If you want SLOW AS FUCK VERSION CONTROL, try using bazaarWITH REDMINE.
Name:
Anonymous2010-03-25 1:29
>>7
it's not my fault some asshole made some closed-source potential malware shit part of the build process for this project.
the same asshole also chose the version control system.
Name:
Anonymous2010-03-25 21:27
1 As a benchmark, Git and Mercurial repositories were seeded with approximately 1500 files totaling 35 M of data. The servers were running in Chicago and the clients in Mountain View (51 ms ping time). The operation of cloning the remote repository (similar to a initial checkout in traditional version control systems) averaged 8.1 seconds for Mercurial and 178 seconds for Git (22 times slower). A single file in the repository was then changed 50 times and the clients pulled the updates. In this case, Mercurial took 1.5 seconds and Git required 18 seconds (12 times slower). When the Git protocol was used instead of HTTP, Git's performance was similar to Mercurial (8.7 seconds for cloning, 2.8 seconds for the pull).
Name:
Anonymous2010-03-25 23:51
>>15
you forgot to bold the last sentence. we all know that git is just as slow.
seriously, what's wrong with cvs (besides being anti-free, of course)? or subversion?
>>16 what's wrong with cvs
It's phenomenally shitty?
>>17
So you don't need some special protocol in order for it to not be fucking slow? Took 'em long enough.
Now all they need is Windows support that's at least half as good as hg, and the scripting flexibility of hg, and cleaning up the gigantic mess of CLI options.
At a point in time when every other person is an unblinking atheist, A-fear-ism offers the most cogent argument against the movement yet. Dispensing with philosophical jargon and pretentious posturing, religious educator Mat Dickie brings a refreshing dose of common sense to the debate. He exposes the ignorant assumptions made by the cynical mind, and reveals the arguments against religion to be anything but "rational". As the title of the book suggests, he considers atheism to be "the irrational fear of religion" - and leaves us in no doubt that we embrace such a negative world view to our detriment...
Who the fuck clones / commits over HTTP in a serious capacity asswipe?
Name:
Anonymous2010-03-26 19:28
>>18 some special protocol
What's the big fucking deal with the git protocol? Don't most versions control systems have their own protocols? The git protocol exists because the way git repos are stored isn't well suited to HTTP, because it stores delta-compressed, whole files into packs, and those packs can't be split by HTTP if the client only needs a few objects from it; whereas Mercurial, I believe, stores reverse deltas, probably in separate files, which is much better for HTTP. But now there's a "Smart HTTP" CGI script to work around the HTTP issues.
Now all they need is Windows support that's at least half as good as hg
msysgit works great, I use it often. Works exactly like the Linux version does, so I'd say that its support is near 100% by now.
the scripting flexibility of hg
Scripting can be done with low-level plumbing in shell scripts, or with any one of the numerous git libraries.
cleaning up the gigantic mess of CLI options.
I'm not sure what you're referring to, but the man pages are pretty clear right now. If you're just complaining that there's too many options, then I'd have to disagree; what's wrong with having features? If you don't want them, don't use them, but let others who need them have them!
>>25 What's the big fucking deal with the git protocol? Don't most versions control systems have their own protocols?
Exactly, different tools for the job. HTTP is for hypertext. Git's protocol is for Git.
>>27
Nah, I have to disagree with that kind of generalization. HTTP outmodes even specialized protocols in many ways (*cough*FTP*cough*.) I don't really care whether you CVS with it, but there's nothing really wrong with doing that.
>>25
I won't argue further on grounds that you are obviously a narrow-minded git purist who refuses to accept the simple fact that other systems might actually do some things better, simpler, and still provide more flexibility.
I don't mind when other people have different opinions, as long as they have the capability to be rational and open-minded about alternatives.
Name:
Anonymous2010-03-29 3:55
>>29
I didn't bash Mercurial, Bazaar, or any other DVCS in >>25, did I? No. I just countered some common complaints about git with my examples with why they're wrong. I, myself, have actually used the three, and my second favorite is Bazaar. I haven't used it for anything, since I'm so accustomed to git, but I liked Bazaar's graphical branch explorer. That was nifty.
Anyways, I'm not narrow-minded, just a defender of my preferred (as in, it's only my opinion, man) DVCS against what seem to me to be unfounded complaints.
Name:
Anonymous2010-03-29 14:41
I'm getting increasingly frustrated with reddit, it's a fuckshit storm full of titmorons. I need to leave it.. but one problem is there is ocasionally really good stuff on it. I don't' know what to do /prog/. help?
As this thread once again sets sail for obscurity I'll take the chance to complain that SCons is slow as fuck as well. It's not even because of the FIOC, it's just built that way.
Name:
Anonymous2010-12-17 1:28
Xarn is a bad boyfriend
Name:
Anonymous2012-06-09 6:58
I don't think GIT is so slow when it comes to cloning. I started cloning a mercurial repo a couple of hours ago and it's still not finished.
I've used both Mercurial and Git and I can say that I like git better. At the time we moved to git, rebasing in hg... let's just say I gave up trying to do it.
Some of the repos on my work are CVS. I once was close to requesting
that unbreakable diamond screen monitor so that I can punch it when I
have CVS rage (and I have it every time I am trying to do something
non-trivial) without damaging company equipment, while also venting
the stress to protect my brain.