Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

On the one hand I'm inclined to agree about monoculture, on the other I've worked at a company where one employee dictated that we use bzr. Basically everything was an order of magnitude more difficult through minimal fault of bzr. None of the third party tooling worked with bzr, or if something did have bzr support it was dropped ages ago.

It's kind of the same situation with Github. The pain in moving away from it as they continue to slaughter the UX is still pretty great.

So yeah monoculture is bad but progress is painful.



Having been that person (albeit before Git was the one true way), I'm sorry. But seemed so nice and had solid externals and sparse checkouts back in the day. And as bad as it got for us, it still felt easier to pick up and use than early Git.


Sure, and I was that person with the svn -> git migration at a different gig. Ideally our behavior matures over time.

For me, at least, two things eventually made git really manageable: github and the O'Reilly book. After those git was a no brainer. Even with git becoming the de facto standard, if you don't need the integrations it's not so unreasonable to use something else (altho bzr's been discontinued).


To me that just sounds like an example of the reason to avoid tight coupling.

If something breaks beacause it cares that I use git and only git, then it's that something that is broken and to be avoided, not the git alternative.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: