March 31, 2004:

A few odd notes from my notefile on offshore outsourcing, which many on the left have gotten to calling "the race to the bottom." As usual, all is not as it seems:

  • We in or close to the IT industry see things as worse than they are in the job market as a whole, because of the number of people who entered the IT industry in the runup to the Internet bubble in the late 90's. We minted a huge number of new MCSEs (helped by Coriolis' Exam Cram series, heh) and other paper-credentialed professionals, most of whom had work until the bubble burst. After that, well, demand was more like what it was in the 1980s, and we will probably never employ that large a number (per capita) of IT people again.
  • Enrollment in CS and other IT-related university programs is down...to about what it was before the runup began.
  • All that said, IT jobs are definitely moving overseas. One reason is that computer and networking technology is advancing much more slowly than it did in the 1990s and before. Computing is now a predictable commodity, and businesses are keeping systems in place for years longer than they used to. It's odd but true that in a great many midsize businesses, the computers that staffers use at work are much older and slower—often mid-1990s Pentium 133s and 200s—than the ones they have at home. Newer computers don't necessarily translate to more productivity, so older systems remain in place, and entrenched technologies from Microsoft, Oracle, and Cisco are so strong a standard that "BDM" (what we used to call "brain-dead maintenance" twenty years ago at Xerox MIS) can be done in India as easily as here. When technology is turning over every year and new tech means more work per unit time, people need to be on site to evolve technology infrastructure. When technology reaches a plateau, cost savings come from reducing people costs, far more than adding new tech.
  • Few people know this (and the press has been almost silent on it) but the US imports jobs from large firms in Western Europe, where labor regulations make people costs significantly higher—and the deployment of labor much less flexible—than here. The Wall Street Journal said in a recent article (not online except in their fee-based service) that about 220,000 jobs came here from outside the US last year. Europe worries more about intellectual property protection in places like India and China than American firms do. What do Europeans know that we don't? (Ha! Been there. Won't talk about it. Use your imaginations!)
  • Another recent article in the Wall Street Journal pointed out that as American IT jobs are outsourced to places like India, labor shortages are forcing local salaries higher, and the automatic reductions in people costs realized by outsourcing are now much less. Cutting people costs by three-fourths through outsourcing is a no-brainer. Cutting costs by 30% or 40% is not.
  • The Democrats have spotted their Golden Issue in this election year, and are whipping up resentment as fast as they can. Some of the smarter businesses see the targets on their foreheads and are pulling back from further outsourcing, citing a growing backlash that the press (who generally take their leads from the left—about 80% of news reporters self-identify as Democrats) has put up in lights in recent months. Kerry, if elected, is promising to make outsourcing hugely expensive for American firms. American workers are already furious over the robber-baron mentality at the executive level, where some CEOs pocket in a year two or three times what mid-level workers make in a lifetime. Populism seems to be on the rise again. I had thought GWBush was a shoo-in. Now I'm not so sure. Whodathunkit?
  • Skilled trades are in short supply, and a lot of bright IT people are finding work in carpentry, electrical work, and other skills that can't easily be sent overseas. This will change the nature of hourly labor in American in ways we can't yet predict. But consider: The immense number of inexpensive single-family homes built immediately after WWII are now changing hands and are in desperate need of rehabbing. Who will do this work? Hint: A lot of paper MCSEs have bought hammers. Wanna make a few million? Start a remodeling business that employs former IT people.

March 30, 2004:

I saw in the Colorado Springs Gazette today that the Chinook Bookstore, a fixture in downtown Colorado Springs for 45 years, is closing forever on June 15. Owners Dick and Judy Noyes tell us that the decision involved competition from Internet sales and simple fatigue, leading them to decide to retire. They refuse to sell the store to someone else, fearing that new owners will not share their dedication to bookselling and their high standards.

Chinook is a wonderful place. Even though I've only been in town a year, I've dropped a couple hundred bucks there on non-technical books. (Their computer section is very slim, which is OK by me. There's always the Big Borders up at Chapel Hill, or for that really geekoid stuff, Amazon.) They're right across the street from the delightfully loopy Uncle Wilber Fountain, and around the corner from the world's densest kitchen store, Sparrow Hawk. According to the Gazette article, the average tenure of Chinook employees is eighteen years.

If this trend continues, the only places with independent bookstores will be towns too small to attract Barnes & Noble, and towns large enough to challenge them with local independents. Denver has Tattered Cover (the LoDo store is breathtaking!) and most other large cities have similar, locally owned operations, but the size of the town needed to support an independent bookstore is creeping upward.

