Jeff's Home Page
Jeff's Resume/Bio
Photo Gallery
Contra Home
About Contra
Contra on LiveJournal
Diary Topic Index
Diary Photo Index

Link to an Entry

Pete Albrecht
Sam'l Bassett
David Beers
M. Covington
Loren Heiny
Jim Mischel
Brook Monroe
Bill Roper
David Stafford
Scott Smith




December 2004
November 2004
October 2004
September 2004
August 2004
July 2004
June 2004
May 2004
April 2004
March 2004
February 2004
January 2004



October 31, 2006: Designing Novels, Part 2: Characters

There was a time when characters didn't matter much in SF. The gimmick and the plot were primary, and the characters were named shadows that moved with the plot like wood chips floating downstream. A lot of the pulp fiction from the 40s and 50s that we now consider unreadable is unreadable for that reason: The hyperdrive is on, but nobody's home.

Although an occasional reviewer will plaster a book for being boring, most of the complaints lodged against modern SF have to do with characters, and what this tells me is that people are now looking for characters more than gimmicks or situations. There's a serious danger here, in that characters without interesting backgrounds or anything to do basically become soap opera, and a great deal of modern SF is costume soap opera, especially series fiction. Situation, characters, and plot are all essential, and in the finest fiction they are woven seamlessly together in a way that makes you gasp when you spin past the last page, wishing it would go on. Of course, one thing that makes good fiction good is that it knows where to stop, even when the readers are left aching for more. The only thing worse than not getting what you want in this case may be getting it.

Alas, nobody's equally good at everything. I'm as bad at crafting characters as I am good at coming up with tech gimmicks. Nonetheless, I've found that I've become better with practice and conscious direction of the characterization process. I need to say here that most of what I know about characterization I learned at the knee of SF writer Nancy Kress, who lived nearby and was a close friend when we lived in Rochester, New York 25 years ago.

Human characters are simulations of human beings. (Most SF aliens are human beings in rubber suits, which is one reason I don't have many aliens in my fiction. Like, almost none.) They have strengths, passions, weaknesses, and personality quirks. You have to be aware of all of these, and keep a character's actions in line with the personality that the reader is given to understand. Good characters grow with the action of the story, as they bump their heads on plot twists and their own limitations, and that may be the toughest aspect of characterization that you'll face.

What I have learned to do, and which works well for me, is to create a sort of character dossier for the most important people in a story. Basically, you sketch out a character's personality. This is not what the person does, but what that person is. It doesn't take a lot of words to do this, and in fact if you run on too long you can get in various kinds of trouble. Here are a couple of dossiers from the novel I'm currently working on, The Anything Machine:

Howard Himmel Banger: Age 14, son and only child of Stella Price Banger. Average height, muscular, shaggy brown hair. Has an inborn talent for pictorial art and illustration but is unsure what sort of career to pursue. Geeky and systematic, he likes to categorize and sort concepts. (His mother is pushing him toward an information science career in content management; basically, to be a librarian.) Prone to nightmares after his father was murdered when he was only seven. A little selfish, less emotionally mature than Gustavia. Not especially confident except when pursuing his passions, and then he approaches mania. Has known Gustavia since they were toddlers in SUNY faculty daycare, and assumes that she will spend her life with him, whether they manage to return to known space or not.

Gustavia Marya "Gusty" Kowalczyk: Age 14, daughter and youngest child of Gustav and Anna Kowalczyk. Short (5'4"), thin, blonde, athletic, gray eyes, pretty without being dazzling. Intends to pursue degrees in economics and business and then teach at SUNY Numenor. Willful, very focused on her studies and her career, and careful lest she make a mistake that will impact her future. Fascinated by animals and the business of agriculture. She struggles with her feelings for Howard, torn between her awakening hormonal attraction to him and her fears that he will not take the place she has outlined for him in her plan for her own future.

The danger in writing character dossiers is not too little but too much. The idea is partly to seed your subconscious machinery with the outlines of a character, and in part to have something to measure a character against as a story unfolds. Both functions depends on how your own subconscious mechanisms work and how you manage them. Characters have a well-known tendency to "grow away" from your original conception of them. This can be a problem, but it can also be borderline miraculous. For example, my early plot outline of The Cunning Blood had no major character named Jamie Eigen. Jamie was at first a nameless spear carrier, a sounding board for Peter while he was in jail. But my subconscious clearly had other plans, and Jamie quickly evolved into perhaps the single most pivotal character in the whole yarn.

Too much detail in a dossier can lead to conflict between what your subconscious wants to do with a character and how you consciously perceive that character. It's best to leave a character dossier a little bit sketchy, so your subconscious mechanisms have room to move. Yes, that's a tightrope walk, and you'll have to zero in on the sweet spot through practice. Again, my intuition is that it's better to say too little in a dossier than too much.

Avoid the temptation to include in a dossier what characters do. Howard discovers the Thingmakers, and creates an index that later evolves into Banger's Big Book of Drumlins ("drumlins" being artifacts produced by the mysterious Thingmakers) but those are plot elements, not character aspects.

Nancy Kress always told me that it's important to understand what each character wants. This is implied by a good dossier, but sometimes it's useful to look at a character from an extreme height: Young Gusty Kowalczyk wants life to be predictable; young Howard Banger wants life to make sense. Most of what they do (and much of the conflict between them) follows directly from those high-level differences. If you can understand your characters that deeply, you can probably do good things with them. So far I'm having a lot of fun, and the story (which is a juvenile; a first for me) is evolving well.

More next entry.

October 29, 2006: Designing Novels, Part 1: The Gimmick

I've been meaning for some time to talk a bit about the process by which I design novels. I don't claim that this is the only way to make a novel, but it's the way my stories come together in those odd back-of-the-head places that I can only barely perceive.

And that's a good place to start: Novels are emergent processes. You don't design one quite the same way you design an electronic circuit. From my experience with my own work in novels (I've written several, though only one is published) and from talking to other people who do it, something like this happens: You toss a certain number of consciously designed elements into the chipper/shredder in the back of your mind, and then you open the gates to see what comes out. Some people toss more consciously designed elements into the chipper than others do. Some people just start writing, and after a few thousand words take a break to see if what has emerged has any value or is just incoherent mulch.

I've tried it both ways, and at many points along the gradient. As usual, there's a sweet spot in the spectrum, where the degree of conscious design is just enough to feed the individual characteristics of the mad chipper/shredder operated by your subconscious. Too much design, and the chipper gums up and the process grinds to a halt. Too little, and the chipper spits back incoherent blather that never coalesces into anything you could call a story with a straight face.

Where do you start? I always begin with what a call The Gimmick, though many call it The McGuffin. I don't like the term "McGuffin," because it implies something that advances the plot without being especially relevant to it. A good gimmick lies at the very core of all the conflict that besets your characters, and it may well be the thing that your readers remember years after reading the story. Gimmicks are extremely important in SF and often in fantasy as well. (Literary fiction generally dispenses with gimmicks entirely, and relies completely on characterization.)

The gimmick in The Cunning Blood is not so much the Sangruse Device (an intelligent nanocomputer) as it is a planet where electrical things don't work. They don't work because of a bacterium-sized nanodevice that homes in on the magnetic fields generated by electrical currents of useful intensity, and breaks down the metal of the circuit's conductive elements until the current stops flowing. Earth authorities (who designed the nanodevice and turned it loose on a nearby Earthlike planet) assume that this condemns those who live on Hell to a sort of pseudo-Victorian gaslight existence. What they forget is that there are more paths to high technology than the electrical ones that Earth followed. Diesel engines, chemical lighting, photochemical amplifiers, mechanical and fluidic computers, fiber optics, they're all in there—and they're all real. Where electricity is crucially important, the Hellions use wires consisting of liquid mercury in thin hoses. The current-sensing nanomachines can't disrupt liquids.

