oh yeah; you know the ruby slippers was wizard of oz, right? not cinders

Ryan, that’s a *great* line. As for the reservations. they’re most valuable now in not scaring developers and architects - “don’t worry we’re only using it for this - you’re safe”. You have to appreciate how threatening disruptive technology is.

“We both know that dynamic languages are going to break into the enterprise at some point.”

I would say they’re already there. My employer has backed Jython/Python for use in the enterprise for years. It has not always been easy, but it’s been great to see the attitudes soften and people come around the idea. It helps that the likes of IBM and Sun (with Jython support in Netbeans) are starting to get people across the chasm. But consider it was all available to the enterprise 10 years ago - it must be heartbreaking to be a Smalltalker.

IBM are *so* strategically smart about open source. PHP is a great choice - alternative technology to JSP/Struts/JSF on the front tier doesn’t conflict with their core J2EE offerings.

i’ve long believed that with obvious exceptions - SABRE’s airline systems, ERP implementations, etc - application development is largely transitive in nature because so many of today’s enterprise development tasks are tactical, not strategic.

in most if not all of these cases, dynamic languages are a better fit for development because of the inherent advantages in resource requirements, developer productivity, etc.

but when evaluating the different options amongst dynamic languages, i do believe it’s important to consider adoption and size of developer community because even tactical applications may need to be upgraded or maintained.

thus i think IBM’s nod towards PHP is a crucially important one, b/c it’s one of the largest enterprise software companies in the world standing up and explicitly singling one of these languages out as investment worthy.

does that mean there isn’t room for Python, Ruby, et al? no. PHP is admittedly an inelegant platform and language, and it’s certainly not perfect.

i do believe, however, that it’s as yet unclear which of the potential candidates will emerge as a mainstream application choice. while they can all find niches, and remain popular, i frankly don’t know if there’s room for both Python and Ruby as Java/.NET-style first-class citizens within the enterprise. maybe, but i doubt it.

but you’re absolutely right in that there’s absolutely room for less. quite a bit of room, in fact.

I’ve been under the assumption for some time now that it would be a single boiling-point technology. RoR would seem to be the best bet for that form of adoption at this point but lately I’ve been thinking it’s happening a little differently.

It seems like there’s a gang of dynamic-language technologies that are amassing around the gates with siege engines and catapults: {C,J,Iron}Python, Ruby, PHP, Groovy, etc. We’re pulling a developer here and another there and then they pull a developer here and another there. It’s extremely rare to find someone who takes the time to evaluate Python or Ruby and then just walks away. Python ate me - two days in the language and I was completely sold. I don’t see any reason this should be different for any other developer.

What we’re going to see is the sloooow seeping of these technologies into different cracks in enterprise development until one day we wake up and realize that 50% of all enterprise coding is done in some form of dynamic languages.

Right now it’s all about making recommendations with reservations:

“Jython is great! …for unit testing and debugging Java code…”

“RoR is great! …for throwaways and skunkworks.”

“PHP is great! …if you just want to do something quick.”

“Cerise is great! …if you don’t need all of the strengths of J2EE.”

Pretty soon we’re going to realize that the “exception cases” we’ve been slotting these tools into make up a majority portion of enterprise development.

There’s still a ton of room for less!