At the end of the day, since I am a balanced individual with more love for people than computer languages, the prevalent emotional attachments I witnessed over Ruby was an obvious red flag for me. I did not get the impression that Ruby could be people-centric. I think the Ruby elitists could do well to pick-up some sales skills.

Point conceded. I do truly hope that Ruby becomes a useful tool for the large-scale application. And it’s only fair to agree that it’s not there *yet*.

@The Universe and Everything

To clarify what I mean by “scaling” in the comments and article, I mean in terms of code base maintainability and modification. I’ll admit that I don’t have any experience at all deploying Rails in a high-volume environment. (and didn’t I mention that I never could get into Rails?) Working with Ruby on anything non-trivial (i.e. larger than your average script) has been nightmarish.

Ruby is quite clean and concise for a lot of things, but then you come across that one random operation you need that’s just incredibly ugly when represented in code (file and directory meta queries come to mind). More than that, the code is often hard to walk visually and the flow is difficult to see. Even on simple scripts to control video encoding (~300 LOC), this can become a problem. Finally there’s the problem with lack of reasonable tools. I won’t deny that NetBeans Ruby has made huge strides. In fact, NB Ruby is just about the best Ruby IDE I think is even possible, but it’s still a Ruby IDE. It’s limited by the raw power of the language, specifically the flexibility of its type system. Ruby tools can unfortunately never be on par with Java, C++ or Scala tools. This is a productivity factor which *must* be taken into account when considering Ruby for a large application.

This is why I think Ruby is a great scripting language, but collapses under its own weight when you “scale” up further.

Ruby is great, Java is great too, as is C, or C++, or Fortran, or Scala, ad infinitum. For the love of God – they’re just tools! If they don’t work for you then don’t use them…

My $0.03 (I put in the extra peeny becuase I thought it was a worthwhile cause).

All that aside, I think that Ruby’s shortcomings in the enterprise sector come mostly from the way it requires a somewhat different approach to, say, Java or .Net deployments. Deploying Ruby applications requires an extensive knowledge, and use of, numerous open source technologies, but more importantly, the setting up of numerous subsystems and daemons, which is not the case with PHP, for example.

Ruby’s ecosystem is not particularly well honed to enterprise development either, but I think this is something that is changing rapidly, and as the commercial deployment products continue to roll out (FiveRuns are making a great start on this front), people will begin to get more faith and Ruby may still, one day, have a chance at fighting on a level playing field with the incumbents. So, yes, I’ll agree Ruby isn’t there, but I’d say it isn’t there yet.

Your only other objection is based on “scaling” in rails but you don’t even understand basic rails code, as evidenced by your example.

I worked for a Fortune 20 company in J2EE for 5 years and have seen my share of scaling issues. We used “mature” and “enterprise” tools that costs $16K per license (our company produces a ton of these tools, so we got them for free as well as direct access to the dev teams working on them). But we still have many many horrific problems. Some we overcame and some we didn’t; we just had to hire a ton of sys admin staff for some projects.

I’ve been developing and deploying Rails apps now for about a year and I don’t see how Ruby/Rails is worse than my J2EE experience.