I could have spun a lot of stories around that one gimmick, and I may do one or two more. The point is that it's a good gimmick, and a memorable one. I put a lot of energy into researching it, and many people have told me that Hell sounds completely plausible and real to them.

You come up with gimmicks by paying attention to the physical world. I read a lot of science, a habit I got into when I was in high school, and Isaac Asimov was publishing a fat paperback full of fascinating science essays every few months. I had a nose for the odd corners of science and technology, especially the ones that never made it into the mass market. Popular Science ran an article about fluidic computers when I was in high school. I read as much about Babbage's mechanical computers as I could as soon as I heard about them. You do this for a few years, and stuff starts to precipitate out of that mass of intriguing facts in the back of your head.

I keep a notefile of gimmicks, and add to it when odd things occur to me. Could an astrobleme have such high walls that the atmosphere would be different inside the bleme than outside? I think so. It's a gimmick. I may never use it, but it's intriguing. Could hacked bacteria separate isotopes of metal salts dissolved in seawater? I don't see why not. Not every gimmick is big enough to hang an entire novel on, but there's no law saying that a novel can only have one gimmick.

Once your gimmick is in hand, you have two ways to go: You can start playing what-if, to see what would happen if the gimmick were made real and turned loose, and then build a plot around the consequences. Or, you can design a cast of characters and see how that gimmick might affect them, which would then suggest a plot. Stories have been done both ways. We'll talk more about that in my next entry here.

October 27, 2006: Duntemann's Brain

My sister Gretchen just sent me what may be an early Christmas present: A slightly squishy, crawling, humming, snarling audioanimatronic plastic brain...with teeth. It arrived an hour ago and was not anticipated, but it's been a lot of fun.

It senses nearby motion and sound, and when triggered, it hums a manic little tune for several seconds while crawling forward about a foot. Then it stops, opens its mouth, and starts snarling and slobbering.

We instantly knew what to do with it: We turned it on and put it down on the floor. QBit bent down to sniff it (I'm sure he was wondering which end to sniff and why there wasn't a tail to sniff under) and when he set it off, he jumped two feet straight up. A second try evoked the same response, and since then he gives it a doleful look from a considerable distance.

We've named it Duntemann's Brain, and it will actually earn its keep around here. QBit has become a very avid table surfer, and any time neither of us is in the room, he will hop up on the kitchen table and start licking the spots on the placemats where I've spilled spaghetti sauce or chicken soup. Then he jumps down with one or more placemats in his mouth, and will drag them off to his lair under the bed if we don't catch him first.

Simply put, Duntemann's Brain will become our kitchen table placemats' bodyguard. In between meals, it will sit there to one side of the salt shaker, waiting for QBit to mount the table. Our only concern is that he will break something trying to get away when it goes off and starts humming and snarling.

Thanks, my good sister. This one's a keeper. Now, who will hack the mechanism to make it sing "Let There Be Peace on Earth" or "I Ain't Got No Body" instead of humming?

October 26, 2006: What Sells on Lulu?

Test Prep. Programming. Diets. Dogs. Lately, a technical manual for the 737 airliner. Almost anything but fiction.

It was a useful ten minutes I spent earlier today, studying the top 100 sellers on, the print-on-demand and digital content gumball machine that I'm going to be using for a number of projects in coming months. Some thoughts:

  • Nonfiction rules. Of the top 100 books, I count only six that are fiction, including one written by a nine-year old. Admittedly, #5 and #10 are fiction, but the next tale down is #45, and the last three are all south of #80.
  • Most of the books have competent or even professional looking covers. It's impossible to know how they came by their covers, but I was expecting a lot of pastel cardboard with words on it. Not so. Good photography, good design—it makes me wonder how many of these books are "long tail" reissues of books originally released for the retail store market with professionally designed covers.
  • Many of the topics are things I would consider useful, if not for myself than for at least a significant if small audience. Examples: That Military House: Move It, Organize It, and Decorate It. Nobody moves more often than military folk. This book is a natural. Or: Intermodal Shipping Container Small Steel Buildings. How to make buildings out of used shipping containers. Hey, if you need to do that, you need this book bad. Or: Plone Live; basically, a user guide for the Plone content server. There's a fair bit of whacko stuff in the top 100; typical of which is How to Become an Alpha Male, dubbed "The lazy man's way to easy success with 20 or more women a month." (Good lord.) But I was amazed at how reasonable and practical most of the top 100 nonfiction titles were. Narrow, yes. (There's an entire book devoted to helping Catholics resist the temptation to masturbate!) But—heh— practical.
  • There's not much turnover at the top. Lulu has a "this week, last week" display that shows how the top 100 titles have changed rank, and (for the most part)...they don't. What was #1 last week is #1 this week. Indeed, of the top 30, only two books changed rank at all. (#5 and #6 changed places.) Most books change rank by one or two positions only. The big exception is The Boeing 737 Technical Guide, which is up 27 positions this week. Consistent ranking over time suggests consistent sales. We aren't told what the absolute numbers are, but I suspect that they're constant, if modest.
I'm pleased with the way Lulu works. It looks like people are having some success with it. Again, this is what the future looks like for niche publishing. I'm going to try uploading a book there tomorrow. Cross fingers. I'll keep you informed.

October 25, 2006: Lulu as a Digital Gumball Machine

I don't think I've talked about this yet, but having researched them in some detail, Lulu is probably the closest we've yet come to a digital "gumball machine." You can publish an ebook (or any kind of digital content, including music, photographs, or software) on Lulu, and simply pay them a 20% commission on each copy sold. No other charges. Most other services that I've researched take a lot more than that; some over 50%. The effective minimum charge per item is $1, on which Lulu takes 20 cents. You can charge less than a dollar if you want, but the commission remains at a minimum of 19 cents.

Costs are a lot higher if you want them to print and fulfill paper, of course, so I'm wondering how many people would be willing to buy an ebook laid out for 8 1/2" X 11" pages, and then print it locally. I've been asked to write an article about my homebrew 2-tube 6T9 stereo amplifier, and if I did I might do it that way: A 24-page booklet with schematics, photographs, and detailed instructions, including a parts list with where to get the parts. You could either read it from your screen or print it on paper so it could lie flat on your workbench. I'd sell it for $2 on Lulu, and make $1.60 on each download. I'd be happy with that. The research is long done, the schematic is already drawn, and writing is easy for me. If I made $100 lifetime on it I'd be happy.

Lulu seems to be doing well, and once the world at large understands that they're doing well, there will be competitors, and author costs will go down. We may finally be coming to a day where niche publishing can actually make money. I'm close to getting my first title mounted on Lulu, and you'll read all about it right here.

October 24, 2006: Another Unknown Earworm Captured

One of the great unsolved problems in computing is what they call "query by humming;" that is, the recognition of melodies based on inexact reproduction of those melodies. Google has allowed me to hunt down a great many odd things, like the old "Elmer the Elephant" kids' show in Chicago, or "Video Village," one of the first modern TV game shows, which was Monty Hall's debut. However, it won't help me with a handful of catchy melodies that I have tucked away in the back of my head, some dating back a long way.

I caught one of these "earworms" recently, though it wasn't easy, and didn't depend entirely on the melody. When I was very young (figure late 1950s) I remember seeing a musical cartoon on one of those ancient Saturday morning "cartoon shows" that had no host or anything but people at the TV station putting on one old theatrical cartoon after another. The cartoon I was thinking of was a mixture of animation and live action, with a real guy playing the piano, and a stop-motion animated big-eared gremlin dancing around on the piano to an extremely catchy jazz tune.

What allowed me to catch the earworm was a single memory: At the end of the cartoon (or maybe the beginning) was the word "Puppetoon" in large letters. Google took that and led me to George Pal, who was also the major force behind the special effects in the seminal 1950 film Destination Moon. (Chesley Bonestell did the background matte paintings.)

George Pal created an animation technique we now call "replacement animation," in which the animators basically create an erector set of different body parts in different positions for an animated character, and reassemble the character for every (or almost every) frame. He did a whole series of Puppetoons using this technique for Paramount in the 1940s, including a couple of pretty famous ones, like "Tubby the Tuba" and "John Henry and the Inky-Poo." I poked around on Google through lists of the numerous Puppetoons. One of them, "A Date with Duke," featured Duke Ellington, a master of jazz piano, which seemed promising given the jazzy piano soundtrack. Sniffing around links containing "A Date with Duke," I landed on this page, and there it was: The dancing gremlin that I remember clearly. (The rightmost of the four stills.)

The soundtrack for "A Date with Duke" was Ellington's own "Perfume Suite," which I researched, and found to contain several movements. I researched each of the movements in turn, trying to find somebody with a sample I could play. The first few I movements I found samples for were too slow to be the tune in question. Then I went after "Dancers in Love: A Stomp for Beginners." The first sample I found was, somewhat oddly, played on classical guitar. It seemed right but I wasn't sure; the part of the melody I remember most clearly wasn't there. After much additional poking around, I finally got here, and played the WMA28 sample of "Dancers in Love." Bingo! That was it! I still haven't found the cartoon—it would be fun to see Ol' Big Ears dance again—but at least I was able to nail a CD of the music itself.

So. Whythehell does it have to be so hard? Why can't I just whistle the melody into my headset mic and let an algorithm name that tune? It took over an hour of detective work (though I will admit, I learned a few things, as I always do during that kind of detective work) and the detective work had very little to do with the melody itself. David Stafford says it's a very difficult problem to solve, and he's solved some doozies. Sooner or later, however, it's gonna fall. Let's hope it's within my lifetime; that's one earworm down, but there are still a few more knocking around in there.

October 23, 2006: Why Executive Salaries Can Explode

I've begun to see a great many head-scratching articles about the explosion of executive compensation in the last eight or ten years, especially at very large corporations, while general staff compensation has stagnated or dropped. The puzzle is why people are puzzled; it seems pretty damned obvious to me. Three forces are at work here:

  • We have gone from a general labor shortage in the developed world to a general labor surplus. Demand for labor peaked right after WWII, and now supply has caught up with demand. This may not be cyclical but structural, in that the demand for labor post WWII was a one-time consequence of industrialization leading population, at least in those places where industrialization was happening. Because labor markets are truly markets, compensation for workers outside the executive suite is dominated by the effect of this labor surplus.
  • People at the top of large corporations in effect set their own salaries, because corporate boards, which set executive compensation, are composed almost entirely of other highly-paid executives or retired executives. They look at executive compensation questions and see their own interests in their answers. Nor do the numbers I see indicate that higher executive pay yields higher financial performance of the corporation as a whole. The market doesn't work here.
  • Although CEO compensation numbers seem high to the rest of us, compared to the overall revenues of massive corporations, a few or even many millions of dollars is small change. Paying the CEO fifteen mil a year won't crash the company, and dropping CEO pay to, say, one or two million a year wouldn't help the company, if revenues are in the billions. On the flipside, when you have tens of thousands of employees, giving them all a significant raise could cost a hundred million dollars or more, which would affect overall corporate viability.

None of this makes it right; it just explains how it's possible. And that said, the curves are not moving in the right direction: Revenues generated by continuing productivity gains are flowing toward the top in an ever more concentrated fashion, probably because, with markets controlling labor costs, it's easy for them to do so.

I don't know what the answer is. I like game theory solutions to problems like this, and one suggestion might be to set corporate tax rates by algorithm, where the lowest tax rates apply when total compensation dollars spent are most broadly distributed across all employees. I doubt that that's politically possible, but it's an interesting notion, and might encourage closer monitoring of executive compensation by shareholders.

October 22, 2006: Gouache, Tempera, and Elmer's Paint

I inherited a lot of things from my mother, but her intuitive knack for fine arts (she painted well and worked in ceramics) didn't come down to me, dammit. So apart from a short no-credit "enrichment" course in college that exposed me to silk screening and acrylics, I have almost no experience in painting. So while it's still unclear what sort of paints were used to paint the very early covers in Popular Electronics (see my entry for October 20, 2006) I learned a great deal about artists' paints by fielding the resulting email traffic. Consensus is that they're not classic watercolors, and may be one of the following:

  • Tempera is a species of water-based paints using a binder made of egg yolk. Tempera doesn't allow the sort of deep color saturation that oil paints or acrylics do, so it can superficially resemble watercolor. If I had to guess what Ed Valigursky was using on those old PE covers in 1955, this would be it.
  • Several people suggested gouache, which is a denser but much crankier species of watercolor, in which the pigment particles are larger and there is less water in the paint. Gouache is more used in geometric Miro- or Klee-style paintings or posters than in classic "fine art" paintings in which individual brush strokes combine to form a holistic image. The nature of gouache paints require that they be used quickly, suggesting a presence in commercial art. However, the drying process affects the appearance of the pigments peculiarly, and they are notoriously hard to control across multiple painting sessions.
  • George Ewing was the only one to mention casein paints, which amount to pigmented and slightly thinned out Elmer's Glue based on a casein binder extracted from milk. I had never heard of these, but research indicates that they have been used since the time of the ancient Egyptians. They cover completely and can be as deeply pigmented as oils.
Michael Covington pointed out that shiny paintings are much harder to photograph than paintings having a matte finish, so commercial art would be more likely to use a non-oil process like tempera. Sooner or later I'll hear from an older commercial artist who can nail the question. We'll see.

October 21, 2006: Odd Lots

  • The cap I mentioned in yesterday's entry is apparently a US Army M-1951 field cap (more formally, the "Cap, Field, Cotton, Wind Resistant, Poplin, M-1951") or perhaps a WWII German army M-43 or M-41 field cap, which look pretty much the same. The M-1951 was used from Korea well into Vietnam, when it was retired in favor of a military ball cap. It's unclear whether the Americans or the Germans invented the cap, though I think it may well have been brought here from a galaxy far far away, where it was standard dress for Empire starship drones. The cap recently re-appeared as part of the Army's BDU (Battle Dress Uniform) though apparently only in camo. Thanks to Robert Bruce Thompson (and several others) for putting a name on it, after which Google did the rest. I like the fold-down ear flap, which could be real handy up here in the mountains.
  • Also pertinent to yesterday's entry, the cover artist on those Popular Electronics covers was Ed Valigursky, who was a very popular SF cover artist in the 1950s and early 1960s, and also did covers for Argosy and Saga. No surprise where that square jaw came from, heh.
  • Karen Cooper says that the covers (again, from yesterday's entry) are not watercolors, and, being an artist, she has some reason to know. I don't think they had acrylics in 1955, and although I'm innocent of art training, I do know that oil paints take a long time to dry—certainly longer than the lead time most magazines give artists for cover art. (I do know a little about that.) Maybe pastels?
  • For years I've been seeing little lines of symbols on clothing tags (especially my shorts) without ever becoming curious enough to go figure out what they meant. Finally (only a few hours before Carol came home from her recent two-week trip to Chicago) I got crazy enough to care, and here's Your Guide to Fabric Care Symbols. Just FYI.
  • Scientists at Indiana University have discovered what may be the first living things for which radiation is an essential part of their metabolic cycles. A colony of bacteria discovered two miles below the surface in extreme northern Canada relies on natural radiation (from uranium deposits) to split water into hydrogen and oxygen. The bacteria then use hydrogen to release chemical energy in reaction with local sulfur-bearing minerals. How the bacteria managed to make their way down into several miles of rock is an interesting question, but the implication is fascinating: Living things can evolve to harvest almost any source of energy, and if they can do it under two miles of Canadian rock, they can do it anywhere.

October 20, 2006: Popular Electronics in Watercolor

As I've mentioned here, I've been gathering up old issues of Popular Electronics to complete my collection, mostly off eBay. The other day I got a very nice batch, most from 1955 and 1956, a time when I was still struggling to keep pieces of fish stick on a fork, and had not yet begun building things in cigar boxes. (That time wasn't far off, though.) What was nice about this batch is that it had clearly been stored all these years on someone's bookshelves, and not out in the garage under a tarp held down by half-empty cans of paint in bad 50's colors.

The covers are thus well-preserved, and remarkable. For the first year and change of PE's existence, its covers were water-color paintings. Some were better than others, and the best of them capture scenes that would be difficult to photograph on demand. Consider, for example, the drama of the August 1955 cover below:

A square-jawed hero and his life-saving Civil Defense 2-way radio—talk about the late-night fantasies of young solder-scorched nerds! Less dramatic but no less indicative of the 50's zeitgeist was the July 1955 issue, which was largely devoted to the uranium prospecting craze of that era. Build your own Geiger counter, go out into the still-empty West, and let the riches of the atom announce its presence into your 2000-ohm headphones:

(By the way, I'm sure that there's a name for that style of cap. What is it? Can I get one? I already have the headphones.)

The watercolors went away in early 1956—my guess is that the artist started demanding a bigger cut—and the covers became photographic. Gadgetry and women, often in extremely unlikely juxtapositions, soon dominated the design. This must be what some calculating New York art director figured would bring in the maximal nerd audience:

Gosh! Radio-telephones for everyone—even redheads! (The article on p. 40 was actually a single page about the new Class B 465 MC Citizens Band, and the redhead was nowhere in sight.)

Skimming the mags themselves has been a lot of fun, and it's interesting to see that the nerd passions abroad in 1956 are still with us: Radio-controlled planes and robots, communications of all kinds, audio equipment, and even computers, though more as objects of awe ("Gigantic electronic brains!") than something to be tinkered together in the basement. I hope to do another entry here on some of the ads in the mid-50s mags. The Knight Kit Space Spanner! CK768 RF transistors! "You can make more money with an FCC license!" American Beauty Electric Soldering Irons!

No, I wouldn't want to go back there, but damn, some of that stuff calls to me!

October 18, 2006: My Personal Security Program

Several people wrote (of yesterday's entry) to ask that if I disapprove of Symantec's security solutions, what do I recommend? And what do I use myself?

First of all, it's not entirely about software, and certainly not about one product. One package from one vendor can never be the solution to everything. What I do is fairly simple, and I have little or no problem with malware. Here's what I recommend, because here's what I do, in decreasing order of importance:

  • Be behind a router. They're cheap, and extremely strong protection from unsolicited outside connections from script malware, which studies have shown scan your IP address every few seconds. Even if you only have one PC in the house, get a router. They contain a Network Address Translation firewall, and most allow you to block specific ports. The one I use is the Linksys 8-port router/switch, model BEFSR81.
  • Don't use IE. Just don't. Almost anything is safer. I have it installed because the occasional nitwit Web site won't render correctly without it, but IE often goes untouched for months on end. Use either Mozilla Firefox (which is free) or Opera, which has also been free for about a year now, and is superb.
  • Make sure "install on demand" is turned off in your browser. Certainly, if you run IE at all, this is desperately necessary.
  • Don't use either "big" Outlook or Outlook Express for email. These are malware magnets largely because they're so popular, but they also have some fundamental problems that make them spam facilitators. (They also use IE to render HTML-formatted email, which makes them vulnerable to most of the same exploits that plague IE.) I used Poco Mail for a long time, but recently abandoned it for corrupting my mailbase. I use Mozilla Thunderbird now, and am quite happy with it, even though Poco has a few nice features that I miss on occasion.
  • Don't surf to porn, warez, or obscure music sites, especially those that offer deals that seem too good to be true. These are the primary source of browser exploit trojans.
  • Research every piece of "free" software you install, thoroughly—and resist installing stuff on impulse. This especially includes browser toolbars and ridiculous crap like those heavily advertized smileys, which are highly malevolent spyware and have gotten some of my nontechnical friends into a world of trouble. If you must fool with such stuff, buy a copy of VMWare Workstation 5 and learn how to use virtual machines (VMs) as software testbeds. I test everything I download in a VM before I ever think of installing it directly on the hardware. Stuff that I end up not using much I just leave in a VM image on my very big hard drive.
  • If you must surf to dicey sites, do so from a browser operating in a VM. I go to ebook pirate sites regularly watching for pirated Paraglyph material, and when I do so, I pinch off a new image of a standard VM and then revert it to the stored image when I'm done, whether it looks like I was attacked or not.
  • Use a low-profile virus checker like AVG. I have used AVG Free for some time now, having also tried Panda and found it too resource-hungry. Viruses are less of an issue than they used to be, especially if you're not so stupid as to open any email attachment that rides in the door, or install warez downloaded from P2P networks.
  • Use a two-way software firewall. I use Zone Alarm and have some some years now. It allows me to control what actually gets out to the Internet, and it lets me know when any app even tries to connect. Much modern software, even if it's completely legitimate, wants to "phone home" for reasons never entirely explained. ZoneAlarm puts an end to that. Also, if you contract some kind of spyware or trojan, ZoneAlarm will let you know when the bad stuff tries to make an outside connection.

That's what I do. Interestingly, I don't do something that makes a lot of sense, and I really should try it: Run Windows as something other than admin. If you create and work from within a limited-permissions account, malware based on exploits (or anything else, for that matter) cannot install itself. You have to reboot into admin to install software, but how often do you actually install software? If you're configuring Windows for a non-technical person who is primarily interested in Web and email, and perhaps word processing, this is a very good thing to do. Unfortunately, some software doesn't work correctly in limited user accounts, and you may not know which software fails until you try it.

Finally, before all my Mac friends start yelling, you can buy a Mac. Macs are largely immune to malware because the machines are fairly rare and sparsely connected (compared to Windows) and malware authors are looking for raw numbers. However, if Apple would ever get its head out of its own you-know-what and decide to become a major player, the bad guys could turn their attention to OS/X. I have not yet been convinced that Mac software is somehow inherently exploit-free—and I think complacency would be a very bad thing.

So there you have it. Note that a lot of my list cooks down to, "Don't be an idiot." It isn't all about software. It's about common sense and a little caution. The best trainable, configurable anti-malware system is you.

