Sunday, May 29, 2011

Engineers Are An Inferior Form of Life

Engineers lack the capacity to synthesize original ideas, which is one of the two pillars of Judgement. Which means engineers are amoral and incapable of discerning good from evil without a market survey. Which will of course return bogus answers.

That's the abstract explanation. Let us examine a concrete example. Let us look at how software engineers decided to handle data.

The History of Unix and Windows

Long ago, software engineers looked at data on hard disks and they asked themselves: what dimensionality is yon data, what is its nature? And they determined it was one dimensional (bytestreams) and they patted themselves on the back and saw that it was good.

Then software engineers looked at data on CRT monitors and they asked themselves: what dimensionality is yon presentation, what is its nature? And they determined it was TWO dimensional (bit-BLOCKS) and they scratched themselves on the head and they said "we have a problem".

But lo, a brave engineer came forth and said: we shall Reduce the dimensionality of the CRT so that it maps the one-dimensional data in two-dimensions. Look, it is simple, Cantor did this! And the engineers toiled for a day and a half and they named their creation the Command Line Interface and they patted themselves on the back and saw that it was good.

Then some evil-doers (called Lusers) made a Feature Request, and they asked: we want to embed this directory data inside of other directory data, can we have this? And the engineers saw that this Feature Request was Easy. So they toiled half a day and they named their creation (now with fractional dimensionality) the Filesystem and they patted themselves on the back and saw that it was good.

Then some evil-doers (incomprehensible unfathomable aliens called Artists) said: look, this is all well and good but I have these things called PHOTOS, what the fuck do Command Lines and Bytestreams have to do with my Photos? And the engineers now had a Real Problem.

But the Engineers' procedures had "worked" so well they had produced something 5% as usable as Symbolics' Lisp Machine or Smalltalk OS, so they said they might as well get on with it and produced Windows. And they patted themselves on the back and saw that other people were calling them delusional fuckers. And they were very surprised.

The History of Lisp Machines and Smalltalk

Needless to say, the approach of genuine systems researchers and designers to data was ... a bit different. For one thing, these weren't engineers. They were experimentalists. And they were either capable of original thought themselves, or if they weren't personally capable of creativity, they were at least capable of recognizing that genuine creativity was something to be cherished, not something to be scorned, crushed and dismissed by claiming it didn't fit a Market Need.

The Researchers' and Designers' approach to data was thus: okay so we've got an encoding of data, but we forgot about it because it's not even remotely relevant. We've got a presentation of data, that's almost relevant but mostly it's misleading so we'll actively put it out of our minds. What we need to do is figure out the NATURE of the DATA ITSELF.

For starters, what dimensionality does this data naturally exist in? Oh, it's K-dimensional fractal data where K varies arbitrarily, hmm that's interesting. Okay, so it looks like we'll need a couple of transformations to encode the data and a completely different family of transformations to present it.

And lo the Researchers approached some engineers and the engineers said: say what? I don't fucking get what you're talking about! What are these "objects" you're talking about? Why would anybody need this? What is this "idea space" you keep talking about?! This is absurd and inefficient! Only hardware exists. ONLY HARDWARE!

Then the researchers scratched their heads and said unto themselves: what we need is to buy some engineers and if they don't DO AS WE SAY then we will FIRE THEIR ASS. And lo this was done and through natural selection the Researchers and Designers finally got some engineers that had faith in their Word, and Lisp Machines and Smalltalk were both invented. And this was magnificent.

And to this day, still the Engineers maintain that Unix and Windows are "good" because they refuse to shut the fuck up and do as they're told!

10 comments:

Anonymous said...

For those of you who read the comments first, don't bother reading this blog post. It is unfortunately a waste.

j said...

Ah, the sugared feeling of you knowing THE ULTIMATE SOLUTION to general mankind problems. Can't beat it eh?

Anonymous said...

Pathetic, ankle-biting commenters.

Richard Kulisz said...

You haven't seen anything. On reddit, where this post was submitted, there was a commenter who said he only read the first half of this post and then bitterly complained that I never explained blah blah ... which I did in the second half of this post.

Anonymous said...

I bet you that half the reason that the engineers got confused about the requests for graphical user interfaces was that they couldn't figure out what the fuck you'd want to do with them. They didn't massively speed up any task, which had been the *very fucking reason* that computers were invented (as far back as Babbage as well!).

They weren't based on sound algorithmic advice, the likes of which dated a hell of a lot longer back than most of the designers' principles. They were a drain on resources, and no matter how much a word processing program or an image manipulator speeds things up now, it isn't by the orders of magnitude that computing speeded up the calculation of things *which were actually bloody useful*.

Also, Smalltalk sucks. Get over it. It eats memory and clock cycles with little - if any - net benefit.

Richard Kulisz said...

It's funny how engineers think speeding up some ridiculously unimportant niche applications in numerical computing by 1000x is "actually bloody useful" whereas speeding up the ENTIRE WORLD ECONOMY by a mere 2x is entirely unimportant.

Why do they think that? Because they've got objective empirical metrics right in front of them. And because they are too fucking retarded to wonder (let alone judge!) whether or not those metrics are any good.

As far as an engineer is concerned, if a metric measures things to 20 decimal places, then it's more important than one that measures things to 2 decimal places. Even if the former is the mean frequency of the sky, and the latter is the age of the universe.

Engineers are the kinds of fucking retards who take "precision is good" to extremes. So in their atrophied minds, the measure of a speedup of a task is GOOD because it's precise. And the measure of something's contribution to the economy is BAD because it's imprecise.

Of course, anyone sane and capable of common sense will merely conclude that since engineers are incapable of common sense then engineers must be retarded.

Thank you Mr Anonymous Coward of an Engineer for your brave illustration of how fucking retarded you are.

Anonymous said...
This comment has been removed by the author.
Anonymous said...
This comment has been removed by a blog administrator.
Richard Kulisz said...

This blog doesn't cater to insects. (indifference)

Anonymous said...

Who DOES it cater to?