As a guy in the book business, I mourn Chinook's passing, but I simply don't know what's to be done. It may be true that online book sales mean the death of independent bookstores, but they also mean that really small publishers (like me, if I ever bestir myself to publish the three books I've already laid out) can stay alive and sometimes even make money.

My guess is that if Dick and Judy Noyes were still in their twenties (as they were in 1959, when Chinook first opened) they might be able to fight on. Is it better to bow out gracefully rather than be humilated into bankruptcy by the unforgiving market? Probably. Will downtown be a less interesting place come July? Absolutely.
March 29, 2004:

Mon Dieu, unpacking. When we packed up our Lowden Road outpost at the beginning of 2003, we used a couple of months' worth of hoarded Wall Street Journals to wad in and around things tossed in boxes. Well, here we are (destinated, as the CBers of the 70s used to say) and all that paper is coming out of the boxes where it had been wadded for more than a year.

The pile you see at left came out of only three or four boxes, including the two big ones containing my rotating parts tower. I know that the other 80-odd boxes in my workshop alone will probably cough up enough wadded Journals to reach nose-high and span the full width of my bowling-alley shaped shop.

In puzzling out what to do with it, I hit upon a partial solution: Send somebody something in a big box. So in a few days, my old Clarion friend George Ewing will receive a Compaq Pentium-350 that I rescued from oblivion and degunked, and the machine will doubtless travel in a far bigger box than it needs to. That takes care of the first four boxes worth. We only have about, hmmm, 200 more.
March 28, 2004:

A few odd lots, written on my newly melted-down-and-resurrected Dell Xeon:

  • A relatively rare astronomical event happened yesterday: Three of Jupiter's Galilean moons could be seen casting a shadow on the face of the big planet at the same time. Pete Albrecht (rightly) bugged me to get my scope out and assembled to see it, but come late Saturday afternoon here, it was not only cloudy but snowing. So I (and I assume most of you) will have to be content with Pete's excellent picture of the event.
  • Is anybody else starting to get irritated by the number of drivers and apps that demand a spot in the taskbar tray? In some it's a configurable option; in many it's a given and can't be changed. My now-deceased sound card insisted on mounting its diagnostic utility in the tray, from which I used it exactly once, to confirm my suspicions that the card was dead. NVidia's latest video drivers mount a tray icon, as does Easy CD Creator, Eraser, NAV, WinZip, and two or three other things. This is getting ridiculous.
  • Back in January, the Long John Silver's fast seafood restaurants announced that they would offer free giant shrimp at every US restaurant to celebrate, if NASA's Mars probes found evidence of saltwater oceans on Mars. Well, NASA has made it official: there was once saltwater on Mars, so on May 10, there will be free shrimp at Long John Silver's. Here's the news release with the details. Now, guys, what about that piece of fossilized Martian rotini? (See my March 5, 2004 entry.)
  • Not much more to report. I have most of my electronics books and magazines on shelves in my shop now, and I'm gradually filling gallon milk jugs with parts that used to live in milk jugs back in Scottsdale. (I recycled 65 milk and pool acid jugs in February of last year just before we left Arizona; see my February 8, 2003 entry. The parts all went into plastic bags, which pack much more compactly than gallon jugs.) I use a gallon of milk in about 11 days, but fortunately my new friend George Ott and his family consume two or three gallons per week, so I already have 14 jugs. This evening I went downstairs just to stick my nose in the jug full of tube sockets and inhale deeply. Ahhh. I missed that!

March 27, 2004:

My friend Pete Albrecht showed my comments (and the Salon article link) from my March 23, 2004 entry about the writing life to his writer friend John Tomerlin, who wrote the thoughtful reply below, and allowed me to reprint it in full. John has known some of the best in the writing business, many of them powers in the SF world. The note speaks for itself, and I won't comment further on it:

Both the original piece and Jeff's comments on it reveal the ambiguous nature of writing fiction for a living.  I began writing because my friends were doing it, because my chosen field, radio broadcasting, had been marginalized by television, because there were heroes for us to emulate—Steinbeck, Hemingway, Faulkner—and because the life-style it permitted was appealing.  Today, conditions are very different.  The best best known and best paid authors are not the best writers.  The best writers are supported by teaching jobs, a situation that insulates them from the harshness of real life, particularly the struggle to make a living writing.  For those of us who took the writing life seriously, the size of the advance was never the most important thing.  The most important things were, in order, (1) to be published at all, (2) to be paid for what we were doing, (3) to find our work on the shelves of libraries, and (4) to receive recognition, either from reviewers or letters from readers, for what one had attempted to do.

