Rushi on Prolog and dairy products
Sunday, October 29th, 2006Like, you tell it, ‘I like cheese,’ and you ask it, ‘What kind of cheese should I get?’ And it tells you ‘Gouda sounds good today.’
Rushi on Prolog and dairy products
Like, you tell it, ‘I like cheese,’ and you ask it, ‘What kind of cheese should I get?’ And it tells you ‘Gouda sounds good today.’
Rushi on Prolog and dairy products
I always finish my Netflix homework over the weekend.
Teresa on homework priorities
<ben> Write it out in pseudo-C, then test it. <rushi> Pseudo-Ruby. <ben> What the hell is pseudo-Ruby? English?! <rushi> Ruby.
In the general case, percent of work completed is a logarithmic function of time spent.
When you start work, you get things done at a pretty good pace. Over time, you start to slow down. As the task approaches completion, you really slow down to accomplishing almost nothing.
This is a good approximation for time significantly less than the deadline. However, as the deadline approaches, the derivative increases rapidly, following the gamma function. Immediately before the deadline, work is accomplished at a nearly infinite rate.
As the deadline approaches, you work faster and faster to get it done; the last hour before the deadline is the most productive time known to man.
I have seen this pattern in almost everything I do, and I’ve been refining this theory over the past few months. I have further noticed that the pattern applies to most of the engineers I know. There are notable exceptions: some people cannot bear to leave things to the last minute, and get as much as possible done ahead of time.
These people are silly.
The next step would be harnessing my own pattern of productivity. False deadlines, drastically inflated requirements (to keep the early logarithmic rise going for a long time), and making that gamma function kick in earlier: all these will let me get more done.
Why am I practicing ‘ichi’? That’s dumb.
Ichi is the kanji for “one”. It looks like this: —
So if there’s something I can’t order at a restaraunt, I usually just point and say ‘kanji ga yomenai desu ga, kore wo onegaishimasu’ [I can't read this, but, one of these, please].
Some take a different approach though. Last night Ryan just said, in English, ‘I’ll have the unagi [eel] nonsense, please.’
Eric on coping in a foreign land
If you were a scientist or engineer working on a technology capable of launching a Singularity, or alternately destroying most life on earth, what kind of warning sign would you put on the wall behind the lab bench?
As Ben demonstrated tonight, Beer + Xilinx is a slightly better idea than Beer + Car and a slightly worse idea than Beer + Scissors.
TritonLink now has a Class Planner (beta). I’d link to it, but you need to be a UCSD student to see it (which is silly).
Messrs. Dye and Pajak have done a very good job. It feels a lot like TritonSchedule (which makes me feel really validated, because a pair of professional UI designers made the same decisions I did), with some nice improvements. They did a very good job on the lecture-discussion selection flow, and the department selection is just glorious. They fixed the grid width in pixels, which makes the positioning a lot easier. It’s all ajax, with almost no latency (on campus wireless); it’s very smooth and fluid. I don’t like that the lecture doesn’t immediately appear when you click it, but so far that’s my only complaint.
When I don’t have a CSE141 lab and an EAP application that need to be done urgently, I’m going to go under the hood and look at their JS code and ajax transfers.
On the whole, this is a Good Thing. My hat is off to ACT today.
I don’t know, but it’s delicious, and I haven’t even tasted it yet!
Ben on Rushi’s mixed drinks