Sunday, 24 April 2011
I have little to say about myself these days. But it's probably necessary to say a bit more than appears on other pages. Um, let's see. What would be useful? Okay, suppose I wanted to help you predict my behavior. What is helpful along these lines?

If you like Myers-Briggs cognitive style classifications, I'm an INTP, which means I'm introverted, intuitive, thinking, and perceiving—versus other categories. I might not seem introverted, but I get a lot of stimulus from thinking, and I have a high need for cognition. My IQ is a few standard deviations above average, which is not exceptional in computing.

I like people and I'm agreeable. I've been told I'm too accomodating and I should say no with more vigor. I did learn how to do that, so the answer is, no. I had that naive expectation other folks are kind when I was younger, but it was just projection. Now I trust little.


Sunday, 24 April 2011
I plan to write about code in C and C++ here. You might get a misimpression I think those are awesome, and that everything else sucks. Not so. I also like Smalltalk and Lisp of the simple sort like Scheme. I have written my own toy versions of those, and expect to do so again. Maybe these will appear here as a roll-your-own scripting language.

I expect both Ruby and Python are awesome. I just haven't used them much. The same goes for other languages. Use whatever works. I write in C and C++ because that's what folks want. I'm not in the business of getting other people in line with my own agenda, but I know that's a normal bias many folks have: to do anything in pursuit of self interest.

If you insist on seeing what I write as advocacy, then assume my bias is the following. If you are stuck writing a lot of C and/or C++, and have to implement everything yourself, you could do worse than to imitate what you see here. And here you might later find a scripting language compiling to C. Do I think that's a good way to get C code you need? Yes.

ends vs means

Sunday, 24 April 2011
Which is more important: end results, or means of getting there? In the tech world this question reads: is the app more important, or the way you write the app? Entrepreneurs say app. Geeks say how you write it. Both agree the end app is essential as the main use case. So it's only a matter of whether you think how code is written also ranks high.

Personally I care about means, so I enjoy making things work more than shipping apps. The latter must occur to pay the bills, but inventing is the part I like. You get more social kudos by shipping apps, so it doesn't matter how you write them. But I don't care about status.

That might help you grasp why I re-invent wheels: it's the fun part. So I like getting work where that's the objective—making new ways to get new kinds of app. Best of all is new apps to help folks make new things: authoring tools. Most boring is a tech focus on enabling people to talk to one another: efficient gossip machines.