Week of 19 February 2007
Update: Friday, 23 February 2007 10:03 -0500
Busy weekend. I got the UPSs swapped out, although I haven't popped the
lid on the old one to clean it out. We had dinner with Mary and Paul
Saturday at the Chinese place. During dinner, we asked them if they'd
like to go out to the pistol range and shoot. They wanted to give it a
try, so we went out to the range yesterday afternoon, burned about
650 rounds of ammunition, and then headed over to the Mexican place
next door for dinner.
Paul had never shot a pistol before, but as it turned out he's a
natural. He started with my .45 ACP Colt Combat Commander, and put two
or three in the bull and the remainder of his first magazine mostly in
the 7 to 9 rings. He also did pretty well with my Ruger .357 revolver.
He started at 15 feet, which doesn't sound like much to anyone who's
never shot a pistol, but believe me when I say that most beginners have
trouble even keeping shots on the paper at 15 feet, let alone putting
them in the bull. Paul soon moved back to the 25-foot range, and later
to 50 feet, which is fairly long range for pistol shooting,
particularly for a beginner. (At 50 feet, I can keep all or nearly all
rounds in the K5 area of an FBI kill-disable silhouette target, but I'm
not going to hit a lot of bulls, particularly when firing combat pistol
style, double-tap and so on.)
Mary started off a bit slower. She'd shot a 9mm pistol a few times in
graduate school, but that was 10 years ago. I let her start with the
.45 ACP, but firing full-bore .45's is a bit much for someone who has
small hands and is less than half Paul's or my mass. I moved her
over to the Ruger .22 bull-barrel target pistol, and she immediately
started doing much better. By the end of the session, she was hitting
the bull regularly. I think her rapid improvement had less to do with
the lower recoil of the .22 than with the fact that the Ruger target
pistol fits her hands much better.
I told Mary that I've taught a lot of people to shoot pistols over the
years, and in terms of natural ability she's definitely well toward the
right end of the curve. She did better sooner than about 90% of the
people I've introduced to pistol shooting. After watching her practice
for only an hour or so, I sure wouldn't want her shooting at me.
Tuesday, 20 February 2007
Oh, yeah. I forgot to mention yesterday that Brian and Marcia Bilbrey
had driven down from Bowie, Maryland to spend the weekend with us. We
had a very relaxing break. Brian also went to the range with us, and
shot extremely well. These guys worry me.
More bad news for folks still stuck on the Microsoft merry-go-round.
Steve Ballmer, faced with poor sales of Vista, has announced that Microsoft will crank up WGA this year.
"[Ballmer] said that there was software within Windows Genuine Advantage which
could be dialled up by the Vole to make it even more tougher for the
pirates to take out.
He promised that the Vole would "really ferret through how far we can
dial it up, and what that means for customer experience and customer
The real problem with Vista, of course, is that no one in his right
mind wants it. Think of it as a service pack for Windows XP that
has no compelling new features, breaks many current programs and
utilities, lacks drivers for many recent and current peripherals, adds
unwanted DRM "features", and costs a lot of money. And Microsoft is
surprised that people aren't rushing out to buy Vista? Duh.
On what may be a related note, Dell is surveying customers about what they want. When I checked this morning, I found the following were the top five requests:
1. Pre-Installed Linux | Ubuntu | Fedora | OpenSUSE | Multi-Boot
2. Pre-Installed OpenOffice | alternative to MS Works & MS Office
3. NO EXTRA SOFTWARE OPTION
4. linux laptop
5. No OS Preloaded
Which makes it pretty clear that the old argument about OSS software
not being available on new systems because there's no demand for
it is entirely bogus.
Someone asked me about AMD's new Athlon 64 X2 6000+. Short answer:
don't bother. This is a stopgap processor for AMD, until they can get
their 65 nm line rolled out. The 6000+ is a 90 nm part that consumes
125W, runs loud and hot, and is beaten in performance across the
board by less expensive Intel Core 2 Duo processors.
