Original humor & comedy articles daily weblog

They Queryin' ERRR'BODY Out Here!

(Yesterday was for Bugs & Cranks love, with the bittersweet story of Luis Salazar's former eyeball in The Cost of Doing Baseball.

And now, look -- look with your special eyes -- at what's going on today.)

I wouldn't have thought that writing technical documentation would be an 'angry' kind of job. Boring, maybe. Soul-sucking and nit-picky and dry, perhaps. But not so much angry.

Apparently, I was wrong.

This afternoon, I was sitting with an analyst in our group who was trying to pull information from a database we keep. He was using a statistical package called 'R'.

(It's called 'R' because statisticians are evidently intent on making the names of their tools as non-descriptive and web un-searchable as possible. I fully expect the next package to be named '4' or '/' or half a letter 'K'.

Maybe I should have recognized this 'angry' thing before, come of think of it. Remind me not to turn my back on any statisticians wielding sharp pencils in future.)

Anyway, R. I noticed our guy had a web page open to the online documentation, and sat down to have a look. Right at the top of the section was an example query. Every database package in the world has these examples; they're meant to illustrate how a real-life question would convert to the particular syntax used.

"We're already doing statistics, for crissakes; our lives are clearly awful enough. Throw us a bone over here."

So, for example, a lot of packages use examples that have to do with music. They assume that there are data tables with information on artists, CDs, songs and the like, and show how to ask the database pertinent and practical questions like:

"How many CDs were published in a given year?" or

"How many groupies were banged backstage during shows in the East Coast time zone?" or

"Who the fuck put Kenny G in here? Are you people shitting me? Kenny G? Really?"

(That last question has no good answer. This is why computers hate us from the bottoms of their cold heartless motherboards.)

But think about it. These examples could be about anything. They're pulled out of thin air, and whatever things you can think of with relationships of some type to other things -- which is, basically, everything -- then that can be your example. And the more familiar and close to home the better -- I've seen queries about movies and actors, beers and breweries, stores and products, zoos and animals, cars and drivers, breads and butters, Laurels and Hardies, locks stocks and barrels, and you get the idea. These can be literally anything.

So what was the query at the top of this R documentation page?

"select row_names, Murder from arrests where Rape > 30 order by Murder"

Wow. Just wow.

Of all the topics, all the queries, with the entire universe of things to choose from, they chose to answer that urgent pressing question on the tips of all of our tongues:

"Yeah, how many places have there been at least thirty rapes, sorted in descending order of murder rate, pretty please with sugar on top?"

I'm not making this up. It's out there in black and white. But honestly -- wow.

I mean, first of all, it's pretty grim stuff. Short of penning examples based on genocide or suicide bombers or something, I can't think of a darker topic for this kind of thing. We're already doing statistics, for crissakes; our lives are clearly awful enough. Throw us a bone over here.

Secondly -- and maybe this is just me -- but I have a policy that I never want to be reading any sort of technical documentation and type CTRL-F 'rape' and find any results.

(It's a new policy, which I set immediately after reading the query above. And I've never actually searched for 'rape' in any documentation I've read in the past.

But I'm sure as hell going to do it now. That shit is crazy.)

And finally -- greater than? With everything else wrong with this example, they couldn't at least look for a spot with less than thirty rapes? Come, now. That's just piling on.

So that's what I learned today. There are some really angry, disturbed and possibly currently incarcerated people writing documentation for statistical packages.

But at least they can teach you how to query a database. Which probably won't help much when you're violated and buried in the crawlspace under their basement stairs. But it's something, I guess.