October 17, 2006: Battling Symantec

I wonder sometimes what poor Peter Norton thinks of the software that now carries his name. I just got back from spending over two hours with a friend, picking fragments of Norton Antivirus out of his machine with a tweezers. I met Peter a couple of times in the 1980s, and he was a completely pleasant guy, if kind of quiet. I used his stuff for a lot of years. I certainly won't use it again, and it has nothing to do with him.

What used to be Norton products are now owned by Symantec. There are two problems with Symantec's security products. (I have no experience with their other stuff.) To wit:

  • The security stuff is so feature-bloated that it will reduce even the fastest machine to a crawl.
  • It is virtually impossible to uninstall in the conventional manner, by way of a Windows uninstaller.

Symantec makes the case that our malware situation is so bad that a really comprehensive security solution can't help but slow a PC down. There is some truth in that, if you feel you have the right to go anywhere online and click on anything you see without consequences, especially if you're running IE. A relatively simple strategy, like being behind a router, turning off browser install-on-demand, not opening arbitrary email attachments, and resisting the urge to visit porn sites will get you 80% of the way there without making your system run like a PII-450.

Uninstallation problems are less excusable. Forget the conventional uninstaller. It just doesn't work. To get rid of Symantec security products you have to download the Norton Removal Tool from Symantec's site and run that—but as I found earlier today, that's no guarantee. When I ran NRT on Dean's machine, it got a certain ways along before telling me that WinFax Pro was installed, and I had to uninstall that first before running NRT. NRT then aborted.

One problem with that: Nowhere that I could see on that system was any least trace of WinFax Pro, and Dean had no memory of ever seeing or using it in the past. That left me the hard way: descending into the Registry to nuke hundreds of Symantec keys, manually disabling seven or eight Symantec autostart services, and deleting hundreds of megabytes of Symantec files. When we were done, the machine ran again like the 2.4 GHz P4 it was born to be. Degunking R Us.

Symantec gets away with this because its stuff is preinstalled on virtually any new machine you can buy over the counter right now, and most people will just pony up the money after the 90 day "trial" (in several senses of the word, heh) than try to pry the thing's ugly little claws out of their systems. Not me, boy. The first thing I do with any new machine is whatever it takes to get every last stain of Symantec off the hard drive. However, since I will probably have to assemble my own machines from now on to avoid the Vista tax, I will avoid the Symantec Disease as well.

Do it if you can. The cure is almost as bad as the disease.

October 16, 2006: Just-In-Time Bookstores

Chris Gerrib posted a link today to an article in the (London) Sunday Times, about what might happen to bookstores if we can ever get the cost of one-at-a-time book manufacturing down to something that competes with conventional book printing and binding.

It's worth reading, if you bear in mind that the author of the piece doesn't completely understand the problem. Much of the disruption in retail bookselling is due to massive discounting by online retailers like Amazon, which forces B&M (bricks'n'mortar) bookstores to meet discounts, at least on popular books. B&M bookstores in turn force some of their infrastructure costs back on publishers by charging for preferred positioning of books. And squeezed publishers then force some of those costs back to authors by cutting advances and royalty rates. (Authors are still looking for somebody to force those costs back to.)

My own opinion is that publishing as an industry still has immense overcapacity from the 90s, triggering a race to the bottom led by large publishers that need to keep staff and facilities busy. Publishing would be well served if about half of publishers went out of business, heh. You first, Alphonse...

Closer to the heart of the problem is the inability to predict sell-through of any given title. That was in fact the topic of a longish article by Jeffrey A. Trachtenberg on the front page of today's Wall Street Journal. (I don't think the piece is online anywhere but the WSJ paid site.) Publisher Henry Holt & Co. took a bath on what they considered a "sure thing," a novel called The Interpretation of Murder. They paid $800,000 for the book (a first novel by an unknown writer) and printed 185,000 copies. Even though early reviewers loved the book, the public did not. We won't know for awhile how many of those 185,000 copies will ultimately have to be pulped, but my guess is, well, most of them.

There are other damaging problems afflicting publishing, primarily the concentration of control in a mere handful of people at two massive retail chains, but the question of capital tied up in roll-the-dice books that may never sell is probably worse. (That is what ultimately put Coriolis Group Books out of business in 2002.) The Times article linked to above suggests that bookstores as we know them will ultimately be replaced by small storefronts at which purchasers can order basically any book, which is immediately printed and bound by a machine in the back room, with a click-to-clunk time of only minutes. People (myself included) have been predicting that for seven or eight years, ever since the Docutech-class of book manufacturing machines hit the streets. Close...but not quite.

Book manufacturing will eventually move to the same premises as book retailing, but with a twist: Books will still sit on shelves for browsing, and purchasers will still roam the aisles, picking out books and carrying them to the checkout counter. The difference is that all of those books were printed in the basement, and as soon as one is sold, the monster book-o-matic downstairs will queue up another one. Front-line staff will constantly grab books from the book-o-matic's output bin and run them upstairs, where they will be re-shelved at the places from which purchased copies were taken. Currently popular books (think John Grisham or Harry Potter) will be printed in tens or twenties rather than ones or twos, but the stock depth will be dictated by an algorithm that monitors sales over time. Obscure titles may not be reprinted and restocked immediately after their shelf copies are purchased, and huge numbers of older and marginally viable titles will not be printed at all until they are actually ordered. The news there is still good: Grab a cup of coffee and in five minutes (depending on machine traffic) The Secret Tofu Recipes of 33rd Order Masons will be handed to you by a beaming young store staffer.

This is not print-on-demand so much as just-in-time manufacturing, and while it will not eliminate wasted paper copies completely, it could cut waste and returns from thousands or tens of thousands of printed copies to something south of a hundred, which is a capital hit any publisher can absorb. We would still have bookstore aisles to meet attractive people in, with books to browse and serendipity around every corner. Publishers will still waste money promoting bad books, but the money wasted will be less. With less wagered on every roll of the dice, publishers may be willing to broaden the base of their frontlists and take more chances on more people. The long tail would get longer and perhaps even fatter, since no book would ever really become unavailable.

I think economics will force book publishing and retailing to a model something like this, but the technical challenges of putting that kind of machinery in every bookstore (and keeping it running!) shouldn't be underestimated. Think fifteen years, not five—and the sooner the better!

October 15, 2006: Nunsuch

Pete Albrecht send me a picture of a toy he found in what's left of the American Science Center catalog, from a place on the northwest side of Chicago where I spent most of my discretionary cash that didn't go for books when I was a teen. The toy is a rubber-band powered gun that shoots little plastic nun figures. It's called (cringe) a "Nunchuck."

The device isn't remarkable; similar toys exist that throw plastic pigs and plastic cats. What I find remarkable about the Nunchuck (and many other nun-toys like boxing nun puppets and wind-up walking plastic nuns that shoots sparks out of their mouths) is this: When was the last time you actually saw a nun? Basically, how is it that young people (under 30 or so) even know what a nun looks like these days?

Nuns were once a part of the American landscape. The big Marshall Field's department store in downtown Chicago (now gutted to irrelevance as a Federated) used to have a separate ladies' room for nuns. I saw the (locked) door reading "Nuns' Lounge" myself on a higher floor, and wondered as a young child what a lounging nun would look like. Everybody remembers Sally Fields' 1970s TV series "The Flying Nun," and "perky nuns" got their own entry (unfairly, I think) in The Encyclopedia of Bad Taste, right after "Pepper mills, huge."