It's funny to read the reviews from sites who obviously want to find
something nice to say about the 6000+. Presumably they're afraid that
if they review it honestly AMD won't send them any more samples.
Everyone seems to be comparing the Athlon 64 X2 6000+ to the Core 2 Duo
E6700, probably because they are both priced at about $500. But the
E6700 simply blows the doors off the 6000+. A better comparison would
have been with the $300 E6600, which still beats the 6000+ in most
benchmarks, but at least not as badly as the E6700 does.
If AMD had priced the 6000+ in the $275 range, it would have been a
competitive processor. At nearly twice that, it's not even in the
Wednesday, 21 February 2007
I'm hard at work on the astronomy book, so postings here are going to
be kind of sparse for a while. Until I complete the astronomy book,
I'll work five days a week on it, and do some work on weekends on the
home chem lab book. I also need to think about my next project, which
may or may not be another book.
Computer books seem to be a dying category. One of my contacts tells me
that Barnes and Noble stores now devote less than 20% of the floor
space to computer books that they did five years ago, and five or ten
years from now it's not unlikely that they'll have reduced floor space
for computer books to almost nothing.
There are many factors contributing to this trend, but the Internet is
certainly the major one. People who would formerly have bought a
computer book now just look up the information they need on Google. The
effect on computer book sales has been dramatic. People who used to buy
half a dozen or a dozen computer books a year now buy one or two, if
that. Many authors who used to do well writing computer books have
moved on to other things, because they're now unable to earn a living
writing computer books.
And it's not just computer books. Non-fiction book publishing in
general is under the gun, and even fiction publishers are starting to
feel the bite. People are reading fewer books nowadays, both those
books that are for enjoyment and those for practical purposes such as
learning about computers. Magazines and newspapers are hurting badly,
with an increasing percentage of their readership (and advertising
revenue) being claimed by web sites. Many smaller towns and even some
not-so-small ones no longer have a daily newspaper, and most magazines
are a pale shadow of what they once were, both in size and circulation.
I suspect that printed books will become increasingly rare over the
coming years, replaced mostly by electronic books. It costs a lot of
money to produce, print, and distribute a printed book, not to mention
dealing with the inevitable returns. As the average sales per title
continue to fall, I think it's likely that at some point it will become
very unusual for a book to appear in printed form. Eventually, only
novels from the top authors will be printed. Bookstores will disappear,
and you'll have to buy your printed books at Wal*Mart.
By then, most people will have a book reader, which will probably be a
device about the size and shape of a paperback book, with a
high-resolution color LCD screen, wireless connectivity, a small
keyboard for entering search strings and so on, and enough storage to
contain 10,000 or more full books, including graphics and images. A
typical book reader will also store a lot of movies and music, of
Given that DRM does not and cannot work, we authors will have to get
used to being paid by only a tiny percentage of our readers. Payment
will become essentially voluntary, and we'll have to hope that enough
readers realize that it's in their own best interest to pay for
material they can download for free. I'd guess that perhaps 1% of them
will pay voluntarily. (About 1% of the readers of this journal
Although 1% sounds like a small number, it's not necessarily
catastrophic for authors. For example, if I publish an e-book that is
downloaded by 1,000,000 people of whom 1% pay, that translates to
10,000 buyers. If each of those buyers sends me $5, that's $50,000 in
revenue for that book. If, say, Pournelle and Niven e-publish a new
novel that's downloaded by 10,000,000 readers, they stand to recoup
$500,000 from their book.
Of course, for many authors the numbers will be smaller. A new mystery
novel by a mid-list author may get only 250,000 downloads and generate
only $12,500 in revenue. Conversely, a very specialized e-book targeted
at a small niche market may be downloaded only 10,000 times, but get
50% voluntary payments at $20 each for $100,000 in revenue.
This new model disintermediates publishers entirely, of course, but it
opens new opportunities for authors. Most authors will be horrified at
the thought of only 1% of their readers actually paying them for the
privilege, with 99% freeloading. That's certainly one way to look at
Another way is to look at the expected value of each copy. Using my
example of 1,000,000 copