Better Than Cats
Speaking of spell-check and porn servers, the paper that we were writing up on krypton background measurements has finally been sent off to a journal, and posted on the preprint server. I encourage everyone to check it out-- it's a real page-turner...
Posted at 11:16 AM | link | follow-ups | 5 comments
Everything I Need to Know I Learned from Spell-Check
The suggested replacement for "eroticism" in the previous post? "Heartaches."
That's either a seriously weird algorithm, or somebody has Issues.
Posted at 11:04 AM | link | follow-ups | 1 comment
Missed Opportunities
Via a passing mention at Grim Amusements, a moderately interesting article on gay rugby players. There's a really interesting article to be written here, or at least an article that would be really interesting to someone who played rugby. Even in the pinko-liberal Northeast college world where I learned the game, it never struck me as a particularly gay-friendly atmosphere, and the clubs I saw in the DC area were not an improvement. The interactions between a gay rugby team and straight clubs ought to be interesting, and it would be interesting to find out what effect (if any) this would have on the various non-athletic rituals surrounding the game. Do they sing the same obnoxious songs? Engage in the same sort of utterly-subtext-free-really-we-mean-it idiocies as the college teams I played for and against?
Sadly, the actual article only touches briefly on these things (in the last section), being written in the basic mode of a typical Salon article, where you end up learning more about the author than about the ostensible subject of the article. This can be done well, but it's a badly overused form, and I'm becoming thoroughly sick of it.
There's a painfully over-analytical tone to the whole thing as well: "Rugby is a blank slate for me: Similar to football in its outlines, it lacks the cultural baggage that comes with being the American sport. I can be tough without feeling like I'm part of a predetermined narrative about American manhood." But it's sort of silly to complain about this, given that it's published in the Village Voice, and that seems to be what they do.
About the most interesting thing in the article was the comment that rugby has seen a huge surge of popularity among gay men since 9/11, when an openly gay rugby player was among the people who rushed the hijackers of Flight 93. On some level, that comes across as desperately shallow, but that's probably more a reflection of my irritation with the form of the article than anything else. Likewise the thought that it's a good thing there wasn't a similar influx of new blood after this incident.
Another amusing tidbit was the statement that there are rugby-themed porn sites that "trade on the eroticism of the sport." Because, you know, wrapping black electrical tape around your head is just plain sexy...
Posted at 10:37 AM | link | follow-ups | 7 comments
Do My Homework For Me
Following on the previous post, I do have one idea for a useful way in which to hand out GMail invites as prizes.
This coming fall, I've agreed to give one lecture as a part of a class on Election 2004. This is a one-off class, with of order 100 students and twenty-odd faculty each giving one lecture on a topic in their general area of interest. I've offered to do "How to Lie With Statistics," which ought to be a rich subject given who the candidates are.
Given the nature of the class, and the size of the class, it seems like it ought to be possible to do some more immediate demonstration of the important ideas. I'm thinking of things like the difference between mean and median values, the improtance of sample size, the effect of selection bias, the difference between correlation and causation, and stuff like that. (Given that list, I obviously haven't written the lecture yet...). With a hundred-odd students, there ought to be a way to demonstrate some of this stuff using the class as a sample. We'll also be using Blackboard for the class web page, so I'll have the ability to ask questions ahead of time, and do some basic analysis of the answers.
The problem is, I don't know what questions to ask. So here's the challenge: What questions should I ask a group of 100 college students at an elite liberal arts college that would be likely to get me a data set that will show the difference between mean and median, and other such effects?
Posted at 11:29 AM | link | follow-ups | 7 comments
Meta-Contest
As a Blogger user, I automatically got a free GMail account. I bet all you people who crowed about switching to Movable Type a year ago feel pretty small now, don't you?
Anyway, the account comes with an apparently infinite number of invites that can be passed on to others, at least judging from the way Kate's been handing them out. Of course, I'm not as nice a person as Kate is, so I'm more inclined to make people dance for my amusement in order to claim them.
Problem is, I'm not as clever as John Scalzi, so I don't have any brilliant ideas for contests to run for GMail invites ("Tell me a funny physics joke that I haven't heard before" is roughly equivalent to "Chop down the tallest tree in the forest with a herring"...). But why should that stop me?
So here's the contest: Come up with something interesting that I should ask people to do in order to get a GMail invite. The best suggestion (or suggestions) gets a GMail invite.
Posted at 11:18 AM | link | follow-ups | 3 comments
Uninterested Readers Are Invited to Piss Off
I've got a bunch of research-type stuff that needs doing today, and I don't feel a really pressing need to blog about any news stories today. I'd be tempted to write about our enjoyable weekend in The City, but Kate already posted a trip report on her LiveJournal. The interested reader is encouraged to go over there and check it out.
Posted at 10:22 AM | link | follow-ups | no comments
Pretty Soon You're Talking About Real Money
the big technie news of the moment is the (mostly) successful sub-orbital test flight by "SpaceShip One" yesterday over the Mojave Desert. The Post article makes it sound like an unalloyed triumph, but the space.com coverage takes a more sober approach (somewhat surprisingly).
Of course, this will ineviatably be held up as a great example of the transformative power of capitalism and private enterprise, as the ship was built in part to go after the X-Prize of $10 million for a privately developed manned space vehicle. I have my doubts about whether this really proves any such thing, though, as the current effort has been funded by Paul Allen to the tune of $20 million or more. Spending $20 million to win $10 million is hardly a canny market maneuver, though it's about par for the course for Allen. This is more of a demonstration of the creative power of amazingly wealthy True Believers than anything else.
(Which is not to say that I hold Allen in contempt, or anything. I'd love to be him when I grow up-- he lucked into more money than he could ever possibly spend, and he's doing his damnedest to spend it all on cool stuff. I admire that in a billionaire.)
I really don't mean to demean their achievement, but the people who are really into this stuff have a gift for snarky public comments that really get my back up. See, for example, many of the comments to this post, or, for that matter, the comments of the mastermind behind yesterday's big flight:
But [Burt] Rutan and his team are the first to do so with no government help -- and not in a titanium machine powered by solid-fuel rockets, but a relatively lightweight craft. It was "a manned space program designed from scratch," Rutan boasted, in four years for $20 million -- about what NASA would spend "on a paper study," Rutan taunted.
Well, maybe. Of course, it's a little hard to say exactly how much NASA would spend on launching a single test pilot to a height well short of Earth orbit, because they're not in that business, and haven't been in that business since 1961...
If you want to try to make the comparison, though, here are some numbers: NASA's Mercury program cost something like $384 million (source) and launched 20 unmanned missions, two sub-orbital flights, and four orbital flights. Rutan's $20 million in modern dollars is roughly 3.2 million in 1960 dollars (source), so he's spent a bit less than 1% of the whole Mercury budget to come a bit short of Alan Shepard's altitutde in four years, rather than two.
I couldn't find any information on the cost of the sub-parts of the Mercury program, so it's hard to do an exact comparison, but it's probably not a big stretch to say that at least half of the Mercury budget was spent before Shepard went up. So on a strict cost-per-milestone level, Rutan is clearly ahead.
But that "from scratch" is a little hard to swallow. The Mercury program was really starting from scratch-- not only did they not have the launch vehicles, they didn't even know what to expect when they got into orbit. Consider this recollection from an older space.com piece:
Glenn, who flew again aboard shuttle Discovery in 1998, remembered a small eye chart taped to his Mercury console. Researchers expected his eyes to swell in microgravity and his vision to blur.
"I was supposed to look at that (chart), and if the letters got blurry, I was to do an emergency re-entry," he said. "We all laugh at that now, but it was a real concern."
That's what "starting from scratch" looks like. Whether he's using their rocket technology or not, Rutan is the beneficiary of close to fifty years of experience in sending humans into orbit, and is spared the costs of finding all that stuff out.
On some level, this is a petty gripe, and I agree that Rutan and his colleagues are to be congratulated for an impressive achievement. Still, the level of rocket-geek triumphalism surounding this flight is just a little hard to take.
Especially since I'm still not particularly clear on what the point of all this is. If space tourism at $10,000 a pop is the best rationale they've got for manned space flight, well... But that's a different rant.
Posted at 4:07 PM | link | follow-ups | [ hide comments ]
The launch itself was not exactly wild- the (admittedly cool-looking) vehicle took off, and sometime later there was a wickedly fast-moving vertical contrail, and then there was a landing...
The libertarians (and the Libertarians) were out in force, though. Ugh.
Nathan L., 06.22.2004, 7:25 pm [link]
More importantly, if Rutan can make a buck selling suborbital tickets, he can fund development of the next generation of vehicle, with increased capabilities. A few iterations of that process, and we'll have low cost access to orbit. That's the point of the tourist flights. Of course, making a profit along the way is also the point, and there's nothing wrong with that.
Most of the suborbital startups view tourism and remote sensing as intermediate business models to keep money flowing in while they are developing orbital vehicles. My own interest in spaceflight comes from a desire to mine asteroids for platinum group metals, but no business plan for extraterrestrial resource exploitation is viable with launch costs at >$5000/lb. Fortunately the ultimate floor for launch costs is more like $80/lb, based on fuel costs plus a reasonable operating margin. The only way to get there is to build lots of rockets, fly them many of times, learn how to operate them cheaply, and design for low cost operations based on that experience. Tourism is the most promising market at this point, and there is certainly considerable demand. It's worht noting that the $10k doesn't just buy the flight, it also buys a week or so of ground school, a ride in a vomit comet type aircraft, and some mementos - it's a whole package deal. At least one of the startups will let you actually fly the ship for the right price (obviously with considerable training beforehand).
Andrew Case, 06.22.2004, 8:01 pm [link]
It's true that NASA hasn't tried to send a person on a suborbital flight since the early 60s. But this is more of a problem than it seems, or at least is the indication of a deeper problem - NASA's spaceflight organization can't fly anything that's not the Space Shuttle, and they're having serious issues with that. Their program management is for shit.
This is not entirely NASA's fault. A lot of the program-funding decisions are made at the level of congressional appropriations, and there's not a lot they can do about that. NASA is one of those agencies whose budget is always on the block, and some of their management idiocies have been the result of the constant funding scramble.
Andrew Case:
The significance of this flight is that it's fully privately funded, and they have follow-on plans that lead to a profitable business. I'm a big fan of spaceflight, but NASA is clearly never going to deliver low cost space access. The standard NASA costing models applied to a vehicle like SS1 yields costs in the neighborhood of $150 million. I know three people who have done the cost numbers for X Prize class vehicles using NASA costing models and all of them come up with prices in the low 8 figures, so I think it's fair to say NASA is not the place to look for cheap access to space.
Um...
Twenty million dollars = $20,000,000. That's eight figures right there.
Is that a typo?
Most of the suborbital startups view tourism and remote sensing as intermediate business models to keep money flowing in while they are developing orbital vehicles.
Yes, but what for?
Manned space flight is cool, and all, but I'm just not really seeing a pressing need for it.
Cheap private unmanned launch capability is probably significantly more useful than manned orbital vehicles.
Chad Orzel, 06.22.2004, 9:18 pm [link]
Rich billionaires are working on that too. Not as much progress yet, but if SpaceX actually does what they want it's going to be very bad news for Orbital, and very good news for anyone who needs to launch things to orbit.
There are three arguments in favor of manned spaceflight, of varying degrees of convincingness. Manned vehicles tend to be much more reliable than unmanned vehicles - if you want to see something scary look at the crash rates for UAVs. If you're trying to do something difficult - say serious searching for past Martian life - it's cheaper to send people, because the latency from the eye to the brain is quite low, which is good for complicated decision cycles. Finally, people would like to go to space in a way that cargo doesn't.
And last, I think that it's closer to entirely NASA's fault than not NASA's fault. Even when they get stable funding profiles, projects are inevitably hugely over budget and behind schedule. You might argue that it's partly Congress's fault because when NASA says "We did all the interesting advanced research, and turning this into something useful will cost X billions more than we thought," Congress inevitably says "We can't afford that, here's some money for some more advanced research." I'd hope that NASA wouldn't deliberately sabotage things to take advantage of this, but I realize there's a bit of naivete there. I wonder if it's gotten to the point where all NASA can do is high-level concept studies...
Jake McGuire, 06.23.2004, 2:23 am [link]
pick up from Usenet, but...
One goal of the X-prize is just to demonstrate that a relatively
low-budget entrepreneur *can* operate in space. There's (I'm told) a
real effect where someone proposes a business plan which involves
putting stuff in orbit, and the investment sources say "Ok, let's work
out how much that'll cost," and they copy down the yearly expense
numbers of the Shuttle program. End of business plan.
Or (less absurd but still fatal) they look at current unmanned
commercial launchers, and say "Ok, you're going to buy one rocket per
launch, and N percent of them will fail. *N* percent? Isn't transport
supposed to have a loss rate of 0.0N percent?"