Some of my friends made money—Matheson, Beaumont, Bradbury—and some didn't—Nolan, Johnson, Sohl.  Myself.  But we all lived the life, taking chances, going hungry at times, having the lights turned off or the phone disconnected during our early years--until gradually, step by step, we found different ways to make it work.  Not one of us, so far as I know, would have traded our lives for all the money in the world, or wanted to have written a Stephen King book.

Not that we regarded ourselves as great writers; but each of us thot he had something to say, and wanted to say it in his own way.  Today, I'm not sure any of us would have made it.  The small markets—pulps, genre novels, magazine short stories—are gone, for the most part; replaced in most instances by highly stylized, action- and not character-driven forms.  The anthology market in television is gone—once represented not only by such series as Twilight Zone, Alfred Hitchcock Presents, and Thriller, but, oddly enough, the series westerns, thru which you could attack any social issue you pleased, from capital punishment to gun control.  Freelance writing for TV is a think of the past, having been replaced with staff writers and hyphenates (writer-story-editor, writer-consultant, etc).

Some time during the years I was growing up and making my living as a free-lancer, the popular advice to would-be writers became, "Don't give up your day job."  That is certainly true today for those who hope to publish fiction.  Back then, you not only gave up your day job, you gave up a normal life, in order to travel, socialize, drink, fuck as many women as possible, and in general try to learn what life was all about.  When you wrote, if you were honest and if you were any good, you didn't worry about who was going to publish or how much money you were going to get.  You worried about if someone would publish.

I feel as sorry today for young writers inspired to write honestly and well, and make a living at it, as I am by anyone who might have loved to race a car they could afford to buy and tune themselves and drive to the track (and home again, with a bit of luck), then use during the week for basic transportation.  All that is gone, probably for good.
March 26, 2004:

Michael Covington's Web diary has moved; bookmark this link. (The link I have above is now correct, though I apologize for being slow to fix it. I'll be slow to do a lot of things for a couple more weeks yet...)

In his March 21 entry, Michael alludes to an extremely critical issue: How we are to know when we really understand something. A lot of people are walking around with complete confidence that they know what they're doing and what they're talking about, when their understanding of important things—including their jobs—may be as thin as paper. I used to run into this when doing job interviews at Coriolis. I learned quickly to get a sense for deliberate fakers, but it took a little more time to develop a sense for people who are honestly mistaken about their own competence. The hardest of all to detect were people who, like me, have a knack for talking and writing clearly, and sound convincing when they put something across, even if all the underlying facts and relationships are jumbled. (Michael calls this "syntactic thinking", meaning that the form is perfect, even if the semantics are completely hollow.) You can give an editing test to copy editors, but managers, yikes—I had a lot of trouble separating wheat from chaff.

We don't teach critical thinking in school anymore, and we're so afraid to call people on their mistakes that our whole culture reinforces false or (I think more accurately put) mistaken understanding. Note that I'm not talking about New Age gobbledegook here (channeling, chakras, and so on) but people who can actually talk a good game about Real Stuff like managing a production department without actually having any deep understanding of the issues involved.

I myself am not immune, though at least I question my own expertise. I've been wrong about a lot of things that I thought I understood, generally when I had too little time to research and practice the knowledge in question. Most recently, I thought I understood sockets programming, but then when I sat down to do some yesterday I realized that (almost) everything I knew was wrong. I then knew I had better go back to Square One and start over. I'm real damned glad that all my tech books (at least) are out of boxes and on the shelf again.

My personal touchstone for evaluating my own understanding is simple, and well-known among writers and teachers: You can't be sure you truly understand something until you can explain it clearly to someone else. Without that, well, it's still an open question.

By the way, Michael's diary is worth seeing for the sake of the spectacular planetary photos he's been posting, taken with an 8" telescope. My friend Pete Albrecht is doing the same sorts of things these days (I've posted some of his photos here) and if I could squeeze another 8 hours out of every day I'm sure I'd be doing the same myself.
March 25, 2004:

We're in the process of cleaning our rental so we can get our security deposit back, and one of my challenges was to get the oil spots off the driveway, where the 4Runner mysteriously leaked from somewhere the Toyota techs couldn't find. I went to Home Depot and asked the guy running a forklift around Janitorial what he could sell me that would get oil spots off concrete. He told me he had some stuff, but a better thing to do was go to Safeway and get the cheapest kitty litter I could find. He told me to use a short piece of 2 X 4 to grind kitty litter into the oil spots, and then when I swept the dust away, the oil would be gone.

This seemed a little fanciful to me—those oil spots were months old—but I gave it a shot. Two remarkable conclusions:

  • He was right. I'm not entirely sure how it works, but when I swept the dust away, the oil was gone. Did the kitty litter soak it up? Or did the kitty litter act as an abrasive and simply grind away the top (stained) layer of concrete? (Or, as my paranoid side began to worry, did the kitty litter dust simply adhere to the oil and make the oil spots the same color as the concrete...for the time being?)
  • He gave up the chance to sell me something expensive because he knew of a cheaper way to do it. God love 'im—I didn't think people like that were still around.

As you can tell, we're making some progress. We sold our moving boxes this morning, and I did a total meltdown and rebuild on my malfunctioning Dell Xeon this afternoon. In the process I discovered that the sound card was stone dead, and this may have been behind a lot of its earlier lockups and other failures. I won't know until I torment it for a few days, but so far so good.

I hope to have some more significant things to say in this space soon. Bear with me.
March 24, 2004:

Little by little things are returning to normal here. Garbage collection began today, and if the sound of the garbage truck ain't the sound of home, what is?

Carol and I noticed an interesting and still-mysterious trend involving our move here: A lot of our electronics went berserk, and some of it died. As the move was only 1.68 miles (less as the magpie flies) and everything worked fine at the rental house, ya gotta wonder. Viz:

  • Our treadmill went nuts the first time I turned it on. The motor started out normally, but then went to maximum speed before I yanked the safety key. Now it won't start at all, and displays an information-free error message on the LEDs. As it's almost 10 years old and a little ragged, I think we'll junk it and get a new one.
  • Our brand-new Sleep Number Bed has a wireless remote, but half the time, when you punch one of the buttons to inflate or deflate the bag, it locks up and displays "Err".
  • My Dell Xeon ran for a day or two here, and then last Friday it refused to boot. I had to repartition the drive and reinstall everything from scratch, a process I'm still not done with.
  • My Dell Dimension woke up last Thursday with all of its icons moved over to the left margin in one huge mass. No other damage that I know of.
  • My Linksys BEFSR41 wired router has apparently lost its ability to hand out local IP addresses, even though its DHCP server is enabled.
  • Carol's Nokia cellphone died while charging last week, and wouldn't turn on no matter what we did. Then, after three days, it returned to life. It won't say whether it descended into Hell in the interim, though I sternly reminded it how close it came to descending into the trash...
Apart from the new bed, we had all this stuff at the rental, and it all worked. It would be just too tinfoil-hattish of me to point out that we now live less than a mile from the NORAD entrance, but I'm out of other theories. More as it happens.
March 23, 2004:

Alana Foster sent me an interesting link to a piece in Salon, the gist of which is a complaint about the current state of book publishing, by a "midlist" author. Midlisters are those who do well in publishing without hitting the NYT bestseller list. "Well" in this context includes critical acclaim and awards and possibly even money, but not the sort of national recognition that, say, Tom Clancy gets.

I'm of two minds about the article. Almost all of what she says about the current state of publishing is spot-on (it's ever more a corporate bean-counter's culture that insists on blockbusters and "selling more of what sells") but her gripes about her success in publishing blew me back in my chair a little. She managed to make a living as a full-time writer for ten years, during which time she received a $150,000 advance, followed by an $80,000 advance, followed by a $25,000 advance (which to her is a pittance) along with numerous other paying assignments. For all that she is bitter about the writer's life and seems to consider herself something of a failure.

Wow.

The largest advance I have ever gotten for a book is $25,000; I did receive one of $15,000, but all the rest were in the $5,000-$8,000 range. I consider myself a smashing success, at least in nonfiction, even though I've worked a day job during all but a few years of my writing career.

I will admit, in agreement with the author, that the view looking forward is not as good as the view looking back. I wouldn't have had any trouble selling my novel ten or fifteen years ago, and there were a lot more magazines back then paying significant money for articles of all kinds, fiction and non. The industry has changed, and the sorts of "easy contracts" she got ten years ago probably aren't coming back.

Authors have to recognize that to be successful going forward, they're going to have to think outside the box and do a lot more work that isn't exactly writing. My suggestion to the author in question is that she become her own publisher. She has a community of adoring fans and knows how to reach them, which is key in self-publishing. My guess is that she could make a better living promoting and selling her books on her own than she's currently making from the big NYC presses. She seems peculiarly hung up on Being a Writer, an attitude I see a lot and don't respect much. The point isn't to be a writer. The point is to sell books and make a reputation. If you have to be a writer/publisher/PR flack/salesman/whatever to get there, well, that's what it takes to get there and you'd better start putting your shoes on. Economical short-run printing and the Internet make things possible today that weren't possible ten years ago, and they will only get cheaper and easier. As soon as we get settled here, I'm going to get back into it myself. We'll see how it goes.
March 22, 2004:

My cousin Rose and her friend Al stopped by for a couple of days on their drive through Colorado, and I showed them all the pictures I had taken during the construction of our new house. I did it in an interesting way: I set up my little EZGo PC on a card table next to our big TV, and ran an S-Video cable from the PC to the TV. I then popped a CD containing all the photos into the EZGo, and put my photo manager into slide show mode. Each click on the left mouse button brought up a new photo, and the TV displayed them all very clearly.

