Re: building foreign packages with pkgsrc

[Ottavio Caruso <ottavio2006-usenet2012%yahoo.com@localhost>]

On Sat, 22 Jun 2019 at 19:13, Jason Bacon <outpaddling%yahoo.com@localhost> wrote:

>
> Given the current pkgsrc user base, it's not clear whether the will
> exists to support less savvy users.?? My hope for pkgsrc in scientific
> computing is to push it forward far enough to ignite interest among the
> most tech-savvy scientists and reach a critical mass of maintainers that
> will get it rolling on its own.
>
> I think this will require two things:
>
> 1. Quick and easy deployment (binary bootstrap kits like the ones I
> provide are probably the best approach).
> 2. Filling in gaps among core scientific packages (e.g. BLAS family
> which Thomas Orgis has nearly finished, MPI, Flang, etc)
>
> I've made a lot of progress on 1) with auto-pkgsrc-setup.?? 2) will take
> a while given the small number of contributors as of now.
>
> There are a lot of reasons to be optimistic, though.?? Pkgsrc continues
> to grow rapidly in general and we have a solid set of development
> packages (compilers, interpreters, editors, etc.).
>
> Whether or not it eventually snowballs as a science tool, it seems worth
> my time to keep nudging it forward.

The fact that there are now at least two (semi)official binary repos
for Linux/Centos (yours and jperkin's) is welcome and commendable. The
way I see it, the obstacles for pkgsrc to be adopted on a large(r)
scale in the Linux world are:

1) The average Centos/RHEL end user started using that distro because
somebody told them it's "corporate Linux", therefore "good", as
opposed to the "bad Linux", the tinkerers, the geeks, and so on (gross
generalisation, I know!). Quite frankly, they have a package
management system that is fast, works pretty well and has tons of
software. They're happy with systemd, freedesktop.org's
semi-proprietary solutions and what have you: because it's corporate,
certified, IBM approved and has all the necessary marketing buzzwords
that make it cool and sexy. I doubt there could be any further
traction at that end of the spectrum.

2) The average (old and new) Linux user is proud of having some binary
package manager that, most times, has been around for 20+ years and
"just works", and is not remotely interested in the intricacies of how
things interconnect at source level. Unless things screw up massively,
in which case...

3) they just install some other distro and try again.

I have noticed that there is a small but determined group of Linux users who:

- are unhappy with systemd/network-manager/gnome-isms, not only for
geeky technical/ethical reasons (Devuan) but for practical reasons as
well;

- are unhappy with this new trend of "64 bit only" releases (Suse,
Ubuntu). Some people need to run i386 binaries for various reasons
(Wine is one of them. I'm following an ongoing thread on the Winehq
forum)

- are generally unhappy with how bloated modern Linux has become.

I think this is a potential user base that could be interested in
pkgsrc as a platform (with or without binary releases).

-- 
Ottavio Caruso