Simply demonstrating that NASA is not the only way to do things is a
big crack in the wall.
As for the unmanned versus manned thing, the point is that if you're
operating a business, you want a decent reliability level for your
*cargo*, never mind the driver. If 3% of the trucks on I-95 rolled and
exploded halfway through Delaware, the solution would *not* be
remote-control trucks. Even if that solved the business problem (which
it totally fails to do), it would itself work against reliability,
because a guy with a wrench can always solve some problems that are
intractable remotely.
Andrew Plotkin, 06.23.2004, 11:22 am [link]
pick up from Usenet, but...
One goal of the X-prize is just to demonstrate that a relatively
low-budget entrepreneur *can* operate in space. There's (I'm told) a
real effect where someone proposes a business plan which involves
putting stuff in orbit, and the investment sources say "Ok, let's work
out how much that'll cost," and they copy down the yearly expense
numbers of the Shuttle program. End of business plan.
Or (less absurd but still fatal) they look at current unmanned
commercial launchers, and say "Ok, you're going to buy one rocket per
launch, and N percent of them will fail. *N* percent? Isn't transport
supposed to have a loss rate of 0.0N percent?"
Simply demonstrating that NASA is not the only way to do things is a
big crack in the wall.
As for the unmanned versus manned thing, the point is that if you're
operating a business, you want a decent reliability level for your
*cargo*, never mind the driver. If 3% of the trucks on I-95 rolled and
exploded halfway through Delaware, the solution would *not* be
remote-control trucks. Even if that solved the business problem (which
it totally fails to do), it would itself work against reliability,
because a guy with a wrench can always solve some problems that are
intractable remotely.
Andrew Plotkin, 06.23.2004, 11:22 am [link]
Twenty million dollars = $20,000,000. That's eight figures
Yup, it's a typo. should be nine figures.
Yes, but what for?
Manned space flight is cool, and all, but I'm just not really seeing a pressing need for it.
There's no real pressing need, just as there's no pressing need for the NFL :-) The importance is lies entirely in the fact that it's a stepping stone to eventual permanent offworld colonies.
Cheap private unmanned launch capability is probably significantly more useful than manned orbital vehicles.
I was just talking about this with some friends, and one of them pointed out that there were four separate times during the SpaceShipOne test program where the ship would have been lost without intervention by the pilot. Unmanned is really hard, and expensive. As another example, the Hubble repair would not have been possible without humans present. Humans are really flexible and capable of tasks robots just can't do. Even really dangerous stuff like deep submergence repair work on oil rigs is still often done with humans, just due to the improved capabilities vs. robots.
Unmanned probes can do a lot, but you can't do really complex tasks with robots, at least not yet.
Andrew Case, 06.23.2004, 11:21 pm [link]
But that's part of the point, I suppose. I think the libertarian space enthusiasts tend to overstate their case outrageously (it sounds suspiciously like the scenario from countless 1940s science-fiction stories, in which the first space flight is funded by a hero entrepreneur and the main obstacle is the tyrannical government out to stop him). Space travel really is difficult; it's not going to magically become easy just because the power of private enterprise kicked in.
But I think they've got at least part of a point here: NASA could have built something like SpaceShipOne at any time, but they didn't; Rutan and Paul Allen did. That they'll come out in the red on winning the X Prize is no surprise; this is an experimental aircraft, not a production model. That comes later.
Also, I don't really see the point of denigrating expensive tourist flights as a motivation for manned spaceflight. As far as I can tell, the best reason for doing it-- really the only good reason-- is that people want to go. If at first only rich people can afford it, well, that's how a lot of things happen; it's better than nobody affording it. I think that it's a perfectly legitimate justification, as long as it's possible to make it financially self-sustaining (which, of course, remains to be seen). Obviously, massive government subsidies for money-losing rich-guy tourism would be a bad idea.
And the other justifications for manned spaceflight really haven't panned out. People can do things that robots can't, but the extra weight and expense required to keep the people alive wipes that out in a cost/benefit calculation. The Hubble telescope was repaired by astronauts, but it was designed from the beginning as a piece and justification of the shuttle program; for the cost of the shuttle program we could have had a bunch of spare Hubbles and just replaced the broken ones.