I couldn't help but think back 40 years ago, to the regular family sessions we had in our livingroom, looking at all the slides my father had taken over the years. There was ritual to it: He set up the slide projector and the screen and arranged all the boxes of slides within easy reach (this was before carousels) and we went through them while munching popcorn. It was a fine way to cement family memories. In fact, many of my memories of our family vacations from the early 1960s center on images my father captured on slides. I doubt I would remember today what Great Aunt Tess looked like (she died in 1964) if not for seeing her every couple of months on that spider-legged screen in the middle of our little living room for years thereafter.

Next time I'm going to use a Wi-Fi media adapter like the Linksys WMA11B, which will allow me to bring down digital camera photos from my file server without having to set up the EZGo as a "projector." We'll lose a little ritual, I guess, but it'll be way easier, and I suspect we'll do it more often. In fact, I'm in the process of scanning my father's old slides, so we'll be recapping those 60's family slide show sessions in a way that might astonish him—but I think he would be pleased.
March 19, 2004:

Carol put an ad in the local paper to sell our moving boxes, and so we've been frantically unpacking boxes so as to get as many flattened out and ready to sell as possible. The ad will run for two weeks, taking us up to the time we surrender the rental house, which is where we'll be selling the boxes. (There's no room for flattened boxes here, trust me!)

Certainly when you open boxes packed long ago, you will find interesting things. Well, yesterday I found something that floored me completely: a 50,000 word fragment of a computer book that I had written and then completely forgotten about.

I mean, completely. I looked at it, scratched my head, and wondered, whenthehell did I do this? Now, my memory is legendary, and I could catalog all the multitude of unfinished SF novels I attempted when I was in college, at least one of which represented close to 80,000 words of text. But Object-Oriented Programming From Square One ran out the back door of my head and didn't even wave g'bye.

The box it was in had been packed in late 1993, just before we moved to our house on Lowden Road in Scottsdale. The box was full of old paper manuscripts of nonfiction I had written in the 80s. I had never opened it after that, and in truth, most of it isn't likely to be useful. The DOS era is over, heh.

After flipping through the manuscript, some of it started to come back to me. I worked on it from early 1989 to the end of 1990, and a lot of it was adapted from material I had published in Dr. Dobb's, Turbo Technix, and PC Techniques. I was inspired to do it by the research I had done in order to write the OOP Guide that shipped with Turbo Pascal 5.5 in mid-1989. The first 40,000 words was a fairly high-level discussion of the ideas behind object-oriented programming, and after that I started some real code examples from Object Pascal and C++. I think that's what stopped me cold: The two languages don't understand some OOP ideas in the same way, and after another 10,000 words I decided to bag it.

There may be a way to teach programming language concepts without actually showing any code, but I have my doubts as to how effective any such method might be. (Even the formidable Don Knuth invented an imaginary language to demonstrate algorithms in his magnum opus.) I learned programming concepts by beating my head against numerous languages, and in looking back, I think having used so many helped me write better code, and also helped me learn the next language faster. I thought Pascal would be my last language, but I may learn one more before I die, and that will be C#. All the others—FORTRAN, FORTH, 8TH, DACL, ABL, BASIC, APL, COBOL, Smalltalk, Tcl/Tk, Ada, Modula 2, x86 assembly, plus enough C and C++ to make me gag—will help me in that task. But could I teach the concepts embodied in all of them without actually mentioning any of them? Interesting question. If there were a market for such a tutorial, I might try it. But there isn't.

Now, there are another two hundred boxes in here. Gotta get back to work. I'll let you know if I find anything else interesting.
March 17, 2004:

[8:30 PM] It was more like 48 hours, but I'm back.

And I'm living in my new house. We have the rental for two more weeks, but it's empty now, and just needs the cleaning service to come through so we can be shut of it.

As for the move proper, it went extremely well. The weather was perfect—60° and sunny—and the whole thing took less than eight hours. The crew of four guys didn't bang up anything, didn't break anything, and didn't even finish the pizzas we bought for them. The moving company was Graebel Moving, and they were stunningly efficient. They were the low bidders, too. I recommend them highly, though as Carol sagely points out, most of it hinges on the quality of the guys who actually come out to load and empty the big truck. Did we just pull a good crew at random? Or are they all that good?

Somebody else can pursue the experiment. I've had it with moving for a good long while.

