The Daily Parker

Politics, Weather, Photography, and the Dog

To Git or not to Git...

I'm torn.

Or I'm a dinosaur. Or I'm a Perceiver. Or I'm a senior software development manager who's sick of changing technologies.

My current drama is between continuing to use Mercurial on one hand, and switching to Git on the other. Both are distributed version control systems, so both enable a load of flexibility in single- or multi-developer workflows. I know that sounds like jargon, so let me explain.

No, there is too much; let me sum up: If you don't have to share every little change you make to a software project, everyone is better off.

The cold war between these two products has created two problems that appear to have nothing to do with each other:

  • We have multiple software projects that we have to continue to support in production while we build entirely new hunks of them (which will take months); and
  • All of the cool tools for integration and deployment work with Git, while not all of them work with Mercurial.

I suppose I shouldn't be surprised. I chose Betamax and Laserdisc as well. I have a real weakness for the best technical solution, even while the popular solution takes the lead. (Both Betamax and LaserDisc had superior audio and slightly better video than the products that defeated them. At least I held off choosing between HD DVD and Blu-Ray until one of them was cold and dead in the ground.)

I digress.

I'm annoyed that Git is moving so far ahead of Mercurial that it's becoming an argument to use Mercurial. I assert this is an argument to popularity, not to logic. But I also get really tired of swimming upstream, and if Microsoft, Bitbucket, and a bunch of other companies are pushing a technology, who am I to blow against the wind?

A conclusion, to the extent possible, will follow shortly.

Comments are closed