On Thu, 12 Dec 2013, Otis Gospodnetic wrote:
Hi,
On Thu, Dec 12, 2013 at 2:00 PM, David Lang <[email protected]> wrote:
On Thu, 12 Dec 2013, Otis Gospodnetic wrote:
Hi,
On Thu, Dec 12, 2013 at 12:21 PM, David Lang <[email protected]> wrote:
Rsyslog has been on git for a long time, creating additional repos on
github is the only new thing.
Sounds like there was/is confusion about git vs. github.
what is it about github that makes it especially collaborative compared
to
'just' git? everything I see is just exposing the capabilities fo git
and a
repo on github is not special in any way from any technical point of
view,
it's just visible on the github website with the fork button rather than
someone having to type git clone manually.
This must have been written up N times before, but I won't google now.
It's not only about Git and not only about forking. It's about being able
to create a PR easily, about seeing a PR (pull request) and the diff
easily, about cross-linking it with issues, about being able to comment on
everything, about being able to review the code before it gets committed,
about being able to click on a file and edit it in place and commit, about
.........
creating a pull request is easy with plain git, and a pull request created
with github doesn't need to be processed on github, it can be processed on
git.
cross-linking on issues would require that we abandon the existing
(searchable and archived) mailing lists, forums, etc in favor of ones that
are dependant on github.
But it's hard to argue with the revolution Github enabled.
Want more patches? Lower the required effort. Github does that.
fine, but that just requires better information on github, not abandoning other
things.
Personally, I'm very much in favor of projects running their own
infrastructure. Things happen to companies and things that they provide for
free now may not be available for free later. Tieing your projects success
to some company like that that has no direct interest in the success of
your individual project is not a great plan.
I know this thinking and am guilty of applying it sometimes. In this "what
if" case I think there is more to gain then to potentially lose.
debateable :-)
The Linux kernel is very collabrative, but it doesn't use github (and in
fact pulls from github are frowned upon, at least at the upper levels due
to trust concerns)
I don't follow Linux kernel dev, but aren't they on bitbucket? I don't
know...
Funny "linux kernel repo" query shows https://github.com/torvalds/linuxas
#2 hit.
the linux kernel project does not use bitbucket as far as I know. They run
their own servers.
the github repo you found is a secondary repo, just like the rsyslog repo
under Rainer's account. In both cases the authoritative public repo lives
elsehwere (rsyslog.com and kernel.org) and the 'real' authoritative repo
lives on a personal laptop.
So if you are happy with the kernel repo, then the problem with the
rsyslog repo is simply a matter of publicity.
It's a big one.
so let's address this problem directly rather than looking at changing other
things in the hope that it will improve this problem.
David Lang
_______________________________________________
rsyslog mailing list
http://lists.adiscon.net/mailman/listinfo/rsyslog
http://www.rsyslog.com/professional-services/
What's up with rsyslog? Follow https://twitter.com/rgerhards
NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of
sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE
THAT.