Nuns vanished from the American scene to a great extent by the early 1980s; I think they were gone from my grade school before that. Many smaller orders of religious sisters have vanished completely, and the larger orders are now mostly reduced to elderly nuns caring for extremely elderly nuns. The sisters who remain do not wear the elaborate habits that our culture remembers; in fact, their dress is often simple and plain enough to be mistaken for that of a soccer mom.

What we may be seeing here is the visible manifestation of a more subtle truth: Nuns were a force to be reckoned with. They were the people who made Triumphalist Catholicism happen, down on the street and in the classrooms, by the pure force of their wills and personalities. Many were very lonely, unhappy women, and that's where the bad press comes from. Not everybody can be a runway model or a firefighter, and not everyone can be a nun. But the women who were born to the manner and stuck with it throughout their lives were genuine heroes.

And I think nuns have stuck to our cultural consciousness because of that heroism, not because of the psychological problems that some had. My own Sr. Maristella (my oldest cousin, now in her 70s) is one of the kindest and most soft-spoken people I've ever met, and although there were some whacko nuns in my own grade school, most of the nuns I had managed classrooms containing 48 or even 50 rowdy kids with an equilibrium (and in some cases a sense of humor) that no one appreciated at the time.

We remember nuns because they are now mythic creatures, much like steam trains, which built America with the same impressive mythic power. Nuns had (and still have) a grip on our imaginations. Heroism and true power do that in a way that deviousness does not. (Quick: Show me a toy that tosses little plastic senators!)

October 13, 2006: Odd Lots

  • One very useful application of Flash lies in posting schematic diagrams. I took the Visio schematic of my 6T9 tube stereo amplifier, exported it from Visio as a .AI file, imported the AI file into Flash, and then exported the diagram as a .SWF file. The SWF is only 22KB in size, which is much smaller than a TIF of similar resolution would be and comes down in a fraction of the time. Take a look. It prints well from Firefox, with the sole glitch that the art does not appear in the print preview screen.
  • I pass this along only because Herman's Hermits' "No Milk Today" is one of my all-time favorite pop songs—and for the sake of an animated bouncing udder. As silly as anything I've seen in some time. Maybe a very long time.
  • Jim Strickland sent me a pointer to this, which I would perhaps consider even sillier if it didn't border on "too true to be funny" territory.
  • Make sure you get one of these before your next all-geek cocktail party.
  • I take a certain amount of heat for building and flying kites. How immature, heh. But lo! I may be getting my revenge. (How soon before this gets into Popular Mechanics?)

October 12, 2006: Resurrecting a Computer Book on Lulu

Back in 2001, Julian Bucknall published The Tomes of Delphi: Algorithms and Data Structures with Wordware Publishing, and it became something of a legend. The book's been out of print for awhile, and used copies are selling on Amazon Marketplace for $200 and up. Whoa.

There's definitely a message in that market. With rights now reverted and in-hand, Julian has republished the book on his own through You can now order that $200 book for $25. Wordware sent me a review copy when it first came out in 2001 (alas, too late to review in Visual Developer Magazine) though it disappeared somewhere along the way. I just ordered a new copy, but my memory of the book is still solid: If you do anything in Delphi beyond stringing components together with one-line event handlers, this book needs to be on your shelf.

That would be good enough news right there, but Julian's gone further: He's written a three-part article on how he went about re-creating and re-publishing the book, which stands as the best single description of the print-on-demand publishing process that I've yet seen. Part 1. Part 2. Part 3.

If you have any least interest in print-on-demand publishing, read them closely and in order. A number of lessons emerge from his experience:

  • Publishing at book length is now possible for almost anyone. It's still a lot of work, but there are no real gatekeepers, as there are in conventional publishing. You don't have to persuade anyone to list or carry your book. You follow the process, and it's published.
  • There is very little real money in print-on-demand (POD) publishing. This has been true since the dawn of POD time. What is said less often is that there is very little real money in conventional publishing anymore, either.
  • The real strength in POD publishing is re-publishing out-of-print books to capture "long tail" demand. Julian's book is exceptional in that regard, in that the book had an unusually good reputation in its first life, rather like Michael Abrash's Graphics Programming Black Book, a 1,340-page behemoth that we published at Coriolis in 1997. Used copies were selling in the $80 range until relatively recently.
  • But that said, POD publishing is extremely length-sensitive. Unit manufacturing cost is 2c per page plus $4.55. Julian's book is 524 pages long, so its UMC is $15.03. That's over twice (and approaching three times) what it would cost to print each copy in 4,000 copy quantities using a conventional offset press. Michael's book would have a UMC of $26. Ouch.

Julian tells us that he has sold 110 copies in one week. That's little short of astonishing, though some of that is probably an initial blip of pent-up demand. Books like this are less version-sensitive than most computer books, so there's a good chance that the book will sell steadily for years to come. And one overwhelming advantage of Lulu-style self-publishing is that the author does not have to warehouse books, process orders and payments, dun deadbeat retailers, or handle the inevitable bookstore returns.

Julian speculates (and I agree) that POD books should go for depth, not breadth. It might be better for all parties if POD books were shorter (100-200 pages) and priced at about $20-$25. (I think that Julian underpriced his book by $5-$8 and consider it a honking bargain.) Based on his experience, I am considering taking my book Degunking Your Email, Spam, and Viruses and breaking it up into three separate books along the obvious topic lines, updating them, and publishing them on Lulu. I don't expect his kind of success with it, but the book reviewed well enough for me to think that it still has some value.

My efforts to publish on Lulu were stymied for months by a weird routing problem that I recently solved (see my entry for September 20, 2006) and it's time to get back on the task. More here as it happens.

October 11, 2006: Review: The Mars Run

William Langewiesche's cover story for the September 2003 Atlantic ("Anarchy at Sea") was a riveting description of high-seas piracy as it happens now. None of this "arrrrr, matey" stuff. It's about multiply nested shell corporations, illusory ship registrations, an utter lack of law enforcement on the high seas, and (especially) assault rifles. Although I've read of "space pirates" in cheap novels for forty years, all of them have leaned toward the kind of nitwit romanticism that has ultimately given us Pirates of the Caribbean and a lot of funny hats.

Not this one.

Chris Gerrib's first novel is about space piracy in a near future (the 2070s) in which commerce among Earth, Mars, and certain asteroids has become routine. The story is told from the point of view of a 19-year-old girl whose disfunctional family cannot help her with conventional college tuition, so she enlists in a sort of space merchant marine academy and becomes a bottom-rung astronaut. On her first run to Mars her merchant ship is attacked by pirates, and she is given a choice: Join us as a pirate, or die on the spot. She joins, intending to escape as soon as possible, and travels on to Mars and back to Earth, where she spends some time in the pathetic Central African Empire (a real country of sorts, albeit one where the emperor had the bad habit of eating his subjects) which is creating a spit-and-baling-wire space navy specifically to prey on spacecraft from developed nations, including the US. Janet eventually triumphs, but it's a difficult, degrading road, which sees several abortive escape attempts and some time spent as a sex slave of the son of the pirate ringleader.

This is a first novel, and for a first novel (I've read more than a few) it works quite well. Chris is a Navy man and has spent considerable time on shipboard, so his descriptions of the nuts'n'bolts of long-haul space travel ring very true. He also has an intuitive grasp of the criminal mind, and we get a clear picture of the pirates as reckless, clumsy opportunists who succeed by a small measure of skill and a great deal of luck—which eventually (as with all criminal luck) runs out.

