Posts from — May 2005

I finished The Demon-Haunted World by Carl Sagan tonight. I had received it as a gift from Mike Cornnell a few years ago, but never got around to reading past the very beginning until this week. The book is excellent on many levels.

It’s difficult if not impossible to summarize the book, so I’m just going to write a bit about the things I liked the most and the things that struck me. At its core, the book promotes the value and importance of science, and the dangers of ignoring it. Specifically, it promotes the teaching and application of skeptical thinking in all aspects of life. Sagan goes into fascinating subjects including UFOs, psuedoscience, inquisitions (some of which were unexpected), literal and proverbial witch hunts, the hydrogen bomb, and science education. I undoubtedly learned a lot in the process of reading the book, and it proved quite thought-provoking. I especially liked how Sagan took a skeptical attitude toward even his own ideas, writing criticism and then defending against it.

The thing that struck me the most (though I’ve seen it before) was statistics like “Sixty-three percent of American adults are unaware that the last dinosaur died before the first human arose”. Even if the statistical error is 30%, it’s still terrible. It’s not like this is any new knowledge, like relativity or string theory. Scary. There’s a rant brewing in the back of my head (not literally, don’t worry) about how school is not challenging enough for American students and how we have to work harder or forget being so powerful in the world. Not tonight, though.

The book isn’t hard to read if you don’t know much about science, it’s meant to be accessible to anyone who enjoys thinking. In the course of reading this book, I came up with a couple of other things I’d like to read. First of all, I’ve always wanted to read A People’s History of the United States by Howard Zinn, so I’ll add that to my queue. Additionally, I’d like to read something about the story of the Manhattan project. Not the science or the military application; stuff about the people involved and the secrecy and such. Any suggestions are appreciated. I’m also interested in learning more about the Cold War, but I’m not quite sure what aspects… let me know if you have suggestions for that, too.

Next book on the list is The C Programming Language. As Barthol so eloquently put it, “GO TO THE EXTREME MAXX AND KEEP READING”. Indeed.

I was thinking about variations on Turing tests. Say you can communicate with some entity via a text chat, how would you determine the following:
- Is the entity a human or an alien from another planet?
or
- Is it a human from present day or a human from the past? from the future?
or
- Is it a human or is it some sort of entity with intelligence beyond that of humanity in present day?

All of these are really variations on a common theme: what question could a human ask without knowing the answer, but be able to verify a correct answer? (This sort of reminds me of Justice Potter Stewart’s “I know it when I see it”, but that’s unrelated) Carl Sagan suggests that a proof to Fermat’s Last Theorem would be an example of a good way to test a higher intelligence (though, according to what I’ve seen online, a proof was completed in 1996, but I’m not quite sure).

A little over a week ago I finished reading The Design of Everyday Things, a book I had been meaning to read for about a year.

The book is generally quite good. First the bad: at times it seemed that it was too repetitive and periodically needlessly dry. With that said, the book was worthwhile… the author codified many things that may be “common sense” but are not nearly as common as they should be. The book taught me how to be more critical of the design of the things around me, why some faucets are harder to use than others, and why I get lost in phone menus often. And maybe it explains why I still have trouble jumping into vi for more than simple editing. (ducks for cover)

The best part was the plethora of examples throughout the text, often with pictures: real-life examples of good and bad design made the book have much clearer applications.

I hope more people who design things will take the time to read this book. Too often people confuse “ugly” and “unusable”… I don’t mind using ugly software, for example, but if it’s just hard to use, that’s a completely different problem. The author may be a little arrogant at times, but considering how many terrible designs are out there, it’s allowable.

I’m now in the process of reading The Demon-Haunted World: Science as a Candle in the Dark by Carl Sagan. I’ll write about it when I’m done, it’s quite good so far.

After that I’m going to finally read K&R’s classic book The C Programming Language.

I have ideas for what to read after that, but I’m not going to plan too far in advance…

I’m considering buying the Linksys Wireless Music System. The software that is used as the server is MusicMatch, which only runs on windows (even if I were on windows, I wouldn’t use it). So, I looked around for linux alternatives and found the hilariously named Twonkyvision Music Server. If that’s not a name I can trust, I don’t know what is.

I wouldn’t mind doing some reverse-engineering in the worst case, anyway. It probably uses RTSP, and I had to “write” some RTSP software of some sort back in the days of network programming.

Oh yeah, I graduated today. The weather was nice, Maya Angelou’s speech was very good… I’m not even sure why I liked it, but I did (I rarely enjoy poetry). Graduation this year doesn’t mean all that much to me since I’m not going to stop taking classes, nor am I going to another school. It’s more of a milestone than a transition. I’ll definitely be sad to see some of my friends go, but the majority of them are either staying here or will be nearby.

With that said, Tim is currently in the process of moving to Texas (he should be there tomorrow), which is sad though probably a good change for him. Since freshman year, he’s been closest to my programming abilities (able to work faster than me on a lot of stuff, but not so fast that I get lost), which was very important, and through that we became close friends. I’ll definitely miss him, though I’m confident we’ll be able to keep in touch since we’re always online anyway.

I graduated magna cum laude, though I want to do better in grad school. Who knows how that’ll go, but I really think that I made extracurricular things come before school very often: I know my involvement in band adversely affected my GPA in some cases (and I think it’s worth it).

It was good to spend time with my family this weekend and say hello to friends I hadn’t seen in a while. Most of the summer residents have moved in to the house, which is good (though it’s quite a mess at the moment, but I’ll wait for the dust to settle).

I’m giving an informal presentation on doxygen and maybe subversion at a work meeting tomorrow, plus I have class 4-7 (public speaking!).

I recently downloaded and started listening to piano arrangements of Radiohead songs by Christopher O’Riley. So far, they’ve been excellent. Not only is it a good adaptation of the themes, it’s legitimately enjoyable piano music.

I replaced my blog today. I’m using new software on it now, and I’ll be updating it more regularly. Updates about my projects and other more technical things will continue to gravitate in that direction. Some old content was imported, but not all.

I won’t be updating that as regularly as this, so I definitely recommend using the RSS feed to keep an eye on it. I expect to write about updates to Run as well as some notes about CAPTCHAs within a week.

I just happened to look out the window as I was reading a web page, and I saw a woman in the driver’s seat of a parked car. She opened the door, leaned out, and vomited on the street. Then drove away. Not as hilarious as the “drive by vomiting” I saw last summer, but still rather ridiculous. Vine street has a lot of character.

If you use a headset to talk with someone on the phone, don’t eat a bagel with cream cheese while talking. Whoops.

Back to reading.

I went to bed at 11:30 last night in hope that I’d be able to get up early without a problem. Clearly, this effort cannot be successful. I was awake in bed until about 2 or 2:30, then fell asleep for a bit, got back up at 3:30, and got back up at 5, finally waking up and getting out of bed at 6:20, the time my alarm was set for. Not a great night for sleep, hopefully that’s the last time in a while.

In the morning I helped with registration and then attended the workshop. There were about 25 attendees, from a variety of companies (including Microsoft, IBM and Yahoo) and universities. I learned a whole lot about Human Interactive Proofs through all the presentations and was able to talk with the presenters during the breaks. I’m amazed how fast they can evaluate new ideas, it’s hard to keep up even when I know what they’re talking about. It’s nice to see a room full of smart people from time to time. I especially liked when people from competing companies worked together to try to improve their methods.

(brief technical aside)
CAPTCHAs, a popular type of Human Interactive Proofs are distorted text tests that look like this:

and is used to differentiate humans from computers (these puzzles are hard for computers but relatively easy for humans).

After all the presentations and panel sessions during the day, there was a banquet which gave us a chance to talk with all the attendees and enjoy some really good food (and some not as good wine, but that’s ok). I sat next to the guy who worked for AltaVista with the group that made the first CAPTCHA, he gave a presentation at dinner.

I’m definitely glad I was able to make it to this, I’ve learned a lot about HIPs, CAPTCHAS, and the academic world.

HIP is an acronym for Human Interactive Proof, and is sometimes pronounced like the word hip. I started laughing quietly when the Microsoft presentation on computerized attacks of these problems discussed “breaking HIPs” and “having strong HIPs”. Oh, man, the image analysis community is hilarious.