One more political observation, which I hope won't piss anyone off this time

[maradydd]

As most of you have probably picked up on, I'm among that minority of computer scientists who actually writes code, and often prefers it to writing papers (much to the chagrin of my advisors and colleagues). I enjoy my theoretical work, but if I spend too much time on theory alone, the joy turns hollow; I want to build things that people can use.

Comments

[feonixrift]

I solved this in my head with the idea that maximizing effective solution of actual problem has its own mathematical elegance. Too bad this only works inside my head.

[maradydd]

I know what you mean. There's also a diminishing-returns issue with wicked problems, in that the point of diminishing returns itself is often not easy to recognise. :-/

[cnoocy]

I think you're about two steps from running into the halting problem.

[maradydd]

Unfortunately this is nothing new :(

[morbid_curious]

*nod* Both the Mathematician and the Engineer want to find a solution, but one is looking for a theoretical solution and the other an operant one. One wants to produce an idealised model, the other an implementation roadmap. When everything's working well, there's a synergy where a goal is identified and a path found. If not, then both are frustrated at the other's lack of clear goal/path inhibiting their path/goal. Of course, then there's the within-condition problems of different groups of engineers squabbling about the optimal route, and different groups of mathematicians squabbling about the optimal destination. Everyone has a slightly different model, and different conditions they deem sufficient for validating and verifying conditions under them

[maradydd]

but one is looking for a theoretical solution and the other an operant one. One wants to produce an idealised model, the other an implementation roadmap

Well, I think most political Mathematicians want to see their models implemented -- but are willing to spend a lot longer refining those models before implementing them than political Engineers find practical, and indeed perhaps longer than is practical.

"While we wait, people are dying," says the Engineer. "I know, I know," says the Mathematician, "but please let's not implement something that will make more people die!" Which is an extreme way of looking at it, and that extremism gets jumped on and distorted in ugly, dishonest ways by the Sarah Palins of the world. The situation isn't made any easier when people with an axe to grind jump in with completely made-up bullshit designed to prompt emotional reactions and get people to shut off their reason.

Everyone has a slightly different model, and different conditions they deem sufficient for validating and verifying conditions

[morbid_curious]

The short short version: research into understanding and improving virtual learning environments.

The slightly longer version involves capture and analysis of a bunch of quantitative data about learner experiences to complement and validate existing qualitative methods (e.g. pre- and post-tests, semi-structured interviews, questionnaires and direct observation). Hopefully this will allow us to better understand what people are actually experiencing in such environments, and thus develop better design techniques for providing effective instruction and learning opportunities.

The deadline's the day before I visit the U.S. for a month, and I'm also co-writing a paper for an ACM SIG submission. So it's a Fun Time All Round right now :-)

[maradydd]

Cool. Do you know steer? He's a mathematician in London who's been an instructor before, and ISTR he's used a variety of different virtual learning environments, both closed-source and open. He might have some useful observations.

I have a bunch of cascading deadlines from mid-September through mid-October, so I feel your pain :-/

[nibor]

I think this just comes down to At what point is the cost of the improvement not worth the benefit of the improvement - keeping in mind that part of the cost of the improvement is the opportunity cost of not using it already, and the impact of delivering now is that the improvement cost tends to be higher after initial deployment. This means that some functions would be worth implementing at initial ship time and not necessarily worth tacking on later, for example. It's more rare in the programming world, but in the world of politics I think that's a significant factor

[maradydd]

I think that last bit tends to get thrown away as politicians fear loss of "control" over their projects once they're set into the wild

You've put into practical terms something that bothered me enormously about a trend we started to see during the Bush II years, where certain bills (e.g., the Military Commissions Act) had provisions that attempted to restrict the scope of judicial review over the ensuing laws. Happily, we have a Supreme Court that's more than happy to say "no, fuck YOU" to that kind of weak attempt at an end-run, but the fact that it happened at all is troubling.

[whswhs]

Something very likely related to this can be found in Jeffrey Tucker's "A Political Theory of Geeks and Wonks," to be found at http://www.lewrockwell.com/tucker/tucker82.html . I'm a total geek, though by considerable effort I can manage to think about the wonkish aspects for short intervals.

[maradydd]

Thanks for the link -- I hadn't read it before.

It probably says something about just how much of a geek I am that the Mathematician/Engineer divide seems to apply at the geek end of the spectrum and vanish into meaninglessness at the wonk end.

[amberite]

Where HAS all the rum gone?

...I sympathize. And this is a really clean, succinct analysis of what makes politics tick. The frustration of large-scale systems is, in general, the fact that any fix is a matter of percentages: there's no scrapping an entire system and re-building it from the ground up, but there are patches -- and a patch might make the code a great deal more functional for a large percentage of the user base, but might also as a side effect break it for some small percentage who are busily trying to use a bug as a feature.

[maradydd]

but might also as a side effect break it for some small percentage who are busily trying to use a bug as a feature

Or might also break it for the percentage who aren't trying to exploit the system, but who rely on some aspect of it working the way it does. With respect to health care, one example is folks like whswhs -- independent artists, writers, musicians, consultants, entrepreneurs, &c who are in a position where they can afford what they need now, but if forced to shell out an extra 12% of their income would be put at a serious disadvantage. Back when whswhs was independently insured, he wasn't able to see a dentist or an optometrist because he couldn't afford the co-pays. Uninsured, he pays cash on the barrelhead and gets the care he needs -- and under the proposals I've seen most recently, in order to keep doing that, he'll be forced to pay a fine. It's a lose-lose situation. My inner Mathematician and my inner Engineer both balk at that

[amberite]

I didn't mean that necessarily as a malicious exploit, and totally agree that some of us rely on bugs as features because that's what's there. I actually meant more that if the present system is a piece of software, it's worse than Windows Vista -- and as such, I just don't think there's any bugfix that can be MacGyvered in without breaking the system for someone, and that is one of the things which deeply sucks about it. Situation Normal is All Fucked Up.

(If I can see any one thing that needs to be fixed in the situation of independent artists, btw, it's that the flat 15% self-employment tax should be taken out and shot, picked up and checked for signs of life and then shot again. At least insofar as it affects people whose income is close to or below the poverty line.)

[maradydd]

Vista's actually a remarkably good analogy. Windows has been hamstrung for a long time by its (perceived?) need to support legacy hardware, and Vista was the first Windows in a long time to say "no, guys, we're just not going to support some of this stuff any more." But the one-size-fits-all solution ... still doesn't fit a lot of people all that well. Linux, on the other hand, manages to avoid this problem for the most part by not assuming a one-size-fits-all solution; the 2.4 kernel is still under development for users who for whatever reason find it better-suited to their needs, the rest of us have moved on to 2.6. (And for those of us with really special needs, compiling a custom kernel or adding new kernel modules isn't especially difficult.)

You might have noticed I take a rather Linuxy attitude to politics. ;)

And, yeah, that flat tax has screwed me pretty hard in the past too.