20 July 2008, 6:16 UTCLink: Charlie's Diary: Magical thinking
Charlie's Diary: Magical thinking … the belief that because doing something about climate change (and environmental degradation and peak oil and the whole dismal litany) is better than doing nothing, any particular something they can point to clearly must be done, however irrelevant it might be to dealing with the underlying problem. It generates make-work, an annoying wheel-spinning tail-chasing pursuit of distractions, at the cost of grappling with the very real and very serious problems we face.
As someone says in the article he links to, "I don't really mind too much what your plan is, but it's got to add up." If you have no numbers of your own and a couple of calculations on the back of a napkin are enough to show that your plan is either not remotely feasible or insignificant, you're not helping your cause.
(Helpful link: the units program can help with a lot of calculations by keeping track of the units of measurement for you. Use the -v option for clarity. If you're on a legacy system that doesn't come with it, use an on-line version.)
[permalink] ‣ 2 comments ‣ keyword: link
18 July 2008, 12:52 UTCThe id of the Net
If the Internet is a group mind, the id resides at 4chan.
(With thanks to T for discussion.)
27 May 2008, 15:35 UTCDeath Freedom Day
Nothing is certain but death and taxes.
Analogously to Tax Freedom Day, could one also calculate a "Death Freedom Day"?
It would be based on the relative speed of medical research; currently, I believe that's about 10% — our remaining life expectancy is pushed back by about a month for every year we live, so that net aging is 11 months of the year.
The twelfth month is free...
In the US, Death Freedom Day would be round about Thanksgiving.
(Actuarial escape velocity is achieved when Death Freedom Day falls before the 1st of January. That would entail an order of magnitude improvement of medical research output; so close yet so far away.)
[permalink] ‣ keyword: immortal
17 May 2008, 15:50 UTCFrozen Evolution, now in English
For those of you I've ranted at rather inexpertly about "frozen evolution" or "frozen plasticity", the book is now available in English.
(Disclosure: Jaroslav is a 2nd cousin of mine.)
[permalink] ‣ keyword: plug
11 May 2008, 15:11 UTCIn defence of "off the scale"
An oft-maligned trope of TV and film science fiction is when some sort of measurement is off the scale. It seems to me, though, that this is somewhat unfair. After all, by convention, what we see on the screen are the exceptional circumstances. The sensors and other measuring devices are designed for normal circumstances. If they go off the scale when confronted with a demigod, say, it's only to be expected.
The characters' attitudes to the situation should perhaps be portrayed as a bit more negative; more a sense of annoyance than awe. After all, a reading off the scale ("off-scale high" is the usual term, or "off-scale low" if it's the other way) basically means that (a) we have no idea what we're facing, other than that it's big (small) and (b) depending on the sensor, it will probably need a calibration check, possibly repair or replacement. It certainly can't be relied upon until at least a rudimentary check is made that it still works.
Even in everyday life, off-scale readings are not that uncommon. When you get in your car, most of the gauges will show off-scale low until you turn on the ignition, while the engine temperature will stay off-scale low for some minutes after starting. All measuring devices have a certain range, if you exceed that, they no longer work. Sometimes permanently. Being deafened by noise or blinded by light is the same thing; we don't even need gadgets here...
It's also important to record off-scale readings as off-scale, rather than picking the nearest number or something. After all, you have no idea how far off-scale you might be, how inaccurate. Even in our space program so far we've had a case of that; an off-the-scale temperature during ground testing was recorded as 100°F, because that was the end of the scale. That was a little high (it was supposed to be 80°F), but not too bad. In reality, the temperature was some 800-1000°F, perhaps more, severely damaging the unit. One of several links in the chain leading to the Apollo 13 explosion.
11 May 2008, 3:41 UTCSelling the invisible
Good design is invisible. See for instance this Bruce Sterling video. When using a GPS navigator, you don't want good instructions per se, you want "I didn't have trouble finding the place". When using a word processor, you don't want "Clippy really helped me", you want "I had no problem writing my letter". Writting the letter may or may not have been helped by Clippy popping up and suggesting things, but it should've been unobtrusive, forgettable that he did so. When wearing a suit of clothing, you don't want people to go "nice outfit", you want them to go "handsome fellow". Good design for all these things means that they are essentially invisible.
But — how do you sell invisible?
You lie about it is one option, observed by T. You pretend that visible features are the important ones, maybe even have a demo mode that attracts attention in a way the real usage mode does not. It's not a good option, but all things considered it could be worse.