Matt McIrvin, 06.24.2004, 9:16 am [link]
As for the unmanned versus manned thing, the point is that if you're operating a business, you want a decent reliability level for your *cargo*, never mind the driver. If 3% of the trucks on I-95 rolled and exploded halfway through Delaware, the solution would *not* be remote-control trucks. Even if that solved the business problem (which it totally fails to do), it would itself work against reliability, because a guy with a wrench can always solve some problems that are intractable remotely.
Some of this is surely because this hasn't been a priority of the space program we've got. Research focussed on a low-cost and reliable launch vehicle would be a Good Thing, I agree. That's really what you want if you're looking to put satellites up, after all-- as Matt notes, the added weight and expense of keeping people alive in space tends to undercut whatever you save due to the flexibility of humans doing the maintenance.
Similarly, part of the reason we don't have the ability to do remote maintenance of satellites and the like is that the stuff we've been putting up is based on the assumption that there will be people there to do the maintenance. That's part of how the Shuttle program justifies itself, after all.
If you started from the presumption that all maintenance would be remote, you'd design things differently, and end up with something that could be maintained remotely without requiring constant human tweaking. You might end up with a system that does slightly fewer things than what you'd make under the current system, but you'd probably also end up with something that was more reliable on the whole.
I think there's a lesson to be learned from the shift in the emphasis of the unmanned probe program-- when NASA stopped focussing on the gazillion-dollar everything-plus-a-spare-kitchen-sink probes, and started doing smaller, simpler craft, they discovered very quickly that they can do some pretty amazing things with relatively simple robots. And they can make more of them for the same money, so the loss of one or two isn't as crippling.
Also, what Matt said, at least where he agreed with me.
Chad Orzel, 06.24.2004, 11:17 am [link]
Agreed that the SS1 was not a stunning technological achievement, libertarians dramatically overstate their case, and private enterprise is not a panacea.
But I really think that you're dramatically underestimating the dysfunctionality of NASA's spaceflight organization, as opposed to aeronautics or space exploration (satellites). Matt mentioned the X-15, but didn't point out (or realize) that NASA can't do the X-15 any more, either. They even tried (X-34, take 2) and failed spectacularly. The list of started, publicized, and faded from view programs is too large to go into here; the only ones that have been even marginally successful are ones that NASA inherited from the military, i.e. DC-X and X-40A.
I also think that you're underestimating the hardness of remote maintenance of satellites - after all, the Shuttle can't reach GEO or polar LEO, which combine for the vast majority of satellites by expense, but no one has figured out a way to service those satellites yet.
Jake McGuire, 06.24.2004, 12:51 pm [link]
That was the party line, but the actual lesson has been far more ambiguous than that. The problem is that for NASA, every loss is a public relations disaster, regardless of how cheap the probe was-- a million dollars still sounds like a lot of money to most people. Many of the simpler probes have been fantastic successes, but the "faster, better, cheaper" approach was widely blamed at the time for the losses of the Mars Polar Lander and Climate Orbiter. And it can be taken too far: the general consensus now seems to be that the European Beagle 2 Mars lander was just too lowball an effort to have a good chance of succeeding.
Meanwhile, great returns are starting to come back from Cassini (a remnant of the giant-probe days), and JIMO, if it happens, will be bigger and more elaborate than anything yet launched.
Matt McIrvin, 06.25.2004, 8:10 am [link]
COMMENTS ARE CLOSED.
Please visit Uncertain Principles' new location at ScienceBlogs to comment.
Look at recent history. X-33: $1.5B down the drain, pile of parts in a hangar to show for it. X-34: $300M down the drain, two airframes in storage, zero flight data. X-43: one successful flight to date, superseded by military scramjet research, follow-on cancelled. Alternate Access to Space: cancelled, and boy would it have been useful about now. X-38: cancelled, and likewise. Various rocket motor development programs: cancelled. X-37: getting to unpowered flight tests has taken three times as long as predicted; orbital flight tests have been getting farther away since the program started. The last NASA access-to-space program that can be called successful was the DC-XA back in 1996, and that was largely inherited from BMDO.
I will, however, join you in berating those who push space tourism at $10k a pop. I mean, I'd like to spend a week in some space hotel, and hope to be rich enough to pay $10k (or $20k) to do so. I'd also like to see us find out if there is or was life elsewhere in the universe, and sending people to Mars seems like the best bet for that. Either of these is going to require much cheaper access to space, and Rutan is the first person to actually fly something making progress toward that goal in quite some time.
Jake McGuire, 06.22.2004, 6:16 pm [link]