attic-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Hervé BOUTEMY <herve.bout...@free.fr>
Subject Re: attic-test.git created
Date Tue, 01 May 2018 03:21:42 GMT
Le lundi 30 avril 2018, 22:43:12 CEST Jan Iversen a écrit :
> Sent from my iPad
> 
> > On 30 Apr 2018, at 22:28, Hervé BOUTEMY <herve.boutemy@free.fr> wrote:
> > 
> > Le lundi 30 avril 2018, 21:03:24 CEST Jan Iversen a écrit :
> >> Sent from my iPad
> >> 
> >>> On 30 Apr 2018, at 20:43, Hervé BOUTEMY <herve.boutemy@free.fr> wrote:
> >>> 
> >>> I don't get what's the issue with creating a test Git repo, that is
> >>> independent from anything else, and that was discussed on the ML
> >> 
> >> There are no issue as such, it is just that lately we have seen a lot of
> >> things happening very uncoordinated. In this case it happened crossing
> >> another (slightly older, but with positive comments) to move the
> >> production
> >> to git.
> >> 
> >> I would have preferred a more organized way, e.g. making sure a test site
> >> (or build bot) will have someone to maintain it, before just doing it.
> >> However this is just my opinion, which, even though I did the
> >> maintenance,
> >> seems to count very little.
> >> 
> >> Please lets stop this discussion, done is done, and I (as
> >> chair/pmc/committer) will have to accept the consequences.
> > 
> > there are no consequences to creating a test git repo
> 
> I think you mean there are no technical consequences, which are totally
> right, but there are other types, like e.g. the effect on the community or
> personally for people doing the maintenance. This action was for me the
> newest drop in a long chain of rain, but please I am NOT thinking or
> suggesting you did anything wrong, I am just tired.
looks like a wrong interpretation on intents, IMHO
Given the misunderstandings on svnpubsub&gitpubsub&CI&... that is clearly 
visible in our discussions on the way things can work in the future, doing 
some tests to better get common understanding is useful (as seen with my own 
https://github.com/asf-attic/asf-attic.github.io/ parallel test).
I don't see the benefit of limiting tests: relax, decisions remain to do before 
working on real production Attic site *with common confidence*


perhaps one misunderstanding to fix is to clearly establish that changing the 
rendering engine (Anakia vs Perl vs Jekyll) is a topic that is completely 
independant of a switch from svn to Git (with 2 options: separate branches for 
output or one mixed branch)

and Buildbot is somewhat related to the switch from svn to Git, since the 
separate branches option is IMHO really viable when there is a CI, which would 
be a good thing if we want to be able to do simple fixes just by editing 
sources on GitHub, without anything installed or run on the editor's computer

I hope the last 2 points help creating a common understanding and do not add 
confusion

Regards,

Hervé

> 
> rgds
> jan I
> 
> >> rgds
> >> jan I.
> >> 
> >>> I did the exact same at GitHub just because it was outside Apache then
> >>> was
> >>> even more clear that it was a side test to have  chance to get common
> >>> tests
> >>> instead of long discussions where it's clear that a part of the
> >>> discussion
> >>> is that we don't have the same initial culture on the technical parts
> >>> 
> >>> regards,
> >>> 
> >>> Hervé
> >>> 
> >>> Le lundi 30 avril 2018, 19:39:16 CEST Jan Iversen a écrit :
> >>>> Hi
> >>>> 
> >>>> We just received an email that attic-test.git is created, on private,
> >>>> which
> >>>> I cannot relay here due to apache policy (which I agree with).
> >>>> 
> >>>> I can however relay my response, what apart from a “thank you” was
a
> >>>> call
> >>>> to all:
> >>>> 
> >>>> “I must however state that the word “consensus” seems to mean
nothing
> >>>> in
> >>>> our little community. I remember we have discussed transferring the
> >>>> site
> >>>> to git, something I would have done tomorrow based on the positive
> >>>> comments... but attic-test, where did that come from.”
> >>>> 
> >>>> I am concerned about people going in different directions, without
> >>>> trying
> >>>> to get consensus, what the community wants, that way we will never
> >>>> reach
> >>>> a state where we can get a new site online.
> >>>> 
> >>>> PMC members know who made the request in self-service but not who
> >>>> initiated
> >>>> it.
> >>>> 
> >>>> A concerned committer/PMC/Chair
> >>>> jan I
> >>>> 
> >>>> 
> >>>> Sent from my iPad
> >>>> 
> >>>> 
> >>>> Sent from my iPad



Mime
View raw message