dsandler.org

And now, a plug for some seriously slick software by another ex-Be, Chris Liscio.

What a crazy 9 months it has been. FuzzMeasure Pro 2.0 is now ready for human consumption.

FuzzMeasure will perform frequency and phase analysis on your audio hardware so you can see exactly how sound is reproduced. It’s perfect for comparing speakers or headsets to see which one has the frequency response you’re looking for; professionals can use it to fine-tune custom cabinet designs. For actual data collection, FuzzMeasure works just as well with fancy impedance-measuring equipment as it does the built-in mic on your MacBook. The new version is the result of a lot of hard work streamlining the interface and tinkering deep mathy stuff I don’t understand; it’s the best (and cheapest, even if you buy the Pro license) tool out there for this kind of acoustinerdiness.

Dear blog owners, I’ve got some new software you should try:

1. Updated: The powerful TrackBack Validator plugin for WordPress has been revved to version 0.7. This plugin kills almost all existing TrackBack, dead. [I say “almost” because I (coincidentally) received word today of a spammer who sets up real blogs to try to spam people. I think of this as a victory for our plugin: It has forced spammers to, you know, behave like real bloggers. Who’s to say a TrackBack from this guy is spam and not a legitimate link to your blog?]
2. New: Illuminati is a new Internet measurement project from the CoralCDN (”Coral cache”) guys. They’re gathering much-needed statistics (like these awesome graphs) about the edges of the Internet (that’s you!), including NAT and proxy statistics, and they need help from website operators everywhere. Visit the site to learn how to contribute, or (if you’re a WordPress blogger) install my Illuminati plugin for WordPress.

This article (and ensuing thread) about lightweight Sparklines in Ruby reminded me that I’d intended to write a super-lightweight PNG library (for those times when you want a quick and dirty Web graphic and don’t have a proper graphics toolkit handy).

I don’t really speak Ruby yet (any more than I do Japanese…konban wa, watashi wa dsandler), so here’s minipng.py. Requires: Python 2.x, and nothing else. Currently supported: 8bpp RGB, with optional single-color transparency (tRNS). Sample code:

# should result in a black smiley face on yellow background
from minipng import MiniPNG, Color
mp = MiniPNG(9, 6, Color.YELLOW)
for pt in [(2,1), (6,1), (1,3), (7,3), (2,4), (3,4), (4,4), (5,4), (6,4)]:
    mp.plot(pt, Color.BLACK)
pngdata = mp.to_png()

I’ll probably add grayscale, gray+alpha, and RGBA soon (next time I’m bored and near a computer).

Hey, looky: the Buy Now saver got a mention on the SomethingAwful forums.

Independent hacker David Weekly, author of the currently-hot PBWiki (”start a new wiki, as easy as making a peanut butter sandwich”), offers impractical but deeply fulfilling advice: The Intellectual Property Wage Slave, or: Why You Should Quit Your Job. Here’s some more good advice: Get your smartest hacker friends together, and you’re bound to come up with something.

A couple of months ago (as noted here), Joel (he who is On Software) professed his love for Hungarian notation—variable naming with little bits of metadata baked right into the crust. Tasty.

Today he returns to old Magyarország with a new article: Making Wrong Code Look Wrong. In it he details the cargo-cult mentality of the generations who have perverted the original H-notation idea since its invention:

In Simonyi’s version of Hungarian notation, every variable was prefixed with a lower case tag that indicated the kind of thing that the variable contained.

rwMax

prefix

I’m using the word kind on purpose, there, because Simonyi mistakenly used the word type in his paper, and generations of programmers misunderstood what he meant.

I know I already delicious’d this, but it’s just so good. From Why Smart People Have Bad Ideas:

That’s the essence of a startup: having brilliant people do work that’s beneath them. Big companies try to hire the right person for the job. Startups win because they don’t– because they take people so smart that they would in a big company be doing “research,” and set them to work instead on problems of the most immediate and mundane sort. Think Einstein designing refrigerators.