Git is Inconsistent
r6.ca
Git is Inconsistent
1–10 of 82 posts
Re: Git is Inconsistent
#2I agree with you completely, but want to know how this can be fixed in git? Surely there has to be something about the merging algorithm that can be changed to fix this, and if that's the case we can just patch it and move on.
What is the specific problem with the algorithm that causes this?
Re: Git is Inconsistent
#3Hmmm, nobody seems to care:
http://article.gmane.org/gmane.comp.version-control.git/1057...
Re: Git is Inconsistent
#4I agree with you completely, but want to know how this can be fixed in git? Surely there has to be something about the merging algorithm that can be changed to fix this, and if that's the case we can just patch it and move on. What is the specific problem with the algorithm that causes this?
I assume this will reduce the quality of the merge algorithm from a stand-alone point of view, which is presumably a very hard sell.
Re: Git is Inconsistent
#5I wonder how mercurial compares in this aspect. Also, I'll keep using git because for sure, it's a helluva lot better than SVN or CVS (which my company was using when I got there).
Re: Git is Inconsistent
#6To quote Johannes Schindelin [1] :
This all just proves again that there can be no perfect merge strategy;
you'll always have to verify that the right thing was done.
Re: Git is Inconsistent
#7I agree with you completely, but want to know how this can be fixed in git? Surely there has to be something about the merging algorithm that can be changed to fix this, and if that's the case we can just patch it and move on. What is the specific problem with the algorithm that causes this?
I assume this will reduce the quality of the merge algorithm from a stand-alone point of view, which is presumably a very hard sell.
I don't know this is true for sure, perhaps introducing the patch would increase it's quality. If someone offered such a patch we could discuss, instead the article only shows the broken test case. It's almost a darcs plug without and reasoning.
Re: Git is Inconsistent
#8I wonder how mercurial compares in this aspect. Also, I'll keep using git because for sure, it's a helluva lot better than SVN or CVS (which my company was using when I got there).
Same as git, and you'll probably get the same reactions.
"""
In other words, we're already at the point of significantly diminished, possibly negative returns on effort. The last few percent will always require some level of human-equivalent intelligence. I think effort here is much better spent elsewhere, like researching general AI or playing on waterslides.
""" http://thread.gmane.org/gmane.comp.version-control.mercurial...
Re: Git is Inconsistent
#9I wonder how mercurial compares in this aspect. Also, I'll keep using git because for sure, it's a helluva lot better than SVN or CVS (which my company was using when I got there).
Same as git, and you'll probably get the same reactions. """ In other words, we're already at the point of significantly diminished, possibly negative returns on effort. The last few percent will always require some level of human-equivalent intelligence. I think effort here is much better spent elsewhere, like researching general AI or playing on waterslides. """ http://thread.gmane.org/gmane.comp.version-control.me…
The least we can do is detect a situation like this and make it a conflict.
Re: Git is Inconsistent
#10I wonder how mercurial compares in this aspect. Also, I'll keep using git because for sure, it's a helluva lot better than SVN or CVS (which my company was using when I got there).
Same as git, and you'll probably get the same reactions. """ In other words, we're already at the point of significantly diminished, possibly negative returns on effort. The last few percent will always require some level of human-equivalent intelligence. I think effort here is much better spent elsewhere, like researching general AI or playing on waterslides. """ http://thread.gmane.org/gmane.comp.version-control.me…
Thanks so much for this link, this is exactly the kind of analysis I was hoping for. Clearly this is all a bit FUD, and darcs which gets this right, is trying too hard. I wonder how fast the general merge algo that darcs is using to get this right is?