Re: Patch pinging

[NightStrike <nightstrike@gmail.com>]

On Tue, Jun 29, 2010 at 6:14 AM, Manuel López-Ibáñez
<lopezibanez@gmail.com> wrote:
> Still, I don't understand why a shell account is required to start
> working on this. From what I understand, the scripts that need
> conversion are not secret, so anyone can work on them. The bugzilla
> customizations can be accessed with anonymous cvs, so whatever custom
> changes we have made, they are accessible anonymously and they can be
> ported to a new local bugzilla installation. In any case, patches
> against a pristine bugzilla installation would need review. I am not
> sure if copyright assignment is needed for all this. If so, then it
> depends whether NightStrike would agree to sign papers for the FSF
> with his real name. But...

When I first offered to do it, several people told me that FSF
paperwork is not required for this effort.

I presented what I would need - access to the current code, as well as
the database.  That didn't have to be the real stuff, but what I
ultimately did is what jsm28 and iant told me to do.  A snapshot of
the database, as long as it is complete, would suffice.  It's not just
a matter of the bugzilla code, it's the db itself, too.

> NightStrike, even just starting a wiki page and making a list of
> everything that needs adjusting and suggestions on how to implement
> the changes would help incredibly whoever ends up committing the
> changes. Right now, I wouldn't know where to start!

Creating an outline of how to go about things that someone else will
ignore / throw away sounds like a bigger waste of time than trying to
jump through the patch pinging horror.

> This whole issue has focussed in a little problem about the final step
> (installing bugzilla in sourceware.org), whereas there is so much work
> to do before reaching that step, that probably the person that starts
> this work won't be the same that finally installs the new bugzilla
> version. Take the opportunity that people are paying attention right
> now and collect all the information that would be needed for this.
> Then you will know if you need a shell account, a cvs account, a
> copyright or just someone else to do that part on your behalf (I will
> be willing to review and commit stuff).

I already did all of that.  I was ready to start actually creating
incremental patches.

> Yes, I know that contributing to GCC seems sometimes like a fight and
> finding the right loop-hole to achieve what you want. And sometimes,
> you lose. On the other hand, the lack of a strong leadership means
> that many times despite all the resistance and obstacles, you can find
> a way to get done stuff if you manage to find the correct way to do it
> (either by means of technical or social engineering, with the latter
> being sometimes more important than the former)

That's not how you run a welcoming project that's trying to attract
new people and *grow*.  That's a recipe for stagnation.  When it takes
just as much if not more effort to fight your way through beligerent
curmudgeons than it does to do the actual work, then something is
wrong in the state of Denmark.

> Personally, I don't give a fig what your real name is. I don't support
> giving you (or anyone without a paper trail) shell access to
> sourceware.org, but I don't think this is needed to contribute to GCC.
> There is such a long list of useful stuff to be done that I could keep
> a horde of anonymous contributors busy for years.