technology

Following up on the topic of the endocrine disruptor bisphenol A (BPA), it seems that exposure from water bottles and other plastic food and beverage containers may be dwarfed by that caused by a surprising source: cash register receipts and other papers from thermal printers, and also carbonless copy papers. (Thanks Sean McCutcheon for the tip.)

These printing technologies rely on paper impregnated with zillions of microcapsules containing ink or dye; pressure or heat breaks the capsules and lets the ink out. (Fun fact: up until 1970, the dye was often made with polychlorinated biphenyls -- PCBs, delightfully toxic chemicals). In some methods, the paper has to be coated with a developer that reacts with the dye; that's where "phenolic resins" like BPA come in.

Gizmag reports on the explosion of "extreme beers" containing more than 20% alcohol.

Typical beer has around 4 to 6% ABV (alcohol by volume); particularly strong beers using special strains of yeast and careful brewing can get up around 12%, giving quite a surprise to the unwary drinker.

In the 1990s, Samuel Adams upped the ante with their 17.5% ABV Triple Bock (1994), 21% Millennium (1999), and 24-27% Utopia series (2002-2007): beers said to be more like brandy or port than traditional beers, but still made without distillation. (No, I haven't tried any of these yet. You buying?)

Then, just over a year ago, interest in the century-old technique of "ice distillation" heated up. Ice distillation takes advantage of the fact that water freezes more easily than alcohol: freeze beer just right, and you can remove much of the water (as ice) and get a stronger beer left behind. (Note that according to the wik, freeze distillation can also concentrate poisonous compounds like fusel alcohols; so don't try this at home.)

Breweries using this technique have been in an arms race recently, rapidly taking the record from 31% ABV to an astounding 43% -- 86 proof, the same alcohol content Jack Daniels whiskey used to have (before JD's wimp-out of a few years ago).

"I am confident we can get to 50% with all the right qualities,", says Georg Tscheuschner of Schorschbräu, the maker of the 43% "Schorschbräu Schorschbock".

In a ha-ha-only-serious investigation, Robin 'Roblimo' Miller explores the link between the music of Frank Zappa and the GNU/Linux operating system, the Free Software (free as in freedom, not (necessarily) as in price) alternative to corporate bastards like Microsoft and Apple.

And this is why he can say, with total authority, that Zappa's "Dinah-Moe Hummm" is totally about Linux, at least in spirit, while the song "Montana" with its talk of zirconium-encrusted tweezers and dental floss, "is obviously about Mac users."

...

In the early 70s Zappa wrote and performed a song called "Penguin in Bondage," a foretelling of the various anti-Linux lawsuits and threats from SCO, Microsoft, and other evildoers.

Zappa was also a heavy user of the Synclavier, an electronic music-machine that was a precursor to today's "studio on a computer" recording and sound editing software. Today, I strongly suspect Zappa would be using Linux and Ardour for most of his recording and composition.

Computers, as many people know, at root work with numbers in the binary (base 2) system. This is because the on-off nature of digital circuits maps very easily to a series of 1s and 0s.

How, then do we get letters (and other symbols), and fractions, and very large numbers, out of the thing? We need to have ways of encoding them in binary.

For example, we might all agree that 01000001 represents 'A'. It wasn't too hard back in the days when we only let Americans use computers (I kid, I kid) and so had a relatively small number of letters, numerals, and punctuation marks to account for; but as we started to deal with accented characters, and typographical symbols like ©, and then Chinese and Japanese and Arabic and...well, it got complicated, and it's something that programmers often get wrong. Gobbledygook all too often shows up in web pages and other documents.

And on the math side, rounding errors continue to be a problem in many applications.

So here are two useful guides that everyone who writes software ought to read:

• Joel on Software's "The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets (No Excuses!)", dealing with the binary-numbers-to-letters (and other symbols) problem; and
• floating-point-gui.de's "What Every Programmer Should Know About Floating-Point Arithmetic", which explains how we get fractions and very large numbers out of strings of 1s and 0s.

As advocates of nuclear fission use the Gulf oil drilling disaster to claim that nuclear is a safe alternative, we ought to keep in mind the latest news from the Vermont Yankee nuclear power plant, where a leak containing thirteen different radioactive substances was found on Friday -- in a pipe in a hole workers dug to find the source of an earlier leak.

Vermont Yankee officials admitted that they had misled state regulators and lawmakers regarding the use of underground pipes to carry radioactive substances.

According to the plant's owners and the Nuclear Regulatory Commission, the public faced no danger. Whether you believe that the NRC is providing better oversight to nuke plants than the Minerals and Mining Service was to offshore oil drilling, is for you to decide.

Radioactive tritium was discovered around the plant in a monitoring well back in January. According to watchdog group Beyond Nuclear, there is evidence of 15 radioactive leaks at 13 different reactor sites between March 2009 and April 2010, and at least 102 reactor units have had had recurring radioactive leaks into groundwater from 1963 through February 2009.

Controlled nuclear fusion, and "energy amplifier" designs using thorium, may eventually provide practical and safe nuclear power. But waste, safety, fuel limits, and weapons proliferation concerns make uranium and plutonium fission poor choices. We should instead focus our resources on making good use of that large fusion reactor that Providence has located just 93 million miles away.

href="http://www.marketingsherpa.com/sample.cfm?contentID=2529#">Some words of wisdom on animated "splash" intro pages for websites, from Jared Spool, of Macromedia' User Interface Engineering group. Yes, this article is old enough to mention Macromedia, not Adobe; and yet some people still haven't got the message:

Jared said, "When we have clients who are thinking about Flash
splash pages, we tell them to go to their local supermarket and
bring a mime with them. Have the mime stand in front of the
supermarket, and, as each customer tries to enter, do a little
show that lasts two minutes, welcoming them to the supermarket
and trying to explain the bread is on aisle six and milk is on
sale today.

"Then stand back and count how many people watch the mime, how
many people get past the mime as quickly as possible, and how
many people punch the mime out.

"That should give you a good idea as to how well their splash
page will be received. That's the crux of it."

At first hearing it sounds like something straight of Snow Crash: researchers at the University of South Carolina took a cotton t-shirt, treated it with boron, and ended up with a fabric reinforced with nanowires made of boron carbide -- a material used in tank armor.

Sadly, a bullet-proof t-shirt is probably some ways away yet. But the technique produces a strong, light material that could be used not only in body armor but in the manufacture of lightweight fuel-efficient vehicles.

If you've never studied computer science -- move on, nothing to see here.

But if you have, you'll appreciate Mike Davey's realization of a Turing machine.

A Turing machine, for those non-geeks who still with me, is a theoretical entity that models the process of computation. In 1936, Alan Turing -- one of most brilliant mathematicians ever, who helped break the Nazi's "Enigma" code and save Western civilization, and for his reward was prosecuted and sentenced to "chemical castration" via hormone injections for his homosexuality, but I digress -- proposed a mathematical model of computation that every CS student still studies today.

Imagine an infinitely long tape, divided into cells, where each cell can hold one symbol. This tape moves back and forth under a read/write head (or the head moves over the tape, it's equivalent), which can read the symbol under it and/or write a new one. The machine has just enough memory to hold one number, its "state", and a finite set of rules that tell it what to do when it seems a given symbol while in a given state: for example, "in state 17, if you see a 0, move the tape 23 places to the right and write a 1".

Anything that can be computed, can be computed by a Turing machine. You might think, for example, that a machine that used a two-dimensional grid of cells could do more than a Turing machine, but it can be proven that a "TM" can simulate such a 2-d machine, and so compute anything that it can. I wrote a lot of proofs involving TMs when I was in grad school.

Of course this is not a practical thing to build. Turing meant it only as a sort of thought-experiment. That's part of the beauty of Davey's construction: it's absolutely useless, completely a work of math art.

This is minor, but god(dess)(s/es) how it annoys me...

Look at one of your credit cards. Notice how the digits of the card number are displayed in groups of four, something like "4111 1234 5678 9011".

Now, try entering the digits in just that way on some website where you want to buy something using that card. Odds are 50/50 that the site won't accept the input, that it will require you to type them in as the much less legible "4111123456789011".

There is absolutely no reason for this other than laziness or incompetence. Telling the program to remove spaces or dashes is trivial -- a decent program should be able to accept either format. (I've done it for my employer.)

Modern programming languages have a rich array of functions and methods for string manipulation. Learn to use 'em or go home and leave the programming to the grown-ups, kids.

I've got a fairly slow DSL line out here at the Secret Headquaters: a 384k symmetric DSL line. Now that's faster than dial-up, but a lot slower than cable or other DSL services. (I have my line with Cavalier, and other than the speed have been generally satisfied.

I'm just barely close enough to the CO (the telephone company "central office") to get DSL service, and because of the distance have been told I can't get a faster line. So I've been looking at options.

There's cable, but a) Comcast sucks, and b) a cable connection is shared with everyone on your block. There are performance and security concerns with the whole setup.

So I was thinking about Verizon's FiOS. Now, yes, Verizon also sucks, so I was reluctant to consider it, but I figured I'd at least check it out.

So I sent them an e-mail with some questions: technical ones like the availability of static IP addresses, and billing ones about the fees they might tack on. (I do not understand how it is legal for telcos to advertize a $49.95 price and then add a whole bunch of unmandated "fees" on top of that, as much as $20 more. Not taxes, mind you, that their competitors would also have to charge, but "fees" that they choose to charge but don't include in the price you sign up for. How is this not fraud? Grrr.)

The response from Verizon? "In order to provide you with the best customer service, please contact our Verizon FIOS Sales and Customer Service department at (800) 837-4966 Monday through Friday, between 8:00 AM and 6:00 PM Eastern Time."

Uh, no. I took the time to write out my questions so that we could have precise communication. I do not want to wait on hold to talk to a salesdroid in your customer service department. If you are not willing to answer my questions in writing, if this is how you treat a potential customer, then thank you, but no, we will not be doing business.

(So now I'm considering Sprint's 4G wireless. Not as fast as FiOS, but they did get right back to me when I e-mailed them questions.)