we'll i for one don't want to miss out on a java/blog melding of the minds ... oh, and the beer will be relished with delight.

Java == platform independence
XML == application independence
JXTA == network independence

Secure End-to-End Computing

in my ears: RadioHead/Amnesiac/You and Whose Army

I've heard it said, at times, that "things get worse before they get better." I'm unsure as to the story behind this phrase and out of context it is not entirely inspirational, to me at least ... but nonetheless the phrase is entirely appropriate for the bit'o slogging I have been involved with as of late.

Here's the skinny. We got JXTA 2.3 out the door and on time which is fine, noble, etc ... but, in doing so this go 'round I was directly exposed to the herculean effort required to connect all the disparent pieces and effectively making the process appear "seemless." Being a "glass is half-full" kind of a guy, most days anyways, I acknowledge that there is typically always room for improvement and as such spend little time rooting out causes short to say that I too, as an application/framework/etc contributor to the overall deliverable am entirely at fault for not keying in to the "big picture." Put another way, generating a release shouldn't be a life altering event. Optimally, a release, or more precisely the process behind a release, will just be another turn'o the key, another drop in the bucket, just another day, etc. as we all have more interesting and necessary work to chew on. For me, specifically, I was was unable to add the more interesting JavaOne JXTA Lab excercises as I had initially strived for (I'll get 'em in next go 'round), nor shored up the JXTA Ext package as fully as I would've liked, or add kew1 new features to MyJXTA, etc. Making compromises is a part of the gig, I accept that, but, moving forward, I intend to do my steady best to contstrain to the fullest extent possible exercise that are not entirely productive when taking a holistic perspective.

So, why am I not as gloomy today. Turns out the legacy build process "just gave out" as it understandably made a number of assumptions to work it's mojo. Alas, things changed, which in and of itself is a good thing (I'm a *huge* XP fan), and the gears ground to a screaching halt. What now. Long story short, I've effectively factored away 5.5k lines of platform/system specifc scripts/data files/etc in favor of roughly 200 lines of Ant files that rightfully delegates to the projects themselves to do the heavy lifting and as such are free to (largely) vary over time. Further, I can run this process on any system, incrementally, reliably, etc. More work is required but at this point in time I see the light ... or, at least I'm happy believing I see the light. As an aside, there is an interesting downside to taking on this type of problem domain work that I'm not entirely thrilled about but given we have a net gain I'm happy to report "it's all good."

To summarize, when something doesn't feel right, fix it when you can vs. when you have to. Just *$%# do it. The rainbow on the horizon ... why Maven of course.

Java == platform independence
XML == application independence
JXTA == network independence

Secure End-to-End Computing

in my ears: RadioHead/Amnesiac/Dollars & Cents