The great flaw of the book is actually a shortcoming of the medium: We have a size eight novel built on a size six frame. The story sucked me in and moved me along, but there was a bare-bones feeling about it, and I kept wanting more of the background of the future that he's painting here. Janet Pilgrim's character development was present and adequate, but I felt that, given her circumstances, I would have liked to know more about the woman she rapidly becomes once she leaves the Earth. Even with her as the viewpoint character, a great deal happens "off stage."

After talking about it with Chris, the explanation was clear, and something entirely new in SF publishing: The constraints of the print-on-demand publishing medium. The Mars Run was published through, where the unit manufacturing cost (UMC) is directly proportional to the page count. Chris calculated the length of the novel that would allow him to make money while keeping the cover price in line with conventionally published trade paperback SF, and he wrote to that length. As I discovered while writing The Cunning Blood, a given story wants to be told at a given length: What I had hoped would be a 90,000 word novel turned out to be 145,000 words instead. Chris found the discipline to keep the word count down, but at the cost of the story's backdrop fading into a necessary indistinctness.

That's really a vote of confidence from me. The action was straight-line and almost continuous, once we got past a slightly slow beginning. I would gladly have read it at twice the length. There's only one caution: Although the protagonist is only 19, the story has some very violent moments and there is a fair bit of explicit sex, so it is not something you should hand to young readers.

And one other interesting note: Lulu sells the book both in printed form and as a PDF-based ebook. I bought both, and read the novel twice, once in each medium. (I read the PDF on my Thinkpad X41 tablet while shivering in the corner of our unheated RV on our peculiar, blizzard-haunted "late summer" camping trip up in the mountains.) The DRM-free PDF sells for only $2.68, and you get it Right Now. (Shipping time for the print edition is perhaps a day or two slower than what you'd see with Amazon.) I was interested in seeing how good Lulu's order fulfillment is, and was very happy with the results.

187 pages. Paperback $10.96. Ebook $2.68. ISBN 978-1-4116-9973-1.

It's about the future, in more ways than one. Recommended.

October 8, 2006: Finally, Bev Bivens on CD

A very long time ago, I bought the first We Five album (on vinyl—this was 1966) and dropped it on the basement floor, taking a huge chunk off the edge. I had played it maybe twice; if my parents had been home at the time, they might have heard a few brand new words out of me, right up through the floor. Somehow I couldn't bear to buy it again, especially with so much to choose from and so little money to actualize my choice. And I never thought to look for it again until a few weeks ago, and was delighted to find that the first two albums from We Five are now available on a single CD.

Most of you have probably never heard of We Five, and at most you may have heard their sole hit, "You Were On My Mind" on the oldies stations or in K-Mart. The group was in fact a musical framework for one of the most amazing female voices to come out of the 1960s: Beverly Bivens. (Not to be confused with 70s artist Beverly Bremers. Please.) In my estimation, only Karen Carpenter ever topped her for smoothness, purity of tone, and vocal emotion. There is excellent vocal harmony here, especially in "Softly, as I Leave You," but without Bev, there wouldn't be much to comment about.

We Five was odd among 60s groups for a number of reasons, perhaps the most significant of which is that they didn't try to write their own music to pad out their albums, but preferred to do quirky covers of Broadway show tunes, like "I Got Plenty of Nuthin," "Tonight," "My Favorite Things," "Softly as I Leave You," and "Small World, Isn't It?" as well as covers of pop songs made famous by others, like "Cast Your Fate to the Wind." This makes the disc completely listenable, something that isn't always the case with albums. There are a few loser cuts, but I almost found myself hearing them with a pang of nostalgia. Some of the arrangements do not work well, particularly "I Can't Help Falling In Love With You, which is too fast (and has a little too much Spanish guitar) to do justice to the lyrics. But overall, it's a fine collection, and Bev Bivens can certainly make a song work that would simply sink without her amazing voice.

You may have to be an Almost Old Person like me to love it like I do, but hell, that's what blogs are for. Highly recommended.

October 7, 2006: How We Used to Grow the SF Market

Back when Carol and I were in Chicago this summer, cleaning out her mom's house, we found a lot of old books, some of which had been in the basement for a long time and (quite apart from their yellowing and brittleness) just smelled. One was a 50c mass-market paperback murder mystery from 1961, and although the book fed the dumpster, it had a couple of familiar bind-in postcards promoting the Science Fiction Book Club.

Like most people, I subscribed to the SFBC for some years in the 70s, and it had an interesting effect on me: I read stuff I wouldn't have chosen on my own. Some of the lousier titles in that group are long gone, but two of the better ones that I still have are The Winter of the World by Poul Anderson and The End of the Dreams by James Gunn. I don't much like heroic fantasy, and I find cover paintings of impossibly endowed women in brass bras a little off-putting, but Anderson's story sucked me in and I felt that I had gotten my money's worth. I think Gunn's book was the first of his that I ever read, and I liked it enough to hunt down the rest of his material.

So the SFBC was a brilliant concept: Push stuff at people, having given them limited but not absolute choice, and you will broaden their tastes, and thereby broaden your market. The fact that they placed their postcards in books without any least relation to SF (like murder mysteries; who knows where else they went?) was something that wasn't clear to me at the time, but it's just the next highest level of the same concept: Get people curious about what they haven't seen yet. It helped that reading was a popular pastime back then; a lot of people bought stuff "just to have something to read" and weren't especially picky about what it was. Give them limited choice ("choose any three!") and they'll feel like they had a hand in the decision, while the real choice is made back at the central office.

Today, with Amazon allowing us razor-sharp choice in our reading, we tend to narrow our tastes until we start complaining, "Ten million books and nothing on!" It makes sense to read outside your core tastes, but nothing compels us to do that anymore. And no matter how much we may say we value recommendations calculated to our precise tastes (in wine, books, music, and anything else of an aesthetic nature) there is something delightful about stumbling across something you enjoyed but would never have chosen on your own. That was what the SFBC was about, and I wonder if there's any future in such a "surprise me!" mechanism in our heavily networked age.

October 6, 2006: What I Mean When I Say "Flash"

If I learned anything in the wake of yesterday's entry, it's that different people are touching entirely different parts of the elephant when they talk about Adobe's Flash. For most people, Flash is still the animated movie-maker that it's been since the very late 1990s. (And people really really hate Flash splash screens!) But non-artist that I am, that's not the aspect of Flash that interests me.

What I see when I think "Flash" is a mechanism for coding up and deploying RIAs (Rich Internet Applications) that (preferably) contain no animation at all. The platform that Adobe created for doing this with Flash is called Flex. The Flex platform is conceptually similar to AJAX, but seems way less haywire to me.

Flex consists of three major elements:

  • ActionScript 3.0, which is an object-oriented scripting language superset of ECMAScript.
  • MXML, which is an XML-derived markup language for defining UIs and data bindings.
  • The Flex Class Library contains components including controls, data containers, and data protocol managers.

A Flex compiler compiles all three of these elements into a binary file with the extension .SWF. The SWF file is loaded and run by the Flash Player (almost always a browser plug-in) which contains a bytecode virtual machine and a JIT compiler. The jitter compiles the SWF to native code as needed. When you've got a finished and debugged SWF, you publish it to a Web host where people can request it via HTTP and execute it via their Flash Player browser plug-in.

