Geek Answers Handbook

DVCS and data loss?

After almost two years of using DVCS, it seems that one of the inherent “flaws” is random data loss: I lost code that is not clicked, and I know other people who also have it.

I see several reasons for this: duplication of data outside the site (ie, “commits must go to the remote host”) is not built-in, the repository lives in the same directory as the code and the concept of “hack”, while you have something will not be released, "common ... But that doesn't matter.

I am curious to know: have you encountered data loss related to DVCS? Or are you using DVCS without problems? And, related, besides “don't forget to push often”, is there anything that can be done to minimize risk?

+5
source share
3 answers

I lost more data from clobbering uncommitted changes in centralized VCS, and then decided that I really wanted them than from everything I did with DVCS. Part of this is that I've been using CVS for almost a decade and git for a year, so I had a lot more room to solve problems with a centralized model, but differences in workflow properties between the two models are also major factors.

, , " , , , ". ( , .) , , , - " ", DVCS, , , . ( - - , ).

, , , . :

  • , . - f'rexample, , , .
  • - , . , , . (git reflog ; , DVCS. , , .)
+2

DVCS, - ( , ), - DVCS (git ): , , , .

+3

, "" ( , , ).

IMO, , (, , : ), "" : , - . DVCS, (git/bzr/hg), . .

0
source

More articles:


All Articles