|Think about Loose Coupling|
Re^5: Death and Return of TIMTOWTDIby jepri (Parson)
|on Jul 07, 2004 at 16:06 UTC||Need Help??|
That's cool. It's nice of you to swing by and explain what was up, but honestly I'd already forgotten the whole thread. I hope whatever was depriving you of sleep eases up. Personally, I've got a lot of time for people who fly off the handle every now and again. The life experiences that brought me to that way of thinking would take a lot of explaining and wouldn't make much sense to anyone anyway, so you'll have to take my word for it :)
I think I can see where you were you were going with your points.
Natural selection is often quoted by computer people who possibly don't see all its ramifications (see my homenode for why I know more about natural selection than computing :). The popularity of programs can probably be described better by economics than evolution (e.g. programs don't die, can't breed: marketplace theories probably can do better).
I can see your point about pride of writing popular software, but the people who are maintaining it are often in that position because they were among the first, breaking new ground. Everyone used their work and by default they ended up as "the author and maintainer".
The usual economist response to a saturated market seems to be "open up a new market", in this case by developing something so compelling that everyone has to have it, right now. The economics breaks in both our arguments when anyone can become market leader by taking the leading product and improving it slightly.
The trouble I have with discussing software development in any of these terms is that they don't really apply. Software development isn't driven by an evolutionary need to survive and reproduce, nor is it a product striving for market share. If I had to define it as anything, I'd see it as research, taking ideas from the great unknown, and wrestling them into a usable form for others to benefit from.