Blog

  • Hotblack on Business

    January 4, 2005

    Was talking to Sunil via the ole MSN. He started his own business, Insyst, back in the old country a couple of years ago. I picked his brain for some insights, and he generously he gave! Paraphrased slightly, they are:

    • Decide whether you want to position yourself as an art guy who can program, or a programmer who can do art
    • Decide whether you want to do one-off project work, build a product, or componentized project work

    • Find a strong business development team to partner up to

    • If you want staff eventually, start writing davecorp methodology

    • After developing your methodology, think about how to hire people that can be transformed into mini-Daves

    <

    p>I never thought of writing down an actual methodology first. Usually you see this advice expressed as “develop a business plan“. It seems more effective, and even slightly more sinister. Mua ha ha ha!

    Read more
    DSri Seah
  • Another Fine Rant on Computer Science

    January 3, 2005

    In case you missed it, Joel Spolsky has written another interesting article with advice for computer science students. An excerpt (emphasis added by me):

    Learn C before graduating […] if you can’t explain why while (*s++ = *t++); copies a string, or if that isn’t the most natural thing in the world to you, well, you’re programming based on superstition, as far as I’m concerned: a medical doctor who doesn’t know basic anatomy, passing out prescriptions based on what the pharma sales babe said would work.

    Hee hee. Joel rocks! :-)

    Read more
    DSri Seah
  • The Tyranny of Wiki-Democracy

    January 3, 2005

    Wikipedia is pretty darn cool because it’s a group assemblage of knowledge from anyone who wants to participate. It’s self-correcting to some extent, because every addition and modification is visible to everyone else. According to the creator of the Wikipedia concept, though, there are major problems:

    • A lack of public perception of credibility, particularly in areas of detail (what Sanger calls “anti-elitism”)
    • The dominance of difficult people, trolls, and their enablers.

    These are fundamental problems with any organization, not just Wikipedia. When an organization gets large enough, inevitably agendas begin to diverge. In the first case, it’s “who is trustworthy”, and the values that go with that. In the second case, it’s abuse of the system to further personal agenda, and trying to fix it without destroying the qualities that allowed the system to work in the first place. In the US, we feel free to leave and start our own new thing to maintain “mission purity”, which tends to leave the parent organization with the people who are thinking only about themselves and the “rules”, resulting in stasis and/or loss of vitality. The author doesn’t quite say this, but does note that it happens over and over again.

    I have no solutions, except to try to understand motivation versus intention as much as possible, in any circumstance, and make a judgment call whether it’s “for the project” or “for the individual”.

    Much interesting discussion follows in the comments on Kuro5hin.

    Read more
    DSri Seah
  • Video Games and The Line

    January 3, 2005

    ScreenshotCame across this link to JFK Reloaded, which is apparently a simulation of the assassination of John F. Kennedy. It uses video game technology and ballistics to recreate the crime scene. It’s a fascinating concept that, while similar in content to the spate of WWII and Vietnam first person shooters (FPS), generates a greater feeling of horror. Is it because that we actually know the people involved? In a lot of these FPS games, we assume that the characters are nameless fictional characters in a fictional setting, or they’re monsters of historical proportions (e.g. Hitler) in an alternate history. Participating in the re-enactment of the events seems more macabre.

    That JFK Reloaded appears to be a straight-up simulation makes it seem even worse than if it was a twisted parody. On the other hand, from a forensic and training perspective, this is a fascinating product. The US Army already develops and distributes training simulation games in the form of America’s Army, and I expect we will see more 3D simulation of this kind of thing since the hardware capability has been part of the standard PC for at least five years. What is missing is broadly available talent to build convincing 3d worlds; I expect this to change in the next 3-4 years.

    Read more
    DSri Seah
  • Video Codec Shootout 2004

    January 2, 2005

    Via Slashdot: a comparison of hotshot video codecs on Doom9, which is, um, a site for enthusiasts of DVD backup. Regardless, they are in the position to know something about what codecs do well under what conditions, so if you’re bundling an animation up for full-frame, rock-solid playback, you would do well to skim the article and learn what works. The testing methodology is informative in itself, listing tools that you can download and use yourself.

    A poster on Slashdot notes that these tests are run on video source that had been compressed once before (these are DVD backups, so the sources have already been compressed with MPEG2, decompressed, and then recompressed). Still interesting to compare relative performance, I say, particularly if you’re doing homebrew video for work instead of with million-dollar master sources.

    Technologies I wasn’t aware of until now:

    • H.264/AVC, which is coming in the next release of Mac OS X “Tiger”, and is an open standard to boot
    • Wavelet-based video compression
    Read more
    DSri Seah