--
Ola Bini (http://ola-bini.blogspot.com)
JRuby Core Developer
Developer, ThoughtWorks Studios (http://studios.thoughtworks.com)
"Yields falsehood when quined" yields falsehood when quined.
--- Begin Message ---
I tried to post this to the jruby-dev list, but it bounced back, so I'll try
sending it to you directly.
A computer-language implementation is fundamentally a low-level enabling
technology. Its primary audience is developers and other engineering types.
It's true that many of these people who work in enterprises have in the
backs of their minds goals like "sneaking Ruby in." But why would they have
such non-technical goals like that in mind? Primarily because Ruby has
*technical* benefits that they'd like to take advantage of.
So the key drivers in the target market for the JRuby initiative are all
technical in nature. *Always activate the key market drivers.* Make a
product that is technically the best (and the sweetest) possible.
On this basis, the Java 4-vs-5 question becomes: can we get benefits that
will excite our target market by going to 5? Now at this point, Ola, you've
segmented the target market in an important and interesting way: there are
the developers who will eventually *use* JRuby, and those who will
*contribute* to JRuby. For the latter segment, the benefits of 5 are
immediate and compelling. Are they for the former? Answer this question
first, and then ask how important these developers are to the success of the
project. My gut feeling is that they are *very* important, because getting a
lot of satisfied, grateful users is the key to being happy that you worked
on an open source project. And by the way, even a small improvement in
performance is a compelling technical benefit, given that the major rap
against Ruby is slow performance.
What about infiltrating the enterprise through the backdoor? Look at Java
itself in this regard. It's true that Java benefited greatly from Sun's
corporate PR effort. Even so, it faced the same barriers to enterprise
adoption that Ruby does now. Even with Jruby as the vehicle, it seems
reasonable to assume that success (defined as a reasonably large penetration
of Ruby in enterprise shops) will take several years of unflaggingly devoted
advocacy to achieve. Let's not be afraid to do this work!
(J)Ruby will never be brought into enterprises by IT managers. It will be
brought in by programmers. Make the programmers happy and (though it will
take longer) they will make the project happy.
--- End Message ---
---------------------------------------------------------------------
To unsubscribe from this list please visit:
http://xircles.codehaus.org/manage_email
