Comment by jm4 | Hacker News Reader

Comment by jm4 | Hacker News Reader

Parent Full thread

How can the Ruby community splinter between two competing web frameworks? The entire Ruby-using population does not need to agree on a web framework. There can be smaller communities around various projects and frameworks within the Ruby community. In fact, I would think that it would be vital to Ruby's advancement and evolution.

You mention that Rails being the dominant web framework is a strength and in the same sentence remark about how with other languages there are multiple competing frameworks. How, exactly, does this negatively impact developers? On the flip side, how does a merger between Rails and Merb positively impact developers? Developers who don't like Rails had an alternative in Merb and vice versa. How is the elimination of a choice a good thing?

The way I see it, competition here is only a good thing. Why would anyone want Rails to be the only option-- or even the de facto standard-- when it's been shown to be exactly the wrong tool for the job in several high profile scenarios? Don't get me wrong-- I'm not bashing Rails. I'm just saying it's a good thing to have a few more tools in the toolbox. When your only tool is a hammer everything starts looking like nails.

Well, Rack is really changing a lot already with Ruby frameworks. Sinatra is becoming more popular for highly specialized apps, and now you can embed Sinatra and Camping apps into other Rack apps, which includes Rails edge apps.

Basically what's happening is that everything in Ruby is already becoming more modular, shifting toward organized APIs, so this helps move in a direction where you can basically build a customized framework like legos out of a bunch of gems and specialized apps without spending an inordinate amount of time trying to patch things to work together.

This is informative. I had never heard of Rack before. It seems very similar to Python Paste. Neat.

However, I don't see what this has to do with my original comment. It doesn't address any of the concerns I raised about the john_dahl's comment. Specifically, I'm wondering how several competing web frameworks results in splintering of the Ruby community. Isn't competition good? Don't we want a variety of tools that meet different needs?

I frequently see Merb mentioned as an alternative to Rails in response to some gripe about Rails. Developers obviously have needs that aren't met by Rails. So my question is how is this merger a positive thing for developers? What's wrong with having several tools that are used for different scenarios? Of course, 100 competing web frameworks that are essentially all the same doesn't do anyone any good, but surely there's nothing wrong with several if they specialize in different areas.

john_dahl's comment essentially applauds the merger as a unification of the Ruby community and implies that it's a disadvantage to have competition. Is that really a good thing? Developers don't benefit one iota from having a "winner".

I agree that competition is good. There is (was) concern that loud, shallow, infantile arguments of Merb versus Rails were damaging, to an extent. If nothing else, they were really, really annoying, and too frequent. This may be the "splintering" of which john_dahl was speaking.

There are other frameworks, although they get little "press", as they lack an Engine Yard or 37signals hype-machine. In fact, if you visit ramaze.net (one of the more mature frameworks) and scroll down, you can find a listing of all the frameworks of which the ramaze people are aware. Twenty-one, including Merb AND Rails, at last count. (It is likely that not all of them are wholly distinguishable from one another. I can't say, as I haven't tried them.)

the problem of too much choice: http://www.spring.org.uk/2007/09/barry-schwartz-on-why-too-m...

That's a completely valid problem, but at best it's debatable whether it applies to Ruby web frameworks. There's not exactly a plethora of mature web frameworks available. Rails and Merb seemed to be the most firmly established frameworks with a few others like Sinatra, Camping and Ramaze sprouting up.

I think the too much choice problem applies much more when you're installing a Linux distribution and you have to choose from 40 different text editors or IM clients. Who has the time or the inclination to evaluate each one?

Reply on news.ycombinator.com