that ICT has taken, contrary to
> the principles that I was taught, when I was taught software
> development.
> 
> I was taught that the purpose of software development, was to develop
> software for the benefit of the client/user.

Well, that's certainly true when you're getting paid, or working for
the benefit of friends or family.  But you're not paying any of these
people anything for any of this work they're doing.  Many of them are
not paid at all for this, they're doing it for free.  Usually that's
because they enjoy it.

I know lots of people who enjoy using software and hacking on software.
I know only a few people who enjoy packaging software for old releases
and solving the porting issues that come up.

So, users DO gain from snaps, just in less immediate ways.  For
example, packagers don't burn out and stop working in things because
they're tired of dealing with all the problems.  Users of older
distributions get access to newer software, that wouldn't otherwise be
available to them.  Maintainers (in theory) get fewer issues caused by
environmental differences, so they have more time to work on real bugs
or new features for users.  Etc.


Anyway, I'm not trying to convince anyone of anything.  Šarūnas asked
the reason snaps exist, so I provided (what I think is) the main one.