Sleepy insights

[jack]

Apropos to my sleepyness my ideas from last night were not amazing, but all were fairly good. A wonderful quote on the subject from The Universe of Discourse is:

The next morning I remembered nothing but that I had had a "clever" inspiration while guzzling whisky from the bottle. "Oh, no," I muttered, "What did I do?" And I went to the computer to see what damage I had wrought. I called up the problematic part of the program, and regarded my alcohol-inspired solution. There was a clear and detailed comment explaining the solution, and as I read the code, my surprise grew. "Hey," I said, astonished, "it really was clever." And then I saw the comment at the very end of the clever section: "Told you so."

I don't know what to conclude from this, except perhaps that I should have spent more of my life drinking whiskey. I did try bringing a flask with me to work every day for a while, about fifteen years ago, but I don't remember any noteworthy outcome. But it certainly wasn't a disaster. Still, a lot of people report major problems with this strategy, so it's hard to know what to make of my experience.

One's subconcious is sometimes witty. And sometimes effective.

Coincidently, recently, I've got the hang of commenting my code well! Cool.

That blog, by Mark Dominus, is very intriguing. I think fanf or senji linked to it, but every few days he has something to say about something obscure but relevent to every day life, that always makes me think. If I had the time it's a good example of the sort of discourse I would like to output.

Comments

[miss-next.livejournal.com]

I never drink enough alcohol to have that effect, but I do notice I am often more creative when I am very tired, which I suppose is similar. sir_dave also writes funnier poetry when he hasn't slept properly. :-)

[cartesiandaemon.livejournal.com]

I often have my best insights -- and do quite well at drudgework -- when I'm just slgihtly tired or oiled. Sometimes a pint at lunchtime really does help.

I think both are removing wibble and distraction and letting one get on with it!

[simont]

I think that's a good explanation for the drudge-work flowing more easily, but I have a different theory about insight and intuition. When you're tired or drunk, it becomes harder to think coherently; instead of your mind being filled by an orderly sequence of closely related concepts proceeding from a beginning to an end, it instead becomes filled up with random concepts from wildly varied contexts drifting around aimlessly and bashing into one another. And this is fertile ground for creative insights and intuitions: it's one thing to be trying to understand DNA in a lab which has a spiral staircase, but in order to discover the double helix you have to be thinking of DNA and looking at the staircase at the same time. Allowing concepts from all over the brain to collide in a disorderly mess is the chaos from which surprising insights arise.

At least, that's the way it seems to work for me. I don't feel as if I've been genuinely creative very often, but when I do it's usually because two thoughts collided in my head which it had never occurred to me to put next to one another before, and this is certainly something that happens more often when I'm tired.

[cartesiandaemon.livejournal.com]

Hmm. They're related -- the turning off of inhibiting bits of your brain. But in my case for drudgery turning off distractions and in your case for insight turning of embarassment.

That does make sense, cool.

[ewx.livejournal.com]

Commenting your own code is good. Learning to comment other people's opaque code as part of the process of understanding it after they've left is also useful l-)

[cartesiandaemon.livejournal.com]

Yes. The trouble is, you get lots of comments prefaced with a symbol meaning "My best guess at what Fred was trying to do is explained here, but that bit doesn't make sense yet unless he was smoking crack" which helps someone else[1] understand it, but isn't tidy.

[1] I include yourself, later, as someone else.

[ewx.livejournal.com]

If you discover a bit that's mad then perhaps that was the cause of the bug you're trying to track down (presuming you're not reading this obscure code for the fun of it) and you might find it best to make it not-mad along the way...

[cartesiandaemon.livejournal.com]

Yes, that's often wise. I'm generally inhibited by the many other people's bits of code working with it that will be broken if I change it in any non-equivalent way on purpose or by accident, and the feeling that once I start rewriting bad code I'll never stop, and a fear that it's doing something clever which I don't get -- the last true empircally about 1/3 of the time :)

[ewx.livejournal.com]

The example I had in mind I've become probably the main expert on and our project is the main user. So if I break it (more l-) then it's almost certainly me who gets to clear it up.

[cartesiandaemon.livejournal.com]

Sorry, that is, I assumed something of the sort, but I can't easily. But I'm doing something wrong, because most bits of code it isn't clear if anyone owns them or not, so I should be bolder, but it gets so tedious asking *everyone* if anyone knows why it's done like that.

Some people are bolder, but it's rather annoying when fixes one person puts in are deleted by other people. I guess it encourages people to comment them, but it's still a problem.