Original Cin

The always witty, opinionated, and intelligent LummoxJR has announced that he will once again be critiquing ads and recognizing the worst of the worst. I have fond memories of the original Ad-Schlocks, and look forward to this.

Back in college, I knew a person named Kyle (I think) who was absolutely crazy about history. Lies My Teacher Told Me was his favorite book. He thought that history was something everyone should know forward and backwards. I think he knew the exact year that Caesar crossed the Rubicon. When I asked him why history was so important, his answer came shockingly close to the aphorism that “people who don’t know history are doomed to repeat it”.*

There is a problem with this reason to study history – it is a falsifiable statement. If we accept that a knowledge of past events allows one to draw parallels with present situations and make better decisions as a result, then we should observe that historians are better decision-makers and make better predictions about the future than the rest of us. A quick search of Google Scholar reveals that someone has, in fact, done exactly that, although no full articles appear to be available online. I intend to look further into this, and see what people have actually found.

*I suspect that this saying is to history as Keynes’s quote about people being “…the slaves of some defunct economist” is to economics. It sounds really neat, but it’s actually pretty dumb.

Spelled it right this time, and did it on purpose as well:

I like Splenda in my tea, but I don’t like what the company behind Spenda seems to be up to lately. Not that it’s evil, just stupid. Found via the blog of Seth Godin is a story about Spenda’s marketing buying up negative domain names related to the sweetener.

Godin sarcastically asks “Is there enough money in the world to buy enough domain names to keep a determined person from saying something nasty about Splenda?” For those of you not familiar with how domains work, the answer is an obvious no – Bill Gates and Warren Buffet working together, liquidating all resources at their disposal, could not accomplish such a feat.

Actually, there are all sort of negative Splenda domains still available. Consider the following, taken from a registrar as of this writing:

badsplenda.com AVAILABLE
badsplenda.net AVAILABLE
badsplenda.org AVAILABLE
splendablows.com AVAILABLE
splendablows.net AVAILABLE
splendablows.org AVAILABLE
splendaisevil.com AVAILABLE
splendaisevil.net AVAILABLE
splendaisevil.org AVAILABLE
saynotospenda.com AVAILABLE
saynotospenda.net AVAILABLE
saynotospenda.org AVAILABLE
deathbysplenda.com Make Offer
deathbysplenda.net AVAILABLE
deathbysplenda.org AVAILABLE
splendacancer.com AVAILABLE
splendacancer.net AVAILABLE
splendacancer.org AVAILABLE
splendasick.com AVAILABLE
splendasick.net AVAILABLE
splendasick.org AVAILABLE
waronsplenda.com AVAILABLE
waronsplenda.net AVAILABLE
waronsplenda.org AVAILABLE
splendainformation.com AVAILABLE
splendainformation.net AVAILABLE
splendainformation.org AVAILABLE
splenda-addicts.com AVAILABLE
splenda-addicts.net AVAILABLE
splenda-addicts.org AVAILABLE
splendaboycott.com AVAILABLE
splendaboycott.net AVAILABLE
splendaboycott.org AVAILABLE
boycottsplenda.com AVAILABLE
boycottsplenda.net AVAILABLE
boycottsplenda.org AVAILABLE
splendacensors.com AVAILABLE
splendacensors.net AVAILABLE
splendacensors.org AVAILABLE
splendabastards.com AVAILABLE
splendabastards.net AVAILABLE
splendabastards.org AVAILABLE

All those domain names that I came up with off the top of my head, and only one of them is taken! Those Splenda people didn’t do a very thorough job at all. Tsk, tsk. You didn’t even think of blows as a synonym for sucks. I’m sure Urban Dictionary has some good ones too (Oh, splendawack.com is also available). The examples may be taken by someone shortly, but the search space for negative domain names for Splenda is so large that finding them all before someone else does is a futile task.

Disregarding all that about domain names, there’s another problem for people trying to squash negative speech about this product. Already, their actions have been interpreted as an implicit admission of guilt, as seen in the Sustainable is Good link. I’ll admit, first and foremost, that I don’t know how ideas die or how to kill them, except that it’s very, very tough. Shouting the opposite idea from the rooftops, however, rarely works. Implicit in each and every single domain registration no matter where it may be redirected to is the idea that sucralose is a dangerous product, and that there is a conspiracy to make Splenda look good.

My car just stalled and died today, just as I was trying to leave. This sucks.

I decided to give the Slax “Kill Bill” Edition Live CD for a spin on my laptop the other night.

I am impressed with how far Desktop Linux has come. At precisely the point in history when an open platform is required for the future of computing, it seems that two decent choices are available: the partly proprietary but still nice Mac OS X, and Linux.

