Ruby's vocal minority

Esther Schindler over at CIO.com does a great job picking apart some recent data on Ruby adoption. The (Koders.com) data, which is gleaned from language-specific searches on its code repository site, suggests that Ruby interest is up by a factor of 20 since 2004.

However, as Schindler points out, the Koders.com data may simply reveal the obvious (i.e., the Ruby community is vocal) or the not-so-pleasant (i.e., perhaps Ruby users have lots of need to look for information because of problems with Ruby).

Ruby use really isn't all that much. According to Evans Data, which asks developers twice a year about their favored programming languages, only eleven percent of North American developers use Ruby today, for any part of their work...About two thirds use JavaScript in any guise, just for comparison, but somehow that doesn't generate the same kind of passion...

[I]t also might mean that Ruby developers need more help than others do..., whether because the existing software is hard to understand or because their shops don't have a lot of existing in-house expertise. It might mean that there's so much easily-found JavaScript open-source code that they don't need to head to a dedicated search engine for it. Maybe, in their enthusiasm for all things Ruby, they just like to look at code examples.

So maybe Ruby is on a roll, or maybe it's not. The Koders.com data is inconclusive on this point. All we really know is that the Ruby community is vibrant and vocal.

But Ruby isn't about to take over the enterprise, where most of the money in software still resides. Java and .Net still rule the enterprise roost. Along with the imperfect Koders.com data, O'Reilly Media's book data points to an upsurge in Ruby development, but we're a long ways off from massive enterprise adoption of Ruby, at least in any way that threatens Java developers.

[sigzero]

Ruby grew because of the hype that was going along with Rails. Other than that...meh.

[PACSferret]

Maybe the stats just reflect the amount of time Ruby developers spend non-developing?

[odubtaig]

I'm slightly confused, given that JS is client-side and Ruby is 100% server-side, shouldn't it be compared to other server-side languages like PHP or ASP? Given that JS and Ruby are different tools for different jobs I should think that trying to compare them directly is like trying to compare the use of delivery vans against forklift trucks; sure they might complement each other but you wouldn't try to use one in place of the other.

[ACubyy]

If Matt actually read the release from Black Duck Software he might have actually understood odubtaig's point. "The number of Ruby searches has increased by more than 20 times since 2004 and has surpassed alternatives, such as PHP, Python and Perl." The story here is comparing Ruby to PHP and the like. Matt may be surprised to discover that there is a need for different programming languages for different purposes. Were you every actually a developer?

[sam_smoot]

I tend to agree that Ruby growth is relatively flat at this point. Without better/more-robust tools (disclaimer, I started the DataMapper project), I don't see Ruby really diving into the Enterprise either where re-use/components are king, DBA's rule, and denormalized complex data models are the rule rather than the exception (it's arguable whether the complexity is justified, but in my experience it's there just the same)

The "switchers" from .NET/Java have switched. We aren't going to see a lot of new, well seasoned developers joining the Ruby fold. We're going to have to train up the newcomers. The people programming for the first time. This aspect at least seems to be growing steadily, but not at a rate exceeding other platforms as far as I can tell.

On the other hand, 11% (from the Evans data) is very significant. I suspect a lot of that is side-project work, or small scripts around the office ("sneaking Ruby into the system"), but still, nothing to sneeze at. Ruby is a core system language. Developers that use more than 1 or 2 of those on a regular basis are going to be fairly rare.

I'm actually fairly surprised. Considering the number of Ruby developers compared to Java or .NET developers in Dallas for example, I would expect it's actually much closer to half that at best that develop primarily in Ruby.

Still, enough to justify breaking into the enterprise if the tools become available. Which I think will happen one way or another over the next 2 years. Merb + DataMapper + JRuby + Log4x will eventually seep in enough to start making some real in-roads.

Rails itself is destined to compete with ColdFusion. Take that how you will. ;-)