It sucks to be a predator
And you’ve got people. Billions of people walking around like Happy Meals with legs. —Spike
Consider the vampire as a creature bound by the Law of Conservation of Energy.
According to the MadSci Network (how can I not trust these guys?), there are about 700 [kilo]calories in a liter of human blood. A vampire who sucks dry a human adult with five liters of blood is going to get 3,500 calories. One who believes in sustainable humaniculture, drawing at a rate no more than what the Red Cross recommends for whole-blood donors, will get half a liter, i.e., 350 calories, every other month.
So if the vampire needs 3,500 calories a day to sustain his or her undeath (more than the average human adult, but it makes the math easier, and heck, a vampire doesn’t spend all night sitting at a desk), he or she will need to either rotate among six hundred cooperative hosts, or take down one victim every day. In the latter case, even if every victim is consumed after he or she reproduces, and even if vampirism is the only cause of human death, we would need a population of at least ten thousand humans to carry each vampire. By comparison, in the classic study of population dynamics among moose and wolves, the moose-to-wolf ratio ranges from 15–50 moose per wolf.
Moose… hmm. What if we are dealing with emo-pires who refuse, on principle, to feed off another sentient species? It says here that “a 400 kilogram moose has a blood volume of about 32 litres”, so one moose could replace between five and six human hosts; a moose generation is only four or five years, so a stable population of ten thousand humans could be replaced by about a hundred and fifty moose—assuming, crucially, that the vampires would be just as successful at catching the moose as they would be at catching two-legged prey. So Maine’s population of 30,000 moose (according to Wikipedia) could support up to two hundred vampires, while its human population could support only a hundred and thirty.
Regardless of which scenario you choose, given how much impact a single additional vampire has on the food supply, it’s hard to see why any immortal vampire would deliberately turn a human.
If vampire-story authors would pay attention to these questions of population dynamics, they could enrich the genre.
Wanted: digital passports
Teresa Nielsen Hayden, bemoaning how CNN requires you to sign away your digital soul in order to post a comment on its Web site, remarks:
In the United States, we don’t have many laws protecting our personal information. We need more than we have. The alternative, the one we’ll get by default if we don’t do anything, is to have our online identities mediated by Facebook. If the government had proposed an online identity system that prone to holes, leaks, and exploits, we’d have been up in arms.
Which reminds me…
In the Anglo-American legal tradition we’ve always been antsy about government officials saying “papers, please”, but authentication of identity is one of the traditional functions of the state. The paper trail certifying that I really am Seth Gordon, for all transactions where it really counts, terminates in two places: a birth certificate issued by the State of Illinois, and a driver’s license issued by the Massachusetts Registry of Motor Vehicles. (If I had changed my name since birth, of course, that paper trail would have taken a detour through some probate court.)
The governments of Estonia and Lithuania have brought state-authenticated identity into the digital age with systems that combine a smartcard, a PIN, and the OpenID protocol. (My Web browser cannot authenticate the Web site for Lithuania’s national OpenID provider, which suggests there are a few kinks to be worked out of the system.) Why can’t the United States do the same thing?
For financial and medical transactions, this would create a single strong system for logging into the Web sites of multiple banks, credit-card issuers, and so forth. For non-financial transactions, a government-backed identity broker could authenticate a user by revealing the minimum amount of information that a Web site operator actually needs, rather than the maximum amount that some profit-seeking broker wants to share. The government could protect citizens’ privacy by offering them proxy identities: “account 2b740996-9919-11df-80f3-001aa0739303 is associated with a lawful US resident over the age of eighteen and you don’t need to know anything else about them”. And a Web site that accepted any OpenID-based authentication system could let users certify their identity through LiveJournal, AOL, Google, or any other private provider, so people who didn’t want to involve the government wouldn’t have to.
Obviously no system is perfectly secure, and letting a single agency manage hundreds of millions of digital identities raises the spectre of catastrophic failure. However, I believe that the current way we handle digital identity, juggling dozens of half-remembered usernames and passwords, is even less secure, because none of the institutions managing these databases have a strong incentive to do it right, and an attacker can wreak havoc by simply penetrating whichever system is weakest. The alternative to authentication by the government is not authentication by Bruce Schneier, but authentication by Facebook.
