start | find | index | login or register | edit
2004-03-27
by earl, 7573 days ago
It's really nice to see the traction building around Groovy and we can only hope the best. Let's hope that history will prove Bill de hÓra right: "Groovy's legacy will be that it was [the] language that changed how we think about the platform [the JVM, ed.]."

I think, however, that Groovy (at least 'til today) missed one historical chance: to re-introduce the power of the REPL to the programming mainstream. Groovy's "make up", with all it's scripting and other syntactic sugar, is still strongly influenced by the "edit - compile - test" mindset; decent REPL's or other interactive environments (e.g. Beanshell-like stuff) are (still) missing.

The power of interactive environments and "experimental development" is something that deeply changes the approach people take at building software. While I could write on for hours about this topic, I think I couldn't put it more precisely than a software developer of a big financial company put it at a recent seminar I attended:

"Once you've got a productive K programmer, it's like you've got your hands inside the data."

And this is quite true for all interactive environments. This is why Mathematica and IDL and the like meant a huge improvement for a certain set of people. Once you can interactively (and experimentally) work in your problem domain, it becomes incredibly productive. A wisdom long known by Lisp and Smalltalk programmers (and their kind).

While it may be advantegous in some regards (e.g. acceptance-wise) that the development seems to currently concentrate on proper compilation (JVM bytecode generation), I think that improving Groovy's interactive capabilities could not only change "how we think about the JVM" but could deeply change how mainstream programming thinks about building software.

"Libtextcat is a library with functions that implement the classification technique described in Cavnar & Trenkle, 'N-Gram-Based Text Categorization'. It was primarily developed for language guessing, a task on which it is known to perform with near-perfect accuracy."
powered by vanilla
echo earlZstrainYat|tr ZY @.
earl.strain.at • esa3 • online for 8692 days • c'est un vanilla site