As with most compiled languages, you can write source code files in any old text editor and then compile them to SWF with the free command-line compiler that Adobe makes available. But I was delighted to see that Adobe has implemented a Flex IDE using Eclipse. The product is called Flex Builder, and I just installed the trial version this afternoon to take a look. The Eclipse implementation is very good, and even has a Delphi-like drag-and-drop palette for Flex components. The Adobe site has some nice tutorials, and I was able to create a (trivial) app in twenty minutes, allocating abundant time to poke at the IDE and understand what it was I was actually doing. The $750 price of Flex Builder is a little daunting, but I will admit, it's very slick.

So that's what I really mean when I say "Flash": a genuine programming platform where the focus is code and not eye candy. I suspect I should cut the confusion by calling it "Flex." As the Star Child said, I'm not entirely sure what to do with it yet—but I'll think of something.

October 5, 2006: Odd Lots

  • Mitch Kapor is redeeming himself from the embarrassment of his silly look'n'feel litigation days by mounting (and apparently funding) a very promising Open Source replacement for Microsoft Outlook called Chandler. I'm not sure it's quite complete enough to spend much time with (and sheesh, Mitch! You've been at it for three years now!) but I'm watching it closely. I've tried a lot of similar software, but have yet to find anything I could really throw my weight behind. This could be the one. We'll see.
  • I ordered a computer table today from Unfinished Furniture Warehouse, on Stone St. in Colorado Springs. It's basically a hack of a standard trestle-based small dining table, in alder wood. They're taking down the vertical members so that the table surface is 26 1/4" above the floor, and cutting the table depth from 40" down to 32". The table length of 72" will be unchanged. For the stain they're going to match a shelf I took in from one of my bookshelves here. They have a number of finished pieces on hand as examples of their work, and I am mightily impressed.
  • I'm studying the Flash platform, and it strikes me that Flash is hugely better than HTML for anything but the simplest of static Web pages. In fact, Flash is amazingly like Java and .NET, which must be one of the best-kept secrets of the Web programming world. What I knew of Flash came mostly from books that Coriolis published on it circa 2000, which were targeted at graphics designers and seriously shorted the programming side of the platform.
  • Cripes. I'll bet the German police are killing each other to get put on this case.

October 3, 2006: My New Chair

Back in the summer of 1988, when we were still living in Scotts Valley, Carol and I went up to San Francisco just to walk around, and we happened into the Just Chairs office furniture store on Bryant Street. I sat in a few chairs, and somehow one just felt absolutely, unutterably right. I bought it on the spot, even though the damned thing cost me $1200, which was more than we had ever spent for any single item of furniture in our 12 years of marriage.

This almost magical chair was a Grahl Allmatic 8 (above, left), made in Germany, and it became the chair I sat in to do my computing for the following 18 years. Back in 1999, the little hydraulic cylinder in the vertical column split, leaking pale pink oil in a puddle on my (fortunately) non-permeable chair mat. After that the vertical position of the chair began to sink slowly with use, and every couple of days I have to raise it. Since then I've been looking for a replacement. The leather has been cracking for some time, and is completely worn away on parts of the armrests. Just before we left for our (long) summer trip to Chicago, the vertical column seized at its lowest position, and I knew that it was the end of the line for the Allmatic.

Alas, Grahl Allmatic chairs haven't been available in the US for a number of years, and it's unclear whether they're still even manufactured. (The Web site you may see if you search is abandoned, by a dealer who clearly doesn't understand the value of keeping Web sites current.)

So. I searched, and searched, and plopped my butt in countless chairs, to no avail. Then, a few weeks ago, Carol and I were up getting QBit his special dog food at a pet store up on Woodmen Rd, which is a few doors down from a place called Relax the Back. In addition to pillows and bed mattress tops, they had chairs. Out of habit I started testing in my usual fashion, and then....


It was 1988 all over again. The chair in question is the Lifeform Ultimate Mid-Back Executive Chair. It bears more than a passing resemblance to my old Allmatic, and adjusts in more ways. It also costs twice as much, though adjusting for inflation it's only about 30% more than the Allmatic cost in 1988. I don't much care, considering the proportion of my waking hours that I sit in front of this damned screen, hammering on the keyboard. I have a chair again. My rear end is happy. The next step is finding a computer table at my accustomed height (26" from the floor) which has also been a challenge. The table I used until we moved in here was something I cobbled up myself (and ugly) and 26" is not a standard table height. There's a custom furniture place up off Fillmore that says they can do it. I need to get their quote, but it looks good.

Lesson: Computer furniture matters, if pounding a keyboard is how you make your living. It's interesting that I never had headaches of any consequence until we moved in to our new house—which was also when I started using a tall table with a chair that was on its last, um, legs.

More as it happens.

October 2, 2006: Thirty Years

Thirty years ago today, I promised the young woman who had won my heart that I would stand by her no matter what—and I'm going to admit here that I had little clear concept of what I was promising. But how could I? I was an odd young man whose father had dropped out of the game when I was 16, and most of what I knew about life and love I had picked up reading other people's tall tales.

I learned quickly that love, like life, was hard work. Standing by your spouse does not mean fending off attackers, as my silly imagination sometimes suggested. I learned over the years that it means transforming yourself in such a way as to transmute lust into passion, and self into partnership, passion and partnership welded together with a bead of unbreakable friendship. This is not easy, and I'm not entirely sure I know how we did it. If there were attackers, they grew out of my own weaknesses, but I beat 'em. I saw the same process working in Carol. ("Work" is the operative word.)

Thirty years on, we have arrived at middle age transformed, completed, and victorious. Done correctly, that's what marriage is about. It's possible for single people to reach the same level of transformation and completion (though I'm sure it takes just as long) but damn, it's good to walk that path with someone right there beside you!

October 1, 2006: Odd Lots

  • Larry Nelson sent me a pointer to a site where you can download free PDF generators for a great many different kinds of graph paper, including some very odd stuff (for Americans, at least) like penmanship guide paper for Chinese ideograms. Music staffs, ledger sheets, guitar fretboard diagrams, and lots more. Brilliant stuff. All he lacks is "do-it" list paper. I'll request it; surely that's as important as Palmer Method for Genkoyushi!
  • Bill Higgins attended the Rocket Belt Convention in Niagara Falls earlier this month, and wrote brilliantly about it in his blog. I have doubts about rocket belts as a real-world concept, but of course they're what almost all geeks dream about. Read Bill's blog starting on September 23 and on for the next few days. (He actually got to try on—if not fly—the original Bell rocket belt!) Also check out his convention photos on Flickr. Also do read his reposted 1992 essay about rocket belts and other unlikely flying machines.
  • Related to the above is an exhibit of the work of Ky Michaelson, whom Bill Higgins characterizes as someone as close to a grown-up Tom Swift as humanity is likely to produce. Get lost in the site for awhile and see if you don't agree. A rocket-powered wheelchair! Yee-hah!
  • Sandisk has announced 12 GB and 16 GB Compact Flash memory cards, which should be available by the first of the year. This market is being driven by very high resolution professional digital photography, but if you really really need to be able to take 16,000 snapshots on your Kodak pocket camera before dumping to disk, well, here's your ticket.
  • Although the Wi-Fi business has been a little sleepy since Wireless-G finally happened a couple of years ago, I'm seeing an occasional flash of brilliance in terms of new niche products. One interesting approach is to build a USB Wi-Fi client adapter right into a fat vertical gain antenna, as is done in the WaveRV system from Radiolabs. Although positioned for RVs (which is how I ran across it, through chatter on the Yahoo Groups RV Communications group) it would work just as well in a car. There's no microwave cable loss between the client and the antenna, as all signal is carried between the WaveRV unit and the computer through a USB cable.