Do they belong to you? Claim these comments.
Chris Wanstrath
Is this you? Claim Profile »
2 months ago
in 6 Line EventMachine Bugfix = 2x faster GC, +1300% requests/sec on time to bleed by Joe Damato
Great writeup - thanks for sharing the gritty details!
1 reply
skiptree
awesome writeup. Thanks!
4 months ago
in How Open Source has Gone Social with GitHub on YoungTechStars.com
Thanks for the writeup, and good luck with the continuing development of Recess! I saw it launched on HN and think it's a very interesting approach. We're glad to be helping push open source forward :)
5 months ago
in Fine. Git is Awesome. on Caffeinated Simpleton
Good post, but I'd like to point out that Bitbucket has no in-place editing / committing nor a Network Graph for seeing activity among forks. These are two of GitHub's most important features, so I think the "matches feature for feature" bit may not be entirely accurate.
For instance: http://github.com/mojombo/jekyll/network
Bitbucket also does not have commit commenting, another huge GitHub feature.
As you say though, Bitbucket does have features GitHub lacks (like the hg-powered wiki - GitHub's is not git powered) and issue tracker. The two sites take different approaches to patch management as well (patch queue vs fork queue).
Both sites have many similar features but it's unfair to call them equivalent as each has useful, unique features the other lacks.
show all 3 replies
For instance: http://github.com/mojombo/jekyll/network
Bitbucket also does not have commit commenting, another huge GitHub feature.
As you say though, Bitbucket does have features GitHub lacks (like the hg-powered wiki - GitHub's is not git powered) and issue tracker. The two sites take different approaches to patch management as well (patch queue vs fork queue).
Both sites have many similar features but it's unfair to call them equivalent as each has useful, unique features the other lacks.
- 2 points
- Jump to »
Ian Lewis
These features are absent. You are right. But at the end of the day, forks and pull requests are the truely important features of github/bitbucket.
I can live without the other stuff though they would be nice.
In place editing/committing is something I would probably never ever ever do. I like to test that a change works before commiting it. Maybe for documentation it would be ok, though lots of documentation is generated and you would probably want to make sure it looks the way you want it to before committing.
I can live without the other stuff though they would be nice.
In place editing/committing is something I would probably never ever ever do. I like to test that a change works before commiting it. Maybe for documentation it would be ok, though lots of documentation is generated and you would probably want to make sure it looks the way you want it to before committing.
justin
Perhaps I was a bit too abrupt. I understand that they both have unique features that draws some distinction between the two, but bitbucket and github are close enough (in my mind) to make the existence of a central place for people to coordinate on code a non-issue in choosing between the two.
I think it would be pretty cool if they could merge and allow you to choose between using hg or git for a project. That would help the social aspects without (hopefully) hurting any of the technical ones.
I think it would be pretty cool if they could merge and allow you to choose between using hg or git for a project. That would help the social aspects without (hopefully) hurting any of the technical ones.
Loïc d'Anterroches
And at the end of the day, you are stuck with a closed source infrastructure like github that can kick you out the way they want. Github is nice, but you lose the principle of a distributed system by just putting everything in the same bucket. How sad... I am working on InDefero http://www.indefero.net to avoid that, I can use Mercurial and Git and stay independent from those services we don't know if they are going to be here tomorrow.
5 months ago
in git ready » the staging area on git ready: daily git tips
Don't forget that in Git 1.6.1 `git stage` is an alias for `git add` and `git diff --staged` is an alias for `git diff --cached` -- I think using "stage" more consistently is a great idea.
6 months ago
in If you only could follow 10 people on Twitter… on Litany Against Fear
Obviously my list would in @qrush. Also @THE_REAL_SHAQ.
8 months ago
in You keep using that word “distributed”… on Virtuous Code
I completely agree. I'm trying to find out why people think they need GitHub when it's down for anything other than fresh public clones.
We've advertised our Disaster Recovery Guide (http://github.com/blog/175-github-disaster-guide) a few times, but I think not enough people know about it. You can still share and deploy GitHub-based repos when the site is down - that's one of the joys of distributed version control.
We've advertised our Disaster Recovery Guide (http://github.com/blog/175-github-disaster-guide) a few times, but I think not enough people know about it. You can still share and deploy GitHub-based repos when the site is down - that's one of the joys of distributed version control.
12 months ago
in Github is a blood sucking leech. on drawohara
Thanks for posting the correction, Ara. I agree we should make our open source policy more clear.
For the record, a git-svn clone of the codeforpeople SVN repo currently available from Rubyforge is 47M.
For the record, a git-svn clone of the codeforpeople SVN repo currently available from Rubyforge is 47M.
1 reply
drawohara
sigh - the email component of disqus isn't working for me - this is what i'd written:
cool - you guys rock btw - which is of course why i was upset by my interpretation of the sales page ;-)
> For the record, a git-svn clone of the codeforpeople SVN repo currently available from Rubyforge is 47M.
and *none* of my work from noaa is there ;-) thus the rationale for my bait.
cool - you guys rock btw - which is of course why i was upset by my interpretation of the sales page ;-)
> For the record, a git-svn clone of the codeforpeople SVN repo currently available from Rubyforge is 47M.
and *none* of my work from noaa is there ;-) thus the rationale for my bait.
1 year ago
in Dog-pile Effect and How to Avoid it with Ruby on Rails memcache-client Patch on Homo-Adminus Blog
If you're using the cache_fu Rails plugin, you can use reset_cache instead of expire_cache to avoid the 'dog-pile' effect.