dimator | head

It’s been exactly one year since I bought my Canon SD10. It’s a 4MP camera with a small form factor, which is good for stuffing in your pocket at parties and other events. So far, I’ve had a great time with it. I keep it holstered on my hip whenever I expect good times. No optical zoom, but for my uses (close up portraits and party shots), that’s not a big problem. My only gripe: very bad red-eye in a lot of photos.

I see that an SD20 has been released, at 5MP, and the same form factor. Not sure what else is different, though.

Mark Cuban has some good opinions sometimes. He writes:

Could there be anything more confusing and shocking than to read that our country was offering $35mm in aid to the areas affected by the Tsunamis, but that the cost of inauguration parties would be about $40mm ?

For this reason, Cuban thinks the inauguration parties should be cancelled. A great idea, but it won’t happen.

This country is run by corporations. Corporations pay big bucks to get time with politicians. Hell, corporations pay big bucks to politicians, and I think the inauguration parties are arranged mostly for this kind of $$/politican mingling.

For Giftmas, I scored a USB/MIDI interface. I have never been much of an audio head, but I got it in the hopes that I can create some decent music for my GBA games. I think if I stare at the daunting interfaces that most MIDI sequencers have, I might figure out what the hell I’m doing in about a month. Here is a list of OS X audio related software.

The MIDI’s can be converted to XM, which is the format played by the Krawall GBA sound library. At least, that’s how it’s all supposed to work.

I have a bunch of Word documents with various entries in them. I want to pull the data out of these entries. These entries have a specific layout on screen, and by looking at them, one can tell the parts of an entry, its meanings, use, etc. The parts of an entry are usually distinguished by special seperating characters or by certain fonts. There are way too many to manually examine and extract, though. I need a way to script the process of looking at an entry and pulling the data I need. Their structure is fairly uniform, but with enough variation that I can’t simply export the documents to plaintext and infer meaning line by line.

I thought of using Word to export the documents as HTML, so that I could use the mark-up surrounding certain parts of an entry to identify it. If you’ve ever seen HTML generated by Word, you know that its an atrocious mess. I got tired of looking at it right quick.

What I’m doing now: I’m using OpenOffice to import the Word documents and saving them in the OO document format. Looking at this (XML) file (after piping through xmlpretty) is much more pleasant. OO organizes each part of each entry in a very sane manner, such that I can now regexp for a certain style name to get at a certain section of data:

<text:span text:style-name="T21">
DUCT
      </text:span>

I was trying to figure out how I could get an external (XML) file into a Java archive, using an Xcode project. I knew I could manipulate the .jar file after building it, but there had to be a cleaner way to do it, via Xcode. Google turned up no answers, but then I found what I was looking for in a mailing list archive. It was here that I realized something:

The most important skill I have learned, as a developer, is where and how to look for answers. Often times, Google is the answer. It is the most important reference tool I have. Google groups, in particular, is where I hit first, because it has a very good signal/noise ratio. If the answer still eludes me, I know that I have to dig into documentation, mailing list archives, message forums, example files, etc. (Let’s not forget AIM!)

In addition to where to look, one has to know how to look. I’ve learned how to phrase my queries in the most general way to turn up what I might be looking for. I know which search results to skip over, and which to dig into. I know how to use the right terminology when posing a question, as well.

I’d say using Linux for many years helped me greatly, because it forces you to learn how to use information resources. Otherwise, your CD drive won’t work, your X won’t be the right resolution, etc. There’s no book on how to search for answers. This is stuff you have to learn along the way.