It seems like the more apparently deterministic the behavior of a system, the more reliable and robust we tend to consider it.
A shell utility and a lock: Both must respond predictably to inputs.
Is there a threshold of complexity above which this is no longer true?
For reasons not always entirely clear to me, I'm a sometime subscriber to The Economist. The subscription will lapse for a while, after I look at the renewal notice while thinking about one too many agonizingly tone-deaf photo captions or quasi-contrarian-reasonable-fiscal-conservative political stances, and then one day I'll be looking at the list of magazines you can get for your tiny heap of airline miles just before they lapse and I'll think well, this is all basically shit, but I guess I might as well get The Economist again, because they're sort of doing actual reporting and such, and unlike Standard US Market Fundamentalists, they aren't completely out of their brains.
I came home tonight to what I think is the most recent issue. It features The Thinker posed on a ceramic toilet with a large thought bubble which says
Will we ever invent anything this useful again?
and then down below and to the left, just where the toilet ought to be plumbed in to something, is a caption which says "The growing debate about dwindling innovation".
I don't particularly need to pick on The Economist. For all I know, there's a substantive article behind this skull-hurting cover. But the cover itself does strike me as a symptom of something: A kind of detachment from on-the-ground techno-social reality. I mean, it has been the God Damned Gosh Almighty No Shit Science Fiction Future since some time in the mid-1990s, a fact which has been obvious to any reasonably competent observer since, I'm going to say, 2007 at the absolute latest. There remains a lively debate to be had about which GDGANSSFF we're getting (I'm not holding my breath for something on the utopian end of the scale), but here we are.
Saturday morning, I rolled over and shut off the alarm on my phone and got on Twitter for a second to see what was up in the wider world. I saw that Quinn Norton had mentioned something about Aaron Swartz, and then I realized that what she’d said probably meant that he was dead.
Most of the web that I pay attention to just kind of exploded with it after that. This stream of grief and anger. I found myself standing in my kitchen crying over a guy I’d never met.
I never so much as exchanged an e-mail with Aaron Swartz. I maybe commented a time or two on his blog. There might have been other occasions, but the only time I can remember saying anything about his work in public was to write here, back in 2010, that something he’d written about how smart people see the world sounded like bullshit. (It probably was. Smart people are wrong plenty.)
He wasn’t a friend or a collaborator of mine. He was just one of those voices out there - a writer I followed as both of us were becoming adults, a hacker of some renown, a guy whose politics I often liked. I’m writing this entry in Markdown, a lightweight markup language he influenced considerably. He fought the good fight on machine-readability, which matters quite a bit to how I use the web. Like a lot of nerds, I spent a good chunk of the last 6 years on reddit, where I started to learn how to program in earnest. When he got busted for scraping articles from JSTOR from an MIT network connection, I was struck by the similarity to semi-serious plans I’d once made at UNL (hadn’t I started writing a little Perl robot one afternoon?), and by the brutal, sadistic absurdity of the Feds prosecuting anyone for such a benign act.
In the last few days I’ve noticed how important he was to a huge proportion of the people I respect in the technical & political world, the people who are voices out there and hackers of some renown. It turns out he was a friend and collaborator to a lot of the good ones.
Thornton Wilder, interviewed in the Winter 1956 Paris Review:
The problem of telling you about my past life as a writer is like that of imaginative narration itself; it lies in the effort to employ the past tense in such a way that it does not rob those events of their character of having occurred in freedom. A great deal of writing and talking about the past is unacceptable. It freezes the historical in a determinism. Today’s writer smugly passes his last judgment and confers on existing attitudes the lifeless aspect of plaster-cast statues in a museum. He recounts the past as though the characters knew what was going to happen next.