50 Minutes

I'm listening to a Robots in Disguise album called Disguise. I bought it sort of on a whim. It's rather better than I was expecting. Seems to me that their albums are more fun to listen to than than their singles.

I've been chopping chilli peppers.

I strongly suspect that the next time I go to take a whizz is going to Suck Rocks, Hard.
Christmas was fun, I've eaten enough turkey to stuff another turkey. The oven is now full of badly over|under cooked (don't know which yet, have to see how it comes out) biryani. Spent a few hours today wandering around Cardiff (Kaadiff for those who're too familiar with the accent) attempting to purchase videogames, literature, music and (mild) smut. Only really succeeded on the literature, games and smut - couldn't find anywhere selling most of the music I was looking for. I bought a copy of Pratchett's Making Money and spent about half of the (short) train journey back home giggling enough to worry everyone else in the carriage.

Pretty good day.

For the record:

The cutest thing in the universe wasn't all that difficult to find, actually.

Awesome on toast.

I just blew about 116 emails into various peoples' accounts - that's 29 apiece, for four people. All because I broke an SSH login and didn't notice quickly enough. Whoooops.

I can has employment! Yay!

Starting Monday, I'm going to be writing a whole pile of scripts for monitoring a company's web and database servers - all FreeBSD boxes at the moment apparently, though it's possible that that might change. Apparently the current issue du jour is the boxes running out of disk space in /var/ on a semi-regular basis as the logfiles grow, plus they want to know about things like whether or not the database tables are growing at roughly the right rate and so on. Should be fun.

Deadlines suck, narrrr

Title says it all, really.

Waiting for the (worms|flames)

So I've just written an absolutley monster-bastard-grade enormously long email to my coursework group, detailing just how awful every possible approach to our current situation would be. The subtext that I don't bother to mention in this particular email is that, hey, this isn't the only piece of coursework we have to do, and it isn't the only thing we'll have to take the time to study.

The other, slightly funnier, subtext is the identity of precisely who is going to end up having to put the extra work in - specifically, almost certainly me and one other member of our group. Which fact I'm sure is just going to look dandy when the examiners decide how many marks to award us for effective time management and effort delegation... especially since, right now, both of us have each contributed about twice as much time as anyone else in this group so far (bar one third member, who consistently works like Hell but, eh, isn't the best programmer on the planet - though by no means is he the worst, and he seems to be improving constantly and quickly).

So just for the moment, I think I'll sit here and listen to some cute German lady singing "99 Luft Ballons" and wait to see what fallout comes of this. Probably it'll take a while for this particular bombshell to go off, though. It was, after all, a pretty damn long email and hence should take a bit of time to read through.

Might as well get along with some of that other coursework while I wait. I really should be doing that more often that I do. ^_^

By the way, this is shaping up to be very helpful in order to be able to read mathematical notation aloud competently, as opposed to vocalising embarrasing sentences along the lines of "forall i in the natural numbers up to n, squiggle x subscript i equals loopy thing y subscript i".

Update: Naturally it seems that most of my group members are crazy and would love to dive head-first into taking the one path that happens to involve most work by a factor of about two. Great, I love these people. Presumably I should just ignore the feeling of panic lurching around in my gut. In my stomach, even the maternally-fornicatin' butterflies are throwing up right now.

Finishing up a good program is a rush.

Entirely irrelevant: this text box is a lot more daunting to start typing into when it's empty than when there's some text in it alreay.

Excelling in this subject (computer nerdery, particularly programming) is pretty darn awesome. What you've done works, even works well. Exactly how well it works is entirely the result of your own effort, and is entirely your own responsibility. You can improve it arbitrarily by putting in any quantitity of additional effort you feel like. It's a high (*) from awesome things happening and you being in perfect control of them. You make what you intend to see done perfectly and precisely clear to the machine and to the reader - and, in the process, to yourself - and then the machine goes off and does exactly the right thing - because you have told it to do exactly the right thing.

My goal in life, for the moment, is to hit up that rush as often as I possibly can. For the moment, what I'm doing well is managing to blast through, by dint of good coding habits, tasks that I ought to be failing miserably by dint of my awful work habits. What I need to do better right now is improve those work habits so that deadlines are more comfortable and my marks will come out higher as a result. Simply resolving to do better about this isn't going to succeed - I know, I've tried it. I just end up making myself feel moody and annoyed about things. What I'm going to do instead is to try to do things in two big bursts - one at the first minute in addition to my usual one at the last minute. Time in the middle I'm going to worry about less, because I'm usually best off spending that percolating over the problem at hand anyway.

(* oh dear, seems I might be a control freak? Maybe.)

Wheee, hissy fit.

The least fun bit about group projects is probably the infighting. Today I'm vaguely bummed out by an internal hissy fit over exactly what features we end up implementing, and how many corner cases are worth actually getting perfectly right (rather than just letting some corner cases slip in the interests of keeping our programs simple). I personally was the one arguing in favour of being lazy and just going with what we have already written and working. The whole argument terminated with the other participant walking off in a terrifically bad mood. Can't blame them strongly for that, since I myself had already gotten fed up enough to stop listening entirely.

Whining alone won't fix this, though. So instead I'm going to go work out in rough detail how much time each of the possible approaches we could take would cost. I suspect most of the group will quickly shift in favour of laziness when I quantify and explain roughly how much time making things perfect will take, particularly if I compare it to how much time we actually have in which to do everything.

Almost an allegory

Herding cats is trivial, provided you know the trick to it and aren't in too much of a hurry. Just put something the cats want where you want them to go. Shaking a box of kitty biscuits works on some tame cats, firing up a can opener on others. If you want them to stay there, add something fluffy in a warm spot. Cushions and sunbeams rarely fail. It's only feral cats that're ever particularly difficult. Even then, all you need are persistence and fish.

Perhaps no one will notice that I've posted this.

Oh well.

From Alan Perlis' Epigrams on programming comes the following nugget:

"Dealing with failure is easy: Work hard to improve. Success is also easy to handle: You've solved the wrong problem. Work hard to improve."

- something I more or less live by. The case that that quote doesn't handle, however, is a lack of feedback, positive or negative, which I find much harder to handle. You can't even safely default to the eventual "work hard to improve" path, because you don't know where to put more work in. The path of least ambiguity here is to set on fire the person whose responsibility giving feedback is supposed to be, but even that's still slightly ambiguous. After all, they might think (in the short period during which they still can think at all, that is) that you're setting them on fire for some other, entirely unrelated reason.

(* Disclaimer: Setting people on fire isn't (usually) necessary. Most of the time, a quick flame war will suffice, with the additional benefit of not damaging the carpets.)