Shipping hg-git by default?

[Arne Babenhauserheide]

Am Freitag, 25. September 2015, 15:01:15 schrieb Sean Farley:
> 
> Bryan Murdock <bmurdock at gmail.com> writes:
> 
> > On Fri, Sep 25, 2015 at 12:22 PM, Nathan Goldbaum <nathan12343 at gmail.com> wrote:
> >> ...what the usage statistics of git vs hg on bitbucket look like...
> >
> > What do those look like?  My assumption is that bitbucket would still
> > have more hg usage than git?
> 
> You would be very wrong, unfortunately. Git usage is about 87%:

That’s slightly higher than the relation of the general number of
users on either platform, right?

> https://www.youtube.com/watch?v=afibo7fHd20#t=22m46s
> 
> My hope is to rally the troops, so to speak, and build up more of a
> community over the next year.

In my group and with the collaborators in the netherlands we’re
happily using Mercurial, and it just works. I get a question about an
advanced topic once every few months, otherwise people just keep
working.

For me Bitbucket is an integral part of Mercurial usage, and I think
that the missing support for largefiles and obsolete/evolve is a big
roadblock for Mercurial itself.

I am saying this despite having written extensions for hosting
Mercurial on FTP or on Freenet.

My feeling was that when Bitbucket was aquired by Atlassian, that was
a heavy blow to Mercurial, because Bitbucket, the biggest hoster of
Mercurial repositories, started spreading propaganda against Mercurial.

The worst example is that the post by Charles O'Farrel “Why Git?”¹ is
still on the blog without any note, even though people showed in the
comments that it spreads misinformation.

¹: https://blogs.atlassian.com/2012/03/git-vs-mercurial-why-git/

That this factually wrong post is there without even a header which
asks readers to look into the comments for corrections, that’s so bad
that I wrote an article about it,² when someone passed around the link
2 years later as argument that Git is better than Mercurial.

²: http://draketo.de/light/english/mercurial/factual-errors-why-git-atlassian

If you can, please print this article and pass it around to the
marketing folks at Atlassian.

Keeping the article from Charles O'Farrel with plain misinformation
on the page of the biggest Mercurial hoster without even a
clarification is a NO-GO. An absolute NO-GO.

I consider it a blow into the face of every Mercurial contributor —
who are exactly the people who enabled Bitbucket to build a good
hosting solution in the first place. When I read that I thought that
someone in marketing had totally fucked up. When I read it again two
years later, I could no longer shake off the feeling that it’s
intentional: That somewhere in Atlassian someone decided that
Mercurial should die. It felt like a hostile takeover.

Sorry for the rant. That is how it felt.

The first glimmer of hope I saw for Mercurial on Bitbucket was when
Sean was hired to improve the Mercurial support. If you manage to
introduce support for largefiles and evolve in Bitbucket, that will be
a killer feature. Painless binary files and safe collaborative history
rewriting.

Bitbucket is important for Mercurial, because it provides painless
hosting for ordinary people. And I think most Mercurial contributors
would be happy to provide support, if they can be sure that Atlassian
will start treating Mercurial as a real asset and not as a burdensome
legacy to get rid of.

You’re taking on Github, and there is one asset you have which Github
does not and cannot have.

Best wishes,
Arne
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 299 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.mercurial-scm.org/pipermail/mercurial/attachments/20150926/bfb9f7c6/attachment.asc>