Now (lest you think I'm sitting around with my feet up) once the truck leaves, the real work begins. We now have a very nice house that is crammed to the rafters with boxes. The master suite and kitchen are (more or less) functional, and we put the guest room together this evening in preparation for a visit this weekend from my cousins Rose and Al. The garage, my office, Carol's office, and my workshop are still box-choked and non-functional. I've had broadband here since March 5, but didn't even use it until this afternoon, when I finally uncrated one of my computers and got things wired in the corner of my new workshop where they didn't stack any boxes.

We are now facing a labor of several months. It took most of a year to get settled in at our last house in Arizona, much of which was due to hidden defects in the structure that had to get fixed. We have a few defects pending correction here (primarily two of the pocket doors that showed up warped and were on order for two months) but most of the work will be some shelf-building in the basement and a great deal of sifting and tamping. We'll be gathering rummage for St. Raphael's rummage sale early in June, and figuring out where things go by the simple heuristics of living here for a while and seeing how things fit.

Looking forward, we have the challenge of hiring a landscaping company to make the land around the house look nice, but in truth, most of it will be emptying boxes and getting rid of stuff we should have gotten rid of long before we left Arizona.

I'll deal with that as it comes. For now, we're home, and boy, you don't know how good it sounds to say that!
March 15, 2004:

We move tomorrow. Boy, this was a long time coming.

Anyway. I'll be backing everything up and taking the machines down sometime this afternoon, and will probably be out of computer touch for 36 hours or so. (Horrors!) So if you need to reach me urgently, email is not the way to go until sometime on the morning of the 17th. We're in the phone book, and phone service has already been connected at the new house.

We'll both be real damned glad when this is over.
March 14, 2004:

I've been keeping informal records of spam numbers here since January 1, and although I don't have three months' data yet, there are some interesting trends that are good for a short entry:

  • My spam count has grown only slowly in that time, though there have been some weird (and temporary) upticks. In early January I was getting about 500 spams per day. I'm now getting about 530. That's an increase of roughly 2% per month, which isn't much.
  • However, the number of spams that my filters miss is about 10% - 15% per day. In other words, even though I religiously block every spam that sneaks through, an average of 12% come in with new domains or phone numbers that I haven't seen every single day.

Because the absolute number of spams is almost constant over time, this means that 12% of the domains I see every day are new, and almost 12% are (by implication) abandoned. The problem here, of course, is that I don't know, without some sort of automated tracking system, which domains are abandoned, and the size of my blacklist of domains is growing phenominally. (I have blocked over 3,800 domains in the past 2 1/2 months.)

I block domains both in the "from" field (when I know they're spammer domains; believe it or not, I check each one!) and all those embedded in the body of the message, pointing to the web-based pitch. This domain is what I call the "payload." I also block phone numbers on the 1% of spams that arrive with phone numbers as the payload, and on any domain pointing to an "unsubscribe" screen. I used to block on strings in the subject field, but I don't even bother anymore, as almost no spams arrive with usefully blockable text in the subject lines.

The conclusion? Blocking spam based on sender domains or on the payload is increasingly futile. Spammers are buying domains in bulk and (as far as I can tell) using them only once. Spot checking domains with ping, I see a relatively small number of IPs at the other end of a huge number of domains, implying that a client-side black hole would be a really useful thing to build, and when time allows I will try. But mostly, I conclude that filtering by keyword (of whatever sort or emphasis) is a dead issue. Nobody's going to do what I've been doing since January, to still have 15% of my spam slip through. Something else will have to be done—and my next book will summarize what I've learned and provide some advice.
March 13, 2004:

If any of you have links to Jim Mischel's Web diary on your sites, check them. Jim changed hosting services recently and his new service is case-sensitive. My link croaked because of the uppercase "D" in "Diary." The valid link is http://www.mischel.com/diary/index.htm.

While we're talking about Jim, his recent comment puzzling over why politicians fawn over their ideological fringes instead of courting the undecided middle got me to thinking. There are two reasons, actually:

