Ruby on Rails Isn’t All Roses
July 12th, 2007Ruby on Rails is definitely the cool new kid at school right now. Actually, RoR is like the suave, handsome quarterback type who the girls swoon over and the guys want to be until everyone finds out he lives in a rotting shed in the sticks with a goat on a leash in the yard and dad passed out in a broken rocking chair on a makeshift porch. What, you didn’t have that guy in your high school? :) [Hey, I'm not judging him - the high school students are. I'm sure he's very nice.]
I was enamored with Ruby on Rails until I dug into it a bit. No doubt it’s great for certain types of sites but it doesn’t have the general purpose appeal or proven widespread scalability of ASP.NET or PHP. I think the bigger contribution Rails is making is inspiring new scaffolding and code generation projects for ASP.NET and PHP.
Anyway, this post is primarily a bookmark for myself so I remember where to find this article on the scalability problems they’ve had at Twitter running RoR:
July 12th, 2007 at 7:55 pm
http://twitter.com/blog/2007/06/under-hood-at-twitter.html
July 13th, 2007 at 6:32 am
“until I dug into it a bit.” It sure was only a bit.
“the bigger contribution Rails is making is inspiring new scaffolding and code generation projects” you’ve obviously never used rails.
July 13th, 2007 at 5:15 pm
What Chris said. It -can- be that, but there’s certainly no reason why it has to -just- be that. I’d recommend digging into Rails more than you have (which hasn’t been much, apparently).
The Twitter example is perfect, in fact. It’s not Rails that was the problem. It was their three-man team having built something using Rails in a way that didn’t scale. They could have just as easily built something in [your technology here] that didn’t scale properly. Twitter worked it out, apparently, and seems as committed as ever to Rails, in fact.
July 14th, 2007 at 1:07 am
(No need to be snippy, Chris)
Like I said, I dug into it a bit by prototyping some things and did a bunch of research along with that. I’m in no way a Rails expert and I do think Rails has merit. Rails is cool, no doubt about that. But you hear more caveats about Rails than PHP or ASP.NET.
The influence of Rails on the industry is far more important than Rails itself. Rails is the new Smalltalk. It seems perhaps like it will be more commercially successful than Smalltalk has been. Nevertheless, the increased interest in scaffolding, ORM, code gen, etc. for ASP.NET and PHP are due in large part to the influence of Rails. Also the great work around dynamic language support for .NET is, it seems, inspired by the newfound popularity of Ruby and Python.
My experience has been that the people using Rails to date are generally pretty far along the bell curves of motivation, enthusiasm, and aptitude levels (Tim O’Reilly calls them the ‘alpha geeks’). I have gobs of respect for Evan Williams (Twitter) and believe he would put together nothing less than a stellar team. It doesn’t give me comfort to know that a stellar team said “[r]unning on Rails has forced us to deal with scaling issues – issues that any growing site eventually contends with – far sooner than I think we would on another framework.”
I dare say that right now the chances are higher that a stellar team will build a reliable and easy-to-scale system on ASP.NET or PHP than with Rails. I don’t think that will always be the case necessarily. Rails will continue to mature and the collective knowledge base will grow. And remember that I’m talking about chances, not guarantees. Developers of all stripes make bad decisions some of the time. There are fantastic Rails sites and crappy ASP.NET sites.
August 11th, 2007 at 1:34 am
I found another post which is worth looking at that discusses scaling the logger facility, much like twitter does it:
http://www.dotrb.com/2007/8/11/scaling-rails-with-sysloglogger