Cool new renewable-energy idea of the month
It’s a bird! It’s a plane! It’s a one-megawatt power plant!
via Hacker News
PSA: How to recognize that someone is drowning
The Instinctive Drowning Response… does not look like most people expect. There is very little splashing, no waving, and no yelling or calls for help of any kind. To get an idea of just how quiet and undramatic from the surface drowning can be, consider this: It is the number two cause of accidental death in children, age 15 and under (just behind vehicle accidents) – of the approximately 750 children who will drown next year, about 375 of them will do so within 25 yards of a parent or other adult. In ten percent of those drownings, the adult will actually watch them do it, having no idea it is happening.
If you don’t want to be one of those adults, read the rest.
via kdorian
Shorter Steve Jobs
(Disclaimer: I work for Nokia, which competes with Apple in the smartphone market.)
The e-postman always pings twice
My wife called me at work this morning to report that she hadn’t gotten any email, not even in her might-be-spam folder. “I’ll fix it”, I said, remembering some odd “temporary failure in name resolution” messages I had received from some scripts running on the server (the virtual machine which serves this here site, and which also handles mail to ropine.com) the previous day. So I restarted the network on the server and figured that would be the end of it.
(It’s always dangerous when you think you know what caused your problem, and therefore focus all your attention on what you think is the cause, rather than the actual problem.)
Tonight, when I got home, she said she still hadn’t gotten any mail, and I looked at the situation again, and discovered that my mail server was not listening to the rest of the world. I’m not sure why it decided to take an unscheduled vacation—I assume this all has something to do with an unfortunate incident yesterday where I filled up my hard drive—but I brought it back up, and one of the nice things about ubiquity of spam is that it’s really easy to tell when your mail service is working.
I must admit that every time this happens I wonder whether running my very own mail server on a machine (albeit virtual) that I administer by my very own self demonstrates more geek-machismo than practical sense. However, for the time being, I am too lazy to migrate to doing it any other way.
Mail servers that are properly configured, i.e., those not used for sending spam, should just keep trying to send their messages for about five days before giving up, so if you sent us something today, you shouldn’t have to do anything else to make sure we get it.
Bucky would have loved it
I dare say this is one of the most significant innovations in construction since the geodesic dome.
As you may already know, rock, brick, concrete, and similar materials are strong under compression but weak under tension; metal, by contrast, is strong under tension but weak under compression. Thus, in a building whose structural walls are made of concrete, metal rebar might be used to provide tensile strength.
Nader Khalili patented a building system that works along similar lines, but instead of concrete, he used sandbags, and instead of rebar, he used barbed wire. You presumably can’t build a skyscraper with these materials, but by arranging them in domes and then plastering them over, a team of semi-skilled workers can make single-story buildings—small, medium, or large—that meet California’s standards for earthquake safety.
I would kvell to learn how to make a house like this, but it looks like the only place that teaches the technique is in California, and the alumni are either themselves in California or have moved even farther afield.
Why your e-book really isn't worth fifteen dollars
Recently, there was a dust-up between Amazon, an information technology colossus that happens to sell books, and Macmillan, one of the six conglomerates behind just about every book published in the United States. Amazon wanted to price all of its Kindle e-books at $9.99 or less; Macmillan, not wanting e-books to undercut its sales of paper books, wanted a deal where they set the retail price and guaranteed Amazon a fixed proportion of the take. As negotiations bogged down, Amazon, no doubt wanting to Send A Message, decided to remove all Macmillan books from its catalog—not just the e-books, but the paper editions as well. At which point authors and fans whose books had been delisted Sent Their Own Message, along the lines of, Amazon, You Suck. So Amazon caved, and now Macmillan is free to offer $14.99 e-books to the masses.