  • Traitors are punished more severely than enemies. When you get out to the ideological fringes, anger rules supreme and rationality is at best vestigial. A Democrat who were to say that perhaps the war in Iraq was justified would lose votes on his left fringe...and a Republican who were to say that abortion should remain legal would lose votes on his right fringe. Ideologues either don't understand or are too angry to care that staying home on election day is a vote for the other side, and politicians know this.
  • The fringes control most of the money. Some big spending extremists (like Barbra Streisand) are smart enough not to keep their wallets in their pockets if a pol violates one or another partisan sacred cow. A lot of them are not that smart, and the sad thing about American politics is that money rules. The undecided middle, almost by definition, don't contribute much money to political parties, because they haven't decided which one to back, or else (like me) cast every vote on the issues and not by party. (Some business interests have been known to cover both sides, but individuals rarely do.) Your slobbering vote-slaves are the ones who finance your campaign, especially with regulations like McCain-Feingold making small individual donations more important than large, heavily regulated ones.
All that being said, pandering to the fringes does lose votes in the middle, if the middle really cares about the issue at hand. Some time back there was something in The Economist indicating that independent voters tend to be moved strongly by economic issues (including health care issues, which are very much economic issues these days) and almost not at all by moral ones. So Jim's right: Bush condemning gay marriage is a no-lose proposition for him. Where the war fits in I'm not sure; I think the middle is less interested in it than most of us think. The big issues are money & job issues: Health care, taxes, jobs fleeing to India while CEOs pocket ten million a year, in good years and bad. That's where the real battles will be fought, but not until we're a lot closer to November. In the meantime, Bush-bashing on the left and gay-bashing on the right will be very much in style, just to keep the fringies whipped up and forking over the cash to their respective parties.
March 12, 2004:

I'll be a little sparse for a while yet, what with the move and reading page proofs. In the meantime, some odd lots:

  • The big truck comes Tuesday to get our furniture and heavy stuff up to the new house, and in the meantime we're sealing grout, bringing up the fragile stuff, and getting some new furniture delivered. We now have phone and cable service up there. My lathe was moved scant hours before the snowstorm that dumped a foot of sloppy white stuff in our neighborhood a week ago. (It was gone in all but completely shaded areas in about four days.) We have our new mission oak bedroom set and I have my oak desk and computer table. We have begun hanging wall art. By St. Patrick's day, we will be living there.
  • My old friend Bob Halloran from my GT and Starside Engineering days (which are going on 25 years ago, sheesh!) sent me a link to a place that sells steel saucer sleds. While I intend to order one and make a Wi-Fi dish antenna out of it, paying $22 for a sled and then spending a weekend turning it into an antenna seems to deliver a lower ROI than simply shelling out $45 or so for a commercial parabolic grid antenna, like those sold at FAB. They sell a number of parabolics in the 15-21 dBi gain spread for $43-$60 each. Better to do the garage sale circuit in spring to pick up a sled for $5 and just pound the dents out of it...
  • Slashdot aggregated a really good report on Michael Icaza's Mono project, which is an implementation of those .NET APIs that Microsoft has licensed via ECMA under "reasonable and non-discriminatory terms." Because Mono is now owned by Novell, patent licensing for Mono, should that become necessary, has a path and a benefector. I am betting that the standard "Linux desktop" for corporate environments will be Ximian Desktop 2, which is built over Mono. Take especial note of the nascent C# development IDE those guys are working on. Once my own Linux desktop research kicks into high gear late this summer, I'll be playing around with XD2 a lot, and I'll record my progress and comments here.

March 9, 2004:

Whew. I just now sent in the last little bits of the second edition of my Wi-Fi book—a major reason I've been so quiet here for the past few days. As with any book, the final 10% is pure torment, and with us trying to get ourselves moved at the same time, well, yeech. Next time I need to plan this all a little better.

We'll be off press with it about April 3, maybe a little sooner. You'll notice that the title has changed, and the reason may surprise you: It was too long before the gist of the book appeared in the title. You get only so many characters in the various book databases, and "Jeff Duntemann's Drive-By Wi-Fi Guide" was often truncated to something like "Jeff Duntemann's Drive-By Wi". We needed to get all of "Wi-Fi" in there so people would know what the hell the book was about. I will admit that we got some raised eyebrows over the phrase "drive-by," but that was deliberate, to get some attention and appeal to a specific audience: wardrivers. The book was the first to cover wardriving in any detail at all and in a positive way, and the title and the cover design (with a guy in a hat driving a Maserati—Kris, you overestimate me!) was an allusion to that. The wardrivers loved it, but I want to reach a mainstream audience as well, one that may think of "drive-by" strictly in terms of unwired ne'er-do-wells.