First of all, the KDE interface actually looks decent now. Not great, but decent. Much better than any shell I’ve seen from Microsoft or otherwise for Windows XP. Most of the confusion I’ve seen with various X Windows, Window Managers, and KDE settings being redundant or incompatible with one another in the past have been ironed out. The system doesn’t automatically recognize that I have a “Wide-format” screen on boot-up, so things look stretched, but fixing this is just a matter of right clicking an icon and selecting the right setting. The boot time is atrocious (I’d guess slightly more than 5 minutes), but this has nothing to with the fact that it’s Linux/KDE, and everything to do with the fact that it’s running from a CD.

One thing I’ve never managed to get right with Linux is getting onto the internet, and it seems I’m still having trouble with the Wifi card built into the laptop. I’ll post an update later.

As I mentioned in my last post, I’m trying to extract Wikipedia articles from the latest database dump into a set of text files. It’s slow going, with roughly 30000 of the 1.6 million or so article being processed so far.

Truth be told, I’m only interested in maybe a few thousand of those for my purposes (creating a partial mirror), but the foundation doesn’t like people live-loading stuff from Special:Export, so I guess I’ll grab everything instead.

As a part of the script that’s processing the XML, the title of every 1000th article is printed out; here’s what I’ve gotten so far:

Saving article Alexander+VII
Saving article Plum+pudding+model
Saving article BCPL
Saving article Computer%2FTime-sharing
Saving article Carbine
Saving article D.+W.+Griffith
Saving article EpiphenomenalIsm
Saving article Emperor+Kenz%3F
Saving article List+of+freshwater+aquarium+plant+species
Saving article Goa+%28state%29
Saving article Hermitian
Saving article International+phonetic+alphabet
Saving article Kit%E1b-i-Aqdas
Saving article Lepus+%28constellation%29
Saving article Marco+Polo
Saving article Mimas
Saving article Naturalistic+fallacy
Saving article Wheel+%28poker%29
Saving article Password
Saving article R
Saving article Spice
Saving article SAC
Saving article Scrooge+McDuck
Saving article Testimonium+Flavianum
Saving article Universal+Declaration+of+Human+Rights%2FText
Saving article White+Fish
Saving article 1840
Saving article 554
Saving article Ireland%2C+Republic+of
Saving article Tristan+und+Isolde
Saving article 1376
Saving article B3ZS
Saving article Wafer+%28electronics%29
Saving article Sea+gooseberries
Saving article Notes+on+Muscovite+Affairs
Saving article Augusta%2C+Maine
Saving article Fishing+line
Saving article Quantum+immortality
Saving article Century+21+Exposition
Saving article Burali-Forti+paradox
Saving article South+West+England
Saving article Dutch+euro+coins
Saving article 810s+BC
Saving article West+Bank%2FMilitary
Saving article Petropavlovsk-Kamchatsky
Saving article Judas
Saving article Samuel+Delaney
Saving article Hilberts+basis+theorem
Saving article Fabaceae
Saving article National+League
Saving article D%FCsseldorf
Saving article New+Historians
Saving article Brain+aneurysm
Saving article Turkey%2FTransportation
Saving article Algorithmic+complexity+theory
Saving article Baby+Gramps
Saving article Alice+Comedies
Saving article Scrooge+McDuck+universe
Saving article Transnational+issues+of+Israel
Saving article Rational+Expectations
Saving article Atharva-Veda
Saving article Clymenus
Saving article The+Nature+of+the+Firm
Saving article St.+Petersburg%2C+Florida
Saving article Kazimierz+Pu%3Faski
Saving article %C4gir
Saving article I%F0unn
Saving article Education+reform+in+occupied+Japan
Saving article Grant+County%2C+West+Virginia
Saving article Beaver+County%2C+Oklahoma
Saving article Warren+County%2C+New+Jersey
Saving article Washington+County%2C+Mississippi
Saving article GNU+Maxima
Saving article Amber%2C+India
Saving article An+%28god%29
Saving article Sargon+II
Saving article Salhouse
Saving article Thatching
Saving article Kranjska+Gora
Saving article Eubie+Blake
Saving article Tara+Hills%2C+California
Saving article La+Jara%2C+Colorado
Saving article Golden+Beach%2C+Florida
Saving article Abbeville%2C+Georgia
Saving article Bannockburn%2C+Illinois
Saving article Clermont%2C+Marion%2C+Indiana
Saving article Stereolab
Saving article Westwood%2C+Kansas
Saving article Mer+Rouge%2C+Louisiana
Saving article Alford%2C+Massachusetts
Saving article Belding%2C+Michigan
Saving article Heeze-Leende

Those odd characters are popping up because I’ve chosen to URL encode the title, but they should be readable for the most part. What strikes me about this list is how dull and boring it actually seems. You have lots of locations, biographies, a bit of software, at least two articles about Scrooge McDuck, and all sort of mundane things that people have heard of before. Yeah, there are some esoteric and interesting subjects in there, but they’re far outweighed by the dull.