Some people may wonder: why on earth should anyone pay fifteen bucks for a stack of bits that cost nigh-unto-zero to reproduce? In response to this argument, Tobias Buckell walked his readers through what it costs to publish a book, pointing out that even before you consider the cost of paper and ink and shipping, the editors and artists and proofreaders have to get paid. According to Scott Westerfeld, the cost of printing a book is between three and ten percent of list price.
I’m not going to argue with Real Authors (especially authors whose books I like) about the economics of the publishing industry, but I beg to point out that in the free market, prices are set by supply and demand. As a representative of the “demand” side, I think something is seriously amiss here.
I can’t stand to read long documents on an LCD screen, so if I’m ever going to take the e-book plunge, either laptop screens need to improve significantly, or I’m going to get an E-Ink-based reader. The Amazon Kindle and the B&N Nook both retail for $259.00. According to the Consumer Expenditure Survey for 2008, that’s double what the average American spends on reading in a year. In exchange for that $259.00, I would have free access to a whole pile of out-of-copyright and freely licensed documents in e-book form, but since I’m not reading those works now, I don’t consider that access to be worth anything to me. What about all those e-books that I can buy?
I checked out Charles Stross’s titles on both Amazon and B&N. When a Stross title is available as a mass-market paperback, the paperback almost always lists for $7.99 (B&N currently is having a “everyone gets the member rate” sale, so they are actually offering them at $7.19) and the Kindle or Nook edition almost always costs $6.39. Buckell’s Crystal Rain and Ragamuffin are available for $7.99 in either mass-market paperback or Kindle form. The books in Westerfeld’s Uglies tetrology are either $8.99 or $9.99 in paperback, depending on the title; the discount on the e-book version ranges from $1.00 to $5.77 (the latter is for the Kindle version of Extras, the most recent book in the series; I assume that’s a promotional rate).
Let’s generously assume, based on the numbers in this survey, that the e-book versions of the titles I would want to buy will sell for, on average, a $2.00 discount off their paperback prices. At that rate, I’d have to buy 130 e-books before I recoup the cost of the reader itself. And who knows how many times e-book pricing and technology are going to change before I buy another 130 books?
Wait a minute. Did I say it cost $259.00 for an e-book reader? But of course, I want to be able to share my books with my wife, and sometimes we both might want to read different books at the same time. So our household would need to get two readers, the cost of which might be recovered after buying 259 books.
And on top of that, the value to me of an e-book is less than that of a paper book, because with the Digital Rights Management built into the technology, I can’t loan my e-book to a friend (unless I bought it from B&N and the title has “LendMe™” enabled as a special feature), sell it to a used book store, or donate it to a library.
Obviously there are a few people out there who find enough other value in e-books to make them worth the money. (Best-sellers seem to have very steeply discounted e-book editions, so if you’re the sort of person who likes to read a lot of best-sellers as soon as they come out, e-books may be a reasonable choice for you.) Hopefully their dollars will sustain the industry until it can come up with a business model that makes sense to the rest of us. And perhaps there are others who thought the cutting-edge technology they bought was a wise investment, and are now fuming about how their hopes have been dashed, their expectations frustrated. All of this has happened before, and it all will happen again.
In the meantime, paper books are serving me well, so I’ll keep buying them.
Overdrawn at the phone bank
I spent today volunteering for the Coakley campaign, helping it try and snatch victory from the jaws of defeat (after the Democratic Party spent a month going “jaws of defeat? what jaws of defeat? we can’t possibly lose!”—but every other liberal blog in the country has that rant). I told the Organizing for America coordinator that I was interested in canvassing, but not in making phone calls, because enough of my friends, not to mention my wife, had been complaining about the incessant phone calls. (What with the mayoral election in November, the special-Senate-election primary in December, and the election itself today, it has been the season for everybody and his brother to leave us a message explaining how So-and-So is the greatest politician since Pericles and could you please remember to vote on Tuesday.)
