Re: [Librefm-discuss] Contribute to the project

[Ian McEwen]

I haven't been much of a contributor in a long time myself, but my two
cents:

On Sat, Aug 22, 2015 at 08:38:11AM +0000, Diego Agulló Falcó wrote:
> Hello,
>
> As stated, I want to contribute, but I'm finding it surprisingly difficult.
> The project seems to be in a state of semi-abandonment, its source using
> technologies and dependencies from over 5 years ago, many of them now
> deprecated or a thing of the past (ADOdb, the old mysql extension or
> Smarty, to name a few).
>

I believe this is an accurate characterization of the project's state.

>  I don't know which one is the "official" site or how to report a bug
> correctly. I don't even know which one is the canon source for the source
> code:
>
> Sites that may be official (or not):
>  - https://gnu.io/fm/
>  - https://www.gnu.org/software/gnufm/
>
> Source code:
>  - git://gitorious.org/foocorp/gnu-fm.git (as stated on gnu.org, it's
> hosted on gitorious which is long dead)
>  - https://git.gnu.io/gnu/gnu-fm.git (gnu.io)
>  - https://gitlab.com/foocorp/gnu-fm.git
>  - https://github.com/foocorp/gnu-fm.git
>
> Four different sources, all seem legit. Gitlab has its own bug reporter,
> the bug report tool listed on the sites has its certificate expired. An the
> worst thing is that wherever you get the code from, it hasn't received a
> commit for a long, long time.
>

I think that there isn't accurately a "real" home any more; gitorious is
where the real development of the project, when it happened, took place.
Probably any source is as good as any other for the most part.

> Many of the resources under foocorp.net are down, something an account is
> blocked: http://bugs.foocorp.net/projects/fm/wiki
>
> But anyway, I downloaded one of them, installed it and (after disabling
> error displaying) ran it successfully. So, what now?
>
> How can I help? What is the state of the project and is there an updated
> roadmap? Is there a bugtracker with actual issues to solve? Any intention
> of using modern "de-facto" standards (composer, PSR, Twig, maybe even a
> framework like Symfony, Silex, Laravel, you name it).
>

I'd say that anything you feel like you want to do, you should. I don't
know if the people who control the libre.fm installation have a process
for integration of new code (perhaps they'll weigh in), but if there's
any sort of plan it's not seeing much action anyway; replacing it with
your own that you'll do work on will be an improvement.

My own thoughts, not that I'm likely to contribute much myself, would be
that standards/modernization work would be a good first priority as long
as it's interesting enough work to continue (perhaps interleave some
work on things that are more fun). Switching to a full framework I'd
only do if it's a faster route to modernizing the code than starting
from where it is. That being said, at some level you're just creating a
new project inspired by GNU FM/libre.fm, which wouldn't necessarily be
bad either if you have big ideas for where you'd like to go.

> Cheers!

I hope this helps, or at least inspires someone who knows more than me
to jump in :)

signature.asc
Description: PGP signature