I’m currently trying to work with the latest database dump from the English Wikipedia. It’s massive (Slightly under 10 GB uncompressed), and a pain to work with – especially since some of the behavior of PHP file functions with large files is not quite right. So, what I’ve been trying to do is break the XML dump down into sections (I’m losing a small handful of articles this way, less then 10) and then process those chunks into text files, which are then stored in a 3 level directory tree by letters: The “Disgaea” article would be stored as /home/myusername/wiki/d/i/s/disgaea.txt)

In order to create these directories, I used the following function:

function make_directory_tree ($level = 0, $parent = '', $maxlevel = 3) {
global $CONFIG;

if ($level < $maxlevel) {
make_directory_tree($level + 1,$parent . $letter.'/');
}
}
}

Truthfully, there are better ways to do this than a recursive function, but I didn't think there'd be a big difference in performance, so I was surprised by how long PHP was taking to create these directories.

And, when it was all said and done, I had the nasty surprise of finding that there were four levels of directories, rather than three - I had forgotten that the foreach loop meant that the final directory layer wasn't created recursively. Oops. I wanted 19683 directories, and accidentally created 531441. Oh well, set $maxlevel = 2, and try again, I suppose.

What is 666 in base 6? 3030! Proof that the Anti-Christ will appear in exactly 1023 years.
</prediction type=”not serious”>

What got me going on this is that I was thinking of how API keys are generated for online services. Ideally, you’d have an API key that’s short, but not too short, and compact but still portable and human-readable. Expressing the key in base 36, perhaps with a few additional symbols thrown in as “punctuation”, seems like the best balance here.

Now, suppose you wanted to group these keys by series; say, by year. What’s 2007 in base 36? I went to get the answer, but accidentally typed “32″ instead, which reveals that 2007 is year 1UN in base 32 (It’s 1JR in base 36, but 1RR in base 33 – maybe we’ll see a resurgence in the popularity of railroads before December rolls around?).

Oh, and prepare yourselves: the year 2047 is 11111111111 in binary.

After installing WordPress here on the previously redirected nic.dreamhost.com, I had been considering trying to use it to run the blog at http://fancruft.com, which is really just an ugly way of displaying some entries from my Lavos MySQL database.

I suppose I’m glad I didn’t even have a chance to start. Today, just about everyone with WordPress 2.1.1 was scrambling to upgrade it after it was revealed that intentionally malicious code had been placed in it. I’d encourage people to read the original announcement.

I like the WordPress front-end. It shows that someone understands how people want to interact with software. But, after a problem this sever in the underlying PHP, how can you take Automattic’s (not so) subtle dig at PHPBB: “Have you ever been frustrated with forum or bulletin board software that was slow, bloated, and always got your server hacked?”

Why, yes, I think I’d like to NOT have my sites owned by the mob’s spam division very much, thank you.

I recognize that the fact that this was discovered in days, rather than months, is a testament to the “many eyes” theory of Open Source, but an intentional backdoor placed in software by a third party is about as bad as it gets.

As time goes by, I suspect more and more that the Republicans may be in serious trouble for the next Presidential election (in 2008). The Democrats have lined up several interesting personalities, all of whom are more politically savvy and marketable than anyone we’ve seen before in my lifetime. I also suspect that the George Bush Presidency has been an exercise in misapplied logic – it would have been the best damn Presidency ever in a world of Vulcans, and here on earth, it sucked, because it assumed short-term rationality on the part of people who were either completely irrational or willing to forego short-term rationality for some long-term goal.

Just to be clear on my own politics, I generally prefer Republicans to Democrats for office, but have always either considered myself a Libertarian or independent. I’ve found my enthusiasm for the Republican Party to be waning in direct proportion to their increasing abandonment of fiscal issues. I sense that the Republicans are too greatly abusing security theater to the detriment of taxpayers and citizens. True, the Democrats aren’t any better in these respects, but the Republican party has no advantages for me in other respects. It’s become of the party of the bland. All of this has shown up in the party’s real ability to get elected; right now, Hilliary Clinton alone has a 30% chance of being the next President based on the Foresight Exchange claim HRC08. I haven’t looked at claims for the other candidates, but, naively, this seem to indicate to me that the Republicans have only 35% chance of winning the election. With a few percentage points moved around, Clinton could mop the floor with all Republican challengers combined. (Why a prediction market instead of a poll? In short, you get slightly better results asking people what they think will happen rather than what they want to happen; see The Wisdom of Crowds for a good explanation, although I was into prediction markets before prediction markets were [sort-of] cool).