Had some more thoughts about careers. I like *playing* with things. I 
like new languages. I like code that doesn't take more than two weeks to 
write. I like playing with css designs. I like things that I think look 
neat. I like equation twiddling kinds of maths, and I like understanding 
proofs but not so fond of thinking of them.

My working technique tends to be along the line of doing something
full time for a few days, and then moving on. When revising or going
through lecture courses, I'd understand very little of the course
during the course. When complete, I'd sit down, and spent a day or so
going through the whole thing, getting the hang of the *shape* of it,
the point of it. The details hang off a clear understanding of what
the thing is, imo, and learning a lecture at a time covers the shape
of the thing in details.
When it comes to coding, I like to do a similar thing: less of the
shape of thing, and more about getting it working, but the same thing
about immersing myself in it for a few days, until it's done, and then
moving on. Things that don't look like you can do that, I get tired of.
I dislike the proper software eng/planning method of coding. Sometimes
it's obvious that doing things right is the only way of doing a
project and getting it working, but it's usually more obvious to me
what's going on, or what should be, when I make it do so. I've found
that getting stuck into things, and then when I've mushed it up
rather, but have a clearer picture in my mind of what should be
happening, works best for me. This may slow me down rather, because it
basically means restarting everything.

I'm capable of understanding most things, when I put my mind to it;
this isn't as obvious as it might seem; for example not everyone in my
year succeeded in making sense of Martin Richards' ascii diagrams
. But I'm generally not so quick at picking things up as some people
seem to be, as they sit around and chat. I like to think on paper (or
into text files). I understand things better when I have a sense of
the shape of them, it's taken me a while to feel comfortable in my
current position, although I've known all along that I would get the
hang of it eventually.

I also like playing with websites; again---small things. Coherent
whole. And a sense of shape; the shape of a page. And I like the
design/art aspect of making a page pretty and messing around with
CSS. Would I want to do it full time? Would I get irritated if other
people's sense of design were forced on me?

I had some other thoughts, about patterny stuff...
I like automation. I like patterns. I am interested in how to spot 
patterns---I spent some time writing a sitemap program, and in order to 
make it as useful as possible I had to do a good deal of working out how 
to work out the shape of the site, so to speak, beating myself over the 
head with trying to detect the logic of the thing. A combination of the 
way links were found on pages/other pages and the underlying directory 
structure. The way query strings were used to deepen the structure. I am 
fascinated by the idea of being able to detect patterns from data, by 
understanding which data to absorb and which data to reject. 

Every time you go out in a group of people, you see vast quantities of 
interactions and reactions from people. Some things you don't see, 
because you're looking the other way, thinking about other things. Some 
things you hardly notice, because your brain rejects them as 
insignificant before they hit your consciousness. Some of them, by 
keeping, you can connect into existing patterns. Some of them you throw 
away. Recently I have been trying to understand people, trying to guess 
how they feel about things, why they react the way they do. All the time 
my brain has been murmuring 'patterns, patterns'. I feel sure that if 
only I could collect all the data, sift it correctly, organise it, I 
could create a pattern from it which would enable to understand things 
better.

In a more general sense, with my sitemap program, first I had to work 
out what the logic was, and then feed it to a computer. What are the 
possibilities of just giving the computer the data? The logic and 
connections that I've failed to spot; could it spot them? Could it work 
these things out? When I wrote my testing code for my 1B project, one of 
the things that interested me was writing it so that I had to tell it as 
little as possible.

So, being a somewhat lazy, unmotivated fool, I haven't really looked 
into career or research areas related to these issues of data and 
patterns and inference. I don't know what's out there, and I should have 
looked. When I was looking for things to do for my part II project, I 
really wanted to do something more related to the idea of web searching, 
of the quantities of data out there and the difficulties of finding 
things out. Of the possibilities that exist, and the issues of erroneous 
data. I got around to thinking about it too late, and the project got 
contracted to what it was. However, I have some collection of links to 
search. Food for thought.

"The trouble with Big Grandma is that she doesn't forget things. She 
notices everything and it goes into her head and makes patterns. Or 
something. So the more she notices, the more she knows."

Noticing things; thinking things are obvious, people not finding them 
obvious. The clues are all there, but no one picks up on them. Hunches.

Google skills. Finding websites... not necessarily ones google knows 
about yet. Use the knowledge you have creatively.