Re: Plans for git

From: Alexander Leidinger <Alexander_at_leidinger.net>
Date: Mon, 21 Sep 2020 09:21:59 +0200
Quoting Pete Wright <pete_at_nomadlogic.org> (from Sun, 20 Sep 2020  
08:41:13 -0700):

Not responding to Pete directly, but in general to this topic with  
some parts of what Pete considers good as something to hook into.

>>> making quarterly reports about this for almost a years as well. We put out
>>> calls for people to help with the efforts about the same time. We have
>>> tried at every step of the way to be open and honest that this was going to
>>> happen.
>> All developer centric communications....

I fail to see why it is important to non-developers, which (D)VCS the  
developers of the product they use are using. It may be interesting,  
yes. It may have an impact on some (see the announcement of  
deprecation of portsnap), but those which put their craftmanship into  
it, are the important ones. Not to tell that we don't need to inform  
(or to let them repeat all the arguments we provided internally  
already), but the main use case for the VCS is to have those with  
commit privilegues handle version control.

I do not go to a lot of in-person meetings, I just follow the internal  
and the official mailinglists, and there was communication for a  
loooong time (and no, I do not use git for FreeBSD stuff so far, so  
you can not consider myself as someone who is eager to get FreeBSD  
moved to git and as such has an interest in it --- but I do understand  
the reasoning and can agree to it). Any FreeBSD committer who tells he  
was not aware of it, has simply not paid attention to it. For any  
non-committer see below.

> I would argue that quarterly reports are actually one of the few
[...]
> honestly there has to be *some* responsibility of operators to at  
> least make an effort to keep up to date on the status of the various  
> efforts in such a large project.  and as an outsider the idea that  
> comms can only happen on the mailing list isn't the greatest - how  
> am i to know that the idea of one person on the ML carries more  
> weight than another, or one persons opinion is the "official" stated  
> opinion of the core group?

I agree to that. And I agree that the status reports are a nice way of  
getting some kind of inside-information in a central way. And in my  
opinion we gave early enough information about the migration to git.  
Maybe it can be organized, so that some guides for users (again,  
deprecation of portsnap and such) are published first via the status  
reports (and other channels), before the switch to the git-repo  
happens. We have no other official channel which is suitable for such  
way-ahead announcements IMO (yes, we should send a mail to  
freebsd-announce when we switch + an entry in the news section of the  
website, and /maybe/ we should send a mail some weeks before the  
switch too, but so far, I do not think this info should have been send  
to freebsd-announce, or be published on the website).

In my opinion the people which drive this didn't keep it behind closed  
curtains, and they went step by step more public, as they made  
progress. To me it looks like now, that they have something which is  
presentable to the world (and not only to committers), they presented  
it to the world. I do not think we can hold them responsible that we  
do not have "the one official channel" for this (hey... anyone feel  
free to create it for the next big change, and document what shall be  
announced how via this channel).

Bye,
Alexander.
-- 
http://www.Leidinger.net Alexander_at_Leidinger.net: PGP 0x8F31830F9F2772BF
http://www.FreeBSD.org    netchild_at_FreeBSD.org  : PGP 0x8F31830F9F2772BF

Received on Mon Sep 21 2020 - 05:22:17 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:41:25 UTC