And hanging out at the office (of a union local that was participating in the campaign), waiting for the van to come pick up us canvassers, I talked to some of the other volunteers and some of the office staff, and they too were kinda sick of all the phone calls. And this was one of the offices from which people were making phone calls on behalf of the candidate. Seeing the instructions for those folks, I could imagine what kind of technology they were using on the back end; it’s sort of cool. A good Asterisk hacker could probably build it in a month or so.
The procedure for participating in the vote-for-Coakley phone bank went something like this:
- Call a certain toll-free number and type in your ID number on the telephone keypad.
- Listen to a tutorial on how the system works.
- When you hear a beep, it means you are connected to a real live person and not an answering machine. Say “hello” and go right into your vote-for-Coakley script.
- When the call is finished, do not hang up. Instead, press the star key.
- Then type in a code indicating the kind of response you got. (I don’t remember the numbers, but it was something like 1 for “hang up”, 2 for “I already voted”, 3 for “I haven’t voted but I will”… you get the idea.)
- You will automatically be connected to the next voter; go back to step 3.
So if you were, say, a college student in California who wanted to do something for the Massachusetts Senate race (on either side—I assume the Republicans have a similar system), but you had no money to donate, you couldn’t fly out to Massachusetts, and indeed you only had the occasional half-hour between classes… no problem! The party could give you their phone bank’s toll-free number, give you an ID number, and you could do your part to grease the wheels of democracy. And with this technology, a small staff could manage to get useful work out of hundreds (thousands?) of untrained passionate volunteers all over the country. Isn’t that great?
Maybe not.
By making each voter telephone contact so cheap for the caller, even as the irritation level for the callee has stayed constant, the whole enterprise of phone-banking is starting to resemble spam. The only difference is that when you get spam in your mailbox advertising organic penile-enhancement products, you don’t really know who sent it, so you delete it (or send it to your spam filter’s tuning system) and get on with your life. If you get spammed over the phone in the name of some political candidate, you remember that candidate’s name, and not necessarily favorably.
In preparation for next fall’s Congressional elections, party committees would do themselves a big favor by deciding how many times they really needed to contact each voter by phone, setting themselves a limit, and publically pledging not to go over that limit. In some districts, this sort of discipline might force them to turn away volunteers—telling them “if all you can do for us is make phone calls, sorry, we already have enough people to reach everyone we need and we don’t want to overdo it”. It might be hard to turn down free labor, but this is a freebie that comes with a hidden cost.
Re-re-reconsidering nuclear power
I generally have a “meh” attitude towards nuclear power. I’m not one of those folks who thinks that anything with the word “nuclear” in it must automatically be Evil. But as a practical matter, nuclear plants have proven so capital-intensive and so dependent on government subsidy that their value as carbon-free power investments has been drastically oversold. Nuclear-plant operators in the United States don’t have to pay the full cost of their own liability insurance. We still don’t have a permanent location for storing high-level nuclear waste, and from what I can tell, if we ever do choose such a location, taxpayers who don’t benefit directly from nuclear-generated electricity will still have to pitch in to cover the cost of disposal. If nuclear operators had to raise capital, buy liability insurance, and dispose of their waste on the open market, nobody would dare to run a nuclear power plant; if other forms of non-fossil-fuel energy enjoyed the same level of subsidy, we’d have solar panels on every roof by now (even in towns where it rains for 364 days out of the year).
All of which is prelude to my fascination with this news: some researchers and operators are developing nuclear power plants that use thorium, rather than uranium, as their fuel. Thorium-232 is far more plentiful than uranium-235, the waste products don’t need to be stored for as long as the waste from a uranium-fueled reactor, and those waste products aren’t nearly as convenient for someone looking to construct a nuclear warhead.
In fact, there was a lot of research into thorium-based reactors in the 1950s and 1960s. (You may recall Heinlein mentioning the element in several of his works.) So why haven’t we been using thorium-fueled reactors all along? Because back in the day, the government wanted nuclear power plants to produce convenient warhead material as a side effect. With the Cold War over and antiproliferation on everyone’s agenda, new priorities and new research might lead to new models for nuclear power—maybe even models that can compete fairly with renewable sources.
via Hacker News
