Interesting, albeit a bit “light.” I’d say it overlooked the biggest, simplest answer for “what took so long”: inertia.
There’s nothing inherently wrong with inertia when it comes to projects. It’s what keeps things working most of the time (“If it isn’t broken, don’t fix it.) Just because the bigger, better, faster way to do something is out there doesn’t mean a project can or should stop and retool. (Every time a bigger, better, faster car hits the market, I don’t drop everything and buy it.) It’s not practical, it takes time and effort and resources to migrate histories and build systems, iron out bumps, not to mention retraining developers, and developers reconfiguring themselves.
I believe the data and charts are skewed by by the “inertia” of old projects, projects that are still on Subversion (or Git) because it’s still working, or they are simply abandoned/inactive projects. The really damning data would be what VCS new projects are choosing. I believe that data would show Subversion deep in the minority, with Git and Mercurial together well beyond 50%. (Based on my gut, and what I hear and see.)
It seems odd that they don’t attribute any of Git’s rise in popularity to GitHub. While there are other code repository stores, it’s probably safe to say that GitHub is more popular, and has a huge focus on helping people find interesting projects.
All the article says is that GitHub got funding because git was popular, but not that it helped make git popular in any way. GitHub was around for a while before it got funding…
Like @brycied00d said, pretty light. I don’t think the author did much independent research. Or any…
Interesting, albeit a bit “light.” I’d say it overlooked the biggest, simplest answer for “what took so long”: inertia.
There’s nothing inherently wrong with inertia when it comes to projects. It’s what keeps things working most of the time (“If it isn’t broken, don’t fix it.) Just because the bigger, better, faster way to do something is out there doesn’t mean a project can or should stop and retool. (Every time a bigger, better, faster car hits the market, I don’t drop everything and buy it.) It’s not practical, it takes time and effort and resources to migrate histories and build systems, iron out bumps, not to mention retraining developers, and developers reconfiguring themselves.
I believe the data and charts are skewed by by the “inertia” of old projects, projects that are still on Subversion (or Git) because it’s still working, or they are simply abandoned/inactive projects. The really damning data would be what VCS new projects are choosing. I believe that data would show Subversion deep in the minority, with Git and Mercurial together well beyond 50%. (Based on my gut, and what I hear and see.)
It seems odd that they don’t attribute any of Git’s rise in popularity to GitHub. While there are other code repository stores, it’s probably safe to say that GitHub is more popular, and has a huge focus on helping people find interesting projects.
All the article says is that GitHub got funding because git was popular, but not that it helped make git popular in any way. GitHub was around for a while before it got funding…
Like @brycied00d said, pretty light. I don’t think the author did much independent research. Or any…