Už jsme tu kde jsme byli

--as they say, where I did my Fulbright (or "back to square one" as we say in these parts).

I find myself once again delving into the realm of computer programming. This week has begun with a return to where I was in the summer, before I (foolhardily, as it turned out) sought the support of experienced programmers to aid me on my journey to setting up my proofs-of-concept. Foolhardy, because it wasted a great deal of my energy and time, and netted me nothing but frustration. Truth is, I have no desire to become a programmer first and foremost. That said, I need to know enough to direct the efforts of those whom I expect eventually to hire. Nothing better for that than getting my feet wet.

If I were able to find a kindred soul, someone passionately interested in what I'm trying to accomplish, not terribly entrenched in the status quo of the field, eager to try something new, and who just happened to have a complementary set of skills to mine, well... I'd be in heaven. For now, I'll have to stick with planet earth, which means if you want it done (whether right I can't yet say) you've got to do it yourself.

In some ways, this is familiar territory for me. I've described myself (academically) as multidisciplinary to my core. The bane of the interdisciplinary is that one needs (all too) often gain at least a modicum of expertise in multiple areas. One needs to be conversant in multiple dialects and methodologies to be able to find the linkages between otherwise disparate fields of inquiry, linkages that have been overlooked or yet undiscovered. The trick (and it's no small matter) is to maintain a focus on the core of your inquiry so you don't bogged down too easily in great breadth with no depth. Properly done, I think it's like meandering through a woods, rather than sticking to a given path. I wrote a fable of interdiscipline on my old blog a while back that fills out this view.

There is always a balance to be struck. There are great rewards in being self-reliant in this way. But there are also more chances to stumble. Finding the balance, learning enough to carry on the conversation, enough to know when, where, and how to delegate is key. What I'm programming now is simply the tools to prove my ideas work. Once I can drum up interest in what those ideas can accomplish, I can safely hire programmers to make those tools better, and more broadly usable. For now, I'll settle on clunky, as long as I can maneuver it.

I did find a project (by a group of astro-physicists) that may be of great help. They've put together a template library for scientific computing in C++ that can be found here.

Wish me luck.