Total Drek

Or, the thoughts of several frustrated intellectuals on Sociology, Gaming, Science, Politics, Science Fiction, Religion, and whatever the hell else strikes their fancy. There is absolutely no reason why you should read this blog. None. Seriously. Go hit your back button. It's up in the upper left-hand corner of your browser... it says "Back." Don't say we didn't warn you.

Wednesday, September 20, 2006

An open letter to SAS.

Dear SAS,

It's been a while since we've talked. How are you? Are you enjoying being a statistical package? How's your dad, Fortran? Is he still hanging in there? I know people have been predicting his death for some time now but, hell, he just keeps ticking, doesn't he?

Do you remember when we met, back when I began graduate school? I was so young then- not to mention innocent. I expected that in grad school I'd be using sophisticated modern quantitative analysis packages. Hell, at the time I didn't even think I'd be any good at statistics. Then I met you. It was an amazing meeting; my eyes locked onto your interface in the department computing laboratory and I knew- right then and there- that my life had changed. I could sense my dreams of elegant computing collapsing into a screaming pile of flaming debris with your very first words to me, "Syntax Error: Subscript out of bounds." Such magic. Some stats packages were ornery, some were difficult but you, SAS, you elevated difficult to the level of art.

What other package could produce such startlingly incoherent error messages? I can tell you, dear friend, that in years or working with statistics, I have yet to understand even a single one of your errors. Oh, I've figured them out, and solved the problems, but I'll be damned if I've ever actually UNDERSTOOD one of the damned things as it's written. It takes talent to accomplish that, SAS, and as always I am in awe of you. Then there's the actual nature of programming you, SAS: how baroque! How deliciously impossible! What other stats package would include a command so old, it actually refers to punchcard systems?* What other stats package would require a semi-colon after every line of code, if only to make it impossible to debug lengthy programs? What other software, SAS, would be so impossibly inconsistent as to only execute certain programs when they're called from a floppy disk? I must confess, I still appreciate that one. Indeed, SAS, my hat is off to you.

I didn't appreciate you at first, SAS. I thought you were clunky and painful to use- a veritable pestilence sent by a wrathful god to punish we mortals for daring to investigate creation. I cast you aside in favor of other newer and "better" stats packages- packages like Stata. Ah, Stata: I was smitten from the moment I received my first error. Not only was the message informative, it contained a link with which I could acquire more information! I even found that when Stata was unable to provide detailed error messages it generally APOLOGIZED for it! I cast caution to the wind and threw myself into Stata, seduced by a pretty (inter)face and an intelligible error code, forsaking your company.

But that was then, and this is now. I have finally been forced back into your cold embrace, SAS, by a difficulty that I simply cannot solve in Stata. Oh, I'm sure it's possible to do it in Stata but I just don't have the time to figure out how. Normally, this still wouldn't bring me back to you since solving a problem in Stata is about ten times faster than in SAS but I was seduced by the power of pre-written modules that promised a quick resolution.

How was I to know that they would be as insufferable as you? How was I to know that you would avenge yourself upon me by refusing to perform even basic commands correctly? I was stupid. I'm sorry. These past few months spent in the bowels of a laboratory working with you have been... special. My constant entreaties to you to process my data, followed by your resolute refusals, have been the stuff of legend.

Yet, all good things must come to an end. I have finally, it would seem, discovered the alchemical secret to extracting blood from this particular turnip. As we speak my commands are burrowing deep into your mind, like the weird worm thing from "The Wrath of Kahn," forcing you to produce my sweet, rich datasets. Soon I will be able to export this data into other programs that serve more willingly than you. Soon I will once more be able to abandon you to your miserable fate.

So, in conclusion, I hope you rot in hell you irrascible, uncooperative, pile of outmoded, inefficient horseshit. If you were a person I would gladly pay for the privilege or defecating on your face every night while you slept, solely for the joy of listening to your cries of disgusted anguish. I hope that during my lifetime mankind learns how to produce a version of SAS that is capable of actually screaming in torment, so that I may thank you properly for all of our good times. Eat shit and die you festering sore on the diseased ass of a post-modern "analyst."


Drek the Uninteresting

P.S: All my love to Lillian and the kids.

* Seriously, people, what the fuck did you think the CARDS command referred to?


Anonymous Anonymous said...

I was beginning to get a bit upset at your love song to SAS, given that you've never serenaded me so sweetly. But then I got to the end and I am forced to say thanks...I think. -Drek's SF

Wednesday, September 20, 2006 11:30:00 AM  
Blogger Tom Bozzo said...

I won't stand for good, logical FORTRAN being judged guilty by association with SAS.

Wednesday, September 20, 2006 5:18:00 PM  
Blogger Tom Volscho said...

That's funny Drek

Wednesday, September 20, 2006 8:14:00 PM  
Blogger cruffler said...

Speaking from the programmer side of computerland, I could offer an explanation for the semicolon thing but the floppy disk thing has me baffled. I do have several large magnets and if you would just direct me to where this abomination is stored...

Thursday, September 21, 2006 4:19:00 AM  
Anonymous Anonymous said...



Thursday, September 21, 2006 9:25:00 AM  
Blogger Tom Bozzo said...

On the semicolon front, statement delimiters can, theoretically, assist in debugging provided they aren't used to turn one's programming into something that looks like a strem-of-consciousness text.

The alternative, which Stata uses to some extent, of requiring that certain syntax elements occupy their own lines, may have some code legibility advantages. Its virtues otherwise depend on one's love of persnickety syntax rules.

Friday, September 22, 2006 9:02:00 AM  

Post a Comment

<< Home

Site Meter