[om-council] Forwarding email from Ben

Bernhard Rosenkraenzer bero at lindev.ch
Fri May 12 18:02:49 EDT 2017


(Taking Ben off for now because we may want to agree on things in the council before sending 100 individual replies, but will chat with him on irc if I can find him)

I must largely agree -- we (the council) haven't been doing our job lately (whatever happened to the regular irc meetings?), and we (the developers) have worked around the absence of anything else by just doing what needs to be done.
I vote to reestablish the weekly or biweekly council meetings. There's thousands of things to be done and we aren't doing them in a timely matter. The discussion with Gael pretty much died down after Krakow, we haven't been talking about the mail server, we haven't been syncing up on laptop progress, ...
I'll take some of the blame, for obvious reasons I haven't been around as much as I'd like to for most of April - and I know some others are having other things happening too -- but we can't let the project die down completely.

I think some things (esp. where I disagree with Ben) need to be pointed out, replying inline to those:

> 2. The disconnect between developers and rest of Community is so
> profound as to be damaging to the Association and the distribution.

I disagree about this one. The only big disconnect there is between what is realistic (what we're doing) and what is pipe dreams (what many in the community think we should be doing, but what we can't do unless we have tens of extra people).

> FWIW-2: What ever transpired in Hungary has never been disseminated to
> the Community still to this day.

True - I presume it's mostly because we haven't been following up on it ourselves, so there's not that much new to report.

> FWIW-3: I need to get a yes or no decision on the main-testing repo in
> ISO's issue.

My vote:
YES for pre-releases (so testers don't report bugs that have been fixed days ago), NO for final release (so we don't throw test packages at "stupid" users).
But those installing a pre-release are doing so because they want packages that we haven't said are ready for release, so obviously they should have testing repositories enabled.

To some extent the problem has been the decision to remove tested 3.01 and replace it with a 3.02 snapshot (instead of just releasing 3.02 where testing repos would not be enabled).

> FWIW-4: The issue of default file system on the ISO's is similar to to
> the issue about main-testing repo in the ISO's.

Agreed, and I'm in favor of going back to ext4 as default simply because it's more stable. f2fs has some issues with grub integration that make it look bad even when it's working.

> FWIW-6: For those of you who have decided to switch from urpmi to dnf
> perhaps it would have been wiser to include more people in the
> discussion and decision making process

No. Not on this one. Code-wise, urpmi is a steaming pile of crap that should have been killed years ago. Furthermore, we don't have anyone who speaks the language it was written in too well.
Including more people in the decision process there could only result in the usual "WE WANT YOU TO KEEP THIS AROUND FOREVER!!! We don't care if there's something better out there! You CAN'T drop it!" resulting in a core component being broken and unfixable because we can't replace it (people won't let us) but we also can't fix it (don't have people with the right skills).

I'm not the biggest fan of dnf, but it's by far the lesser evil.

And FWIW this has been decided by an even larger group than us (Mageia developers -- their decision to drop urpmi means we can't "steal" their work on it anymore, essentially requiring us to either get more manpower to keep urpmi barely working or to do the sane thing and switch to something that is actually maintained).

> or at least let the Community know that this is going on.

Apparently people don't read the meeting logs -- maybe we can volunteer Ben to summarize them for the rest of the community?

> From the outside looking in it appears
> that a very small group of people are deciding what is best and the
> rest of us can like it or leave.

For some things, it has to be done that way. People at large don't understand how little manpower we have (and maybe that's a good thing, not sure how many people would be happy relying on something maintained by that small a group), therefore have totally unrealistic demands, and on some issues (e.g. urpmi, drakx) don't understand what the actual issue is - and (as we have seen on the i586 issue) never step up to actually do anything about a something they want to keep.

> Again this could have been handled better.

Yes, by communicating it better. But the decision needs to stand.

> 1. We're never going to be able to engage the Community and grow the
> distro if we don't even try. And we aren't trying.

Can we volunteer Ben to try?

> 2. The way we are currently governing the Association and more
> specifically the distribution is exclusionary not inclusive. If we
> wish to grow the distro we need to change this 180 degrees.

I think the perception with Ben (and maybe others) is that we have a functional council that constantly talks and decides things, and doesn't inform people.
The question here is, what is worse? The fact that the council simply isn't active right now (which is what is really going on) or the perception that we're working in secret?

> 4. Would it be possible that all of us, me included, find a way to learn how to communicate with and initialize communications with
> users, testers, people on forums and mailing lists, publications, ect.
> in such a way as to engage and attract participants and grow the
> distro?

In a way we do have a problem there, it's a bit of a negative feedback loop.
We throw stuff at the community, we don't get a response (or we get a response that blocks our work, like "YOU CAN'T DROP XYZ!!! But we won't step up and help maintain it! It's YOUR duty to maintain it! That's what we aren't paying you for!"), so next time we just do what has to be done (potentially making what remains of the community even less likely to step up and do something).
We need to break out of it.

ttyl
bero




More information about the OM-Council mailing list