The book didn't change a great deal internally. My emphasis in the update was on Wireless-G and WPA security. I went through the whole thing and made changes all the way along, but the bulk of the new material involves WPA security and the newer IEEE standards, including 802.11g and 802.11n, which will eventually become "Wireless-N." My soup-box antenna made it in, as did my booze-can antenna. Maybe with the next edition I will include my saucer sled antenna—as soon as I can find a saucer sled. They're not the ubiquitous childhood winter icon they were forty years ago, alas. I guess what I need is a used wok. Time to start hitting the local garage sales...
March 5, 2004:

Some odd lots while waiting for the machinery moving man, or someone like him:

  • Michael Covington now has a Web diary. Michael, being one of the brightest guys on the planet, is well worth reading on a regular basis. Bookmark it here.
  • Carol and I have moved upwards of 150 boxes to the new house, and spent most of yesterday sealing grout. The lathe goes up today, and our new furniture will be delivered tomorrow. This is getting serious, folks.
  • Time reports that 45.5% of Americans now self-identify as Republicans, up from 40% ten years ago. 45.2% now self-identify as Democrats, down from 49% ten years ago. We really are dancing on a razor here, but if I were a Democrat I'm sure I'd be worried.
  • There may not be life on Mars, but there evidently is Italian food. Something that looks a lot like a piece of rotini has been seen by the Mars rover Opportunity. See the image here. Given that NASA is now convinced by chemical analysis of the Martian soil near Opportunity that a lot of water was once sloshing around that part of the planet, the prospects for life are now much better. To avoid the inevitable conclusion that Mars is an Italian colony, let's keep looking. Maybe we'll spot a pierogi, or a knish.
  • While we're talking fringe science, Slashdot reports that there's a new kind of desktop fusion research going on, and it seems to be getting some results. Here's the NYT article; you have to register for it. (I have, and there have been no incidents.) Unlike the ill-fated Pons and Fleischman variety of "cold" fusion, the mechanism here is to heat gas bubbles fantastically with ultrasonic energy until the hydrogen fuses. Or something. Let's just say I'm watching closely, but the champagne is still in the pantry on the high shelf.
  • Ben Oram sent me a pointer to Whois Source, which has the kind of multiple-domain reverse DNS that I was asking about in yesterday's entry. Try this query. Unfortunately, they charge what I consider a lot ($15/month) to see more than three domains per IP, so using it to increase my spam capture rate by a few percentage points doesn't seem like a good deal. I think I may just press on with my plan to implement a client-side IP black hole, as soon as I get moved and some time opens up.

March 2, 2004:

My most troublesome spammer is Aphrodite Marketing, and I'm trying to figure out how to deal with him. I know his name, address, and phone number, not that I really want it. Screaming at him won't help; sensitive souls do not go into the spam business, or if they do, they don't stay long. No, I want to find out what all his registered domains are. He has dozens, or maybe hundreds, and I almost never see the same one twice. I have blocked over 50 already. They all resolve to the same IP: 69.6.68.11.

If there were a way to resolve domains within a mail client (and thus create a client-side black hole under my control) I'd be shut of him, at least until he gets a new IP. But if he keeps registering domains in bulk and using them once or twice, he wins.

I thought I read somewhere that if you have a spammer's nameserver address, you can do some kind of lookup and find out all the domains handled by that nameserver. The nameserver listed in whois is

a.ns.joyrunning.com 69.6.68.10

I've tried various things without success. Any suggestions? A typical Aphrodite domain is numberfound.com. Another is competitiveoath.com. They are legion, and until I can write a utility to block mail based on DNS lookups of IP addresses, he will continue to be a problem.
March 1, 2004:

As you might imagine, I've been thinking a lot about moving logistics recently. The move has begun with a vengeance: I have already moved over 100 boxes up to the new house, including 50 cartons of books. A couple of my friends have told me this is a bad use of my time (and my back) but there's really a method to this madness.

First of all, we need to get as much stuff as possible out of this tiny rental so that the rest of it can move out more easily. We moved a 3400 square foot household into a 2000 square foot house, and it's kind of tight in here. Moving books was an obvious choice, because we have built-in bookshelves at the house capable of holding at least 1200 books. (Carol and I own about 1800-2000 books. I've never done an accurate count.)

I also have to get enough stuff out of the cram-packed garage here to allow a machinery moving company to take my lathe and bench up there this Thursday. (The main move won't happen until 3/15.) This meant hauling boxloads of electronics parts down into my workshop, and rearranging a lot of the other stuff that I can't move on my own. The other 50 boxes I've taken up there have been electronics stuff: Parts, test gear, radios. Last week I set up 12 feet of Turnkey modular wall shelving in my workshop, and the remaining six feet in the garage for lathe tooling, metal stock, and so on.

It's now loose enough in here to get furniture out without bloodshed, and so I can relax a little and concentrate on finishing my Wi-Fi book revision, which seems to be going on an asymptotic curve. That's the way a lot of projects go. About half the book is already typeset, but I still need to finish the two critical chapters focusing on WPA.

So I've been busy. When Carol and I were too beat to cook yesterday and went down to China Wok to get some chow fun for supper, my fortune cookie was most pertinent: "You have had a good start. Work harder!"

Gee, thanks.