Some unusual suspect posted a few weeks back at (fabulous) OSnews a kind of a rant about why maybe we should mourn GeoCities. His claim is that, one way or the other, the mind state a creator has when he is pushing the raw matter around is different, more involved and in some ways more authentic than using an abstract (and arguably artificial) tool that splits the «content» from the piece itself. And, although i don’t buy it — i feel no loss for GeoCities, none whatsoever, despite having used it more than my fair share back in the day, or maybe one of the many copycats — i don’t buy it but i can’t shake the feeling that there is some… some… thing to this whole point.
At some unspecified level, i do feel too that there is some difference in the modus facendi, in the way your brain melts with whatever you are doing, and i do think that… Well, i don’t know. But i accept the possibility that maybe our present Content Management Systems do take something out of our hands while they provide us some facilities.
What really excites my curiosity is: whatever the means is, be it raw html, pen and paper, or maybe glue and scissors, or WordPress, whatever it is it will have an effect in what you create. And there is a big part of the whole creative business that is basically avoiding whatever biases your media brings.
Then again, even if i don’t love WordPress, it is not bad enough for me to go find something else. But.
But maybe our whole culture is at the midst of coming to grips with the whole computer language, that is, maybe we are having too long an adaptation period with the computer tools to create. Not only to write, but everything. Take soap-operas, for instance: lately we’ve been seeing some shows that do some heavy colour correction, something that, while possible with analogical media, has a lot more room for manoeuvre in the computer, so… I love it. I love the colour correcting, even though it is way unreal, but at the same time, i know it is too much. I just know that there had to be a guy somewhere in that production line to just say “less” at some point. The point obviously being not “less saturation” but “how do the new tools change our culture?” How do cheap GPUs change computers? And what is the difference — society-wide — of a computer and a computer+GPU?
Stated so broadly, all those questions amount to nothing. In those terms, the answer is “who cares”. I do think so. Then again, i think those questions, at some unspecified level, will become relevant. Pretty soon.
Actually, the overall impact of digital culture is very much upon us, but i do not think we have the right way of thinking about them yet…
We’ve had so-called «personal computers» for many years, and WordStar was already useful enough that you would not want to write without it after a few weeks getting used. The mythical Lotus 1-2-3 spreadsheet was seemingly useful enough that people would buy computers exclusively for it. And still, computers began appearing in film just after the new apple gave them a fashion spin. Up till last year most “graphical” operational systems where mostly bloated overcomplex monsters, but for some strange reason those systems were what it took to make people feel justified at calling their computers “friendly” (even though no, they do not understand anything at all about them and can’t really say they are at ease using but a few very specific programs). And probably the placebo effect is more important in improved functionality than any other HCI enhancements.
So: back to GeoCities?
Another recent thread at OSnews made the serious-to-some-willingly-dubious-point argument that MovieOS is the future of computation. MovieOS, in case you don’t know, is the computer you see on CSI. It works in unbelievable ways, almost always displaying gorgeous graphics for anything. And the idea is that maybe if computers did work that way we could free our minds of thinking about bits and bytes and instead focus on the tasks, which in turn would make us a better source of bits and bytes for the computer to process (that is, we would be more productive).
First, better visualization is just better. It should be easier to teach our computers to show spiffy graphics — and although there are plenty of toolkits for transitions and fades and morphs and whatnot, you do not see any toolkit that allows you to simply dump some data into a container and let it decide what is a sensible way to display it: Somehow things like that are seen as not useful, they are seen as too constraining.
And being a designer, i know inside-out this argument that a rational presentation can never be achieved by straight-forward rules. And it harks back to the mourning for GeoCities. Rule-based presentation is seen as constraining. This might be yet another outcropping of the old prejudice against rationality. But, as much as i can’t take the raw argument, i know this: writing in WordPress is more constrained than it was in PmWiki. It was actually one of the reasons i gave up on my only-me wiki: I was wasting too much effort with endless hand-tuning. The very choice that created this blog was in a way the choice between GeoCities freedom and M-V-C restraint.
But there is a wrinkle in this: PmWiki also managed my content. It was also a form of dealing with content that removed it from “raw matter”. It maybe was even more removed than WordPress, for in it i had to use (OK, i just loved it) the wiki-esque way of writing, with the codes and ascci-art-esque thinguies. But the feeling i had, when using it, was actually of having a direct relationship with the DATA. I felt like i was smearing the bits, not some random container, not some abstract XML.
So if there is some form of binary matter, and if there are ways you can get close to the machine, what are those ways? I think it all revolves around this: DATA is CODE. For the computer, there is no intrinsic difference between 2k of html, 2k of WordPress SQL and 2k of a.out. It is all current flowing in circuits.
So: when you write in WordPress, those keystrokes provoke the computer to align 0s and 1s in specific patterns, and those patterns constrain your computer into acting into certain ways. For that matter, it does the same to any computer that HTTPs into their server, or something to that effect, it is more complicated but not in ways that matter. So does the WikiCreole in PmWiki. What is the difference? Why did PostScript end up being a Programming Language instead of a Data Format?
At some prehistoric point in time, computers had to be set up to run in the closest you can possibly get to the machine without being a T-1000 yourself: By swapping cables around. Make no mistake: That sucked. A thousand years passed and you could actually program your computer by writing in binary: You wrote up long sequences of binary numbers that you knew by consulting a big table would have the results you wanted. Now at some point someone had the idea to put the computers onto the task of translating the tables and generating the binary numbers. And that idea, as long discussed as it has been, is still full of unseen consequences.
In a way, a compiler is a kind of codec: It takes binary data and spits binary data in a format we prefer. Whether we prefer this new format because it is shiny video on our spiffy LCDs or because it is optimized with the latest GCC flags is immaterial. Computers are good at data translation. When we started piping those translation capabilities into themselves a big Pandora box opened up. And new data formats also allowed new data entry mechanisms. Ergo mice and tablets.
But not only that, for example, the order you play your mp3 is also data. So is whether you prefer to open a browser and then an instant messenger when going online. Or if you accel your mouse more or less. All those are data formats, and whether or not we parse them does not make us “closer” or “farther” from the machine. So, for example, i disagree with Neolander when he says that iPads make computing less general-purpose. So-called virtual keyboards are a joke, really, but that does not mean they are a different kind of data. It is all binary!
Again: both WordPress and PmWiki have the same binary kind of matter, but one of them feels more direct. I think the difference is that one of them is more expressive — as a data-code hybrid.
I have long quibbled with some of my friends who call “raw” html (as in notepad and not DreamWeaver) code. It is not code. It is not a program. But the idea of expressiveness that we use to compare computer languages is kinda useful here.
Again, NAND is by itself a Turing complete language, but Turing completeness, it turns out, is not a very relevant way to judge a computer language. And it turns out almost all trendy new languages seem to include this dark-arts thing called “garbage collection” which basically means it is impossible to see how the program could render into binary form. You cloud your mind from this imagination, something that for example C stimulates you to see in your mind’s eye. Is this more expressive?
And maybe that is what i think is important in computer-culture. This is what i think is important: To rethink and reimagine the very idea of EXPRESSIVE.
I apologize for such a convoluted post, it was a form of soulsearching. I was searching for the idea while i wrote it.
So if neither C is expressive nor Ruby, what then? Haskell? I love Haskell, but it is still a little… greek? I guess probably WikiCreole goes further in that respect than either. There was this guy who wanted to write the genesis in Ruby, maybe we need more of those experiments…