Design badly is worse. Design things that, to quote Bruce's phrase, "hang around the user's neck". I've seen GPS navigators that insist on defaulting to an information-rich, too-distracting screen that's practically unusable without the even more distracting voice instructions. We've all seen how Clippy turned out.
Even moderately well-designed things have their weaknesses. Auto-ranging is one idea that's common but often unfortunate; it interferes with quick reading of rough value, which depends not on the large, easy-to-read digits but on the tiny "V" or "mV" in the corner. Fortunately, in the case of my GPS navigator, it only has two ranges, which are easily enough distinguished (12.3km or 100m). My multimeter, while practically unusable that way, has a button to select a range manually. My GPS navigator will default to the silly screen, but it's easy to switch to the useful-for-driving screen — and it will stay there.
I wonder if there is a better solution.
6 May 2008, 3:34 UTClolinstructions
Non-rechargeable batteries are
not being recharged. ―packaging of electro⌁dice
15 April 2008, 14:34 UTCLimits of invention
Inventions have long since reached their limit, and I see no hope for further developments.―Julius Sextus Frontinus, Roman Engineer, 10 A.D.
We usually quote him in order to giggle at him; but there's one interesting circumstance: he was right. In the following millenium, what significant inventions have the Romans made? Very few, if any. There were minor inventions and incremental improvements, but real progress had to wait until the Empire fell.
The Romans were not set up to do pure research. They were good at engineering and applied research, especially applied to making war. Pure research, not so much. When they came across its fruits, they took them and applied them; but they would never advance it. For a millenium, they used the same math textbook, with all its faults major and minor, never advancing the field.
After a century or so with no pure research, applied research starts going in circles. (And after a decade or so with no applied research, engineering starts going in circles.) There was no domestic pure research, and by the time of Julius Sextus Frontinus, the limits of the expansion of the empire have pretty much been reached. Certainly Greece had been conquered more than a century prior. If he felt that they were starting to go in circles invention-wise, even if he didn't know the cause, he might well have called it.
31 March 2008, 15:38 UTCOf vampires and men
Where there are vampires, cloves make the man.
29 March 2008, 16:18 UTCThrust-forward vs thrust-up
When designing a spaceship, various frames of reference are possible, but two are common: the main thrust direction can either be labelled "up" or "forward".
NASA usually uses thrust-forward [1], but the LEM was thrust-up. Russian craft always seem to have been thrust-up. I have no idea about Shenzhou. In fiction, Lost in Space was thrust-up while Star Trek and Star Wars are generally thrust-forward. It doesn't make much difference, for the most part, but it does make some. For one thing, seats seem to end up raked slightly differently.
For larger vessels (fictional or planned), it will probably make more difference: thrust-up is probably better, because when thrust is activated things will fall in the "down" direction; it's easier to reason correctly about that. With thrust-forward, things will fall "aft" — counter-intuitive, possibly dangerously so. For instance, a passage parallel to the thrust vector will be labelled "up-down" in thrust-up orientation, obviously wrong or at least in need of safety arrangements (railings). In a thrust-forward orientation, it'll be labelled "forward-aft".
[1] NASA-STD-3000 I 8.5.3.2 Directional Designation Design Requirements
[permalink] ‣ keyword: space
27 March 2008, 18:15 UTCElectrolytic Underwater Breathing Apparatus
26 March 2008, 11:58 UTCHollywood death-wish
25 March 2008, 15:00 UTCAbsolute position
25 March 2008, 13:44 UTCAcronym lunch
18 March 2008, 16:41 UTCFormula editor frustration
29 February 2008, 16:49 UTCNetwork Neutrality and automatic telephone exchanges
29 February 2008, 14:30 UTCBuckywell fusion
27 February 2008, 2:43 UTCIntransitive fabber nozzles
12 February 2008, 13:52 UTCThe Age: The music industry is dying
12 February 2008, 2:32 UTCAnonymous analyst criticises anonymity
11 February 2008, 15:42 UTCScientology and the Net; or, the more things stay the same, the more they change
16 January 2008, 2:59 UTCBiometric aphorism
10 January 2008, 9:05 UTClinux.conf.au Open Day — 2 Feb 2008
15 December 2007, 7:20 UTCGoogle Earth and cache-filling
20 November 2007, 7:12 UTCLink: Larry Lessig's TED talk
14 November 2007, 15:10 UTClinux.conf.au earlybird registrations closing
14 November 2007, 12:27 UTCGrand Sale… Grand Piano, that is
4 November 2007, 5:17 UTCMacOS vs Linux: a difference of style
31 October 2007, 14:03 UTCLink: That Sneaky Exponential
29 October 2007, 11:58 UTCBoat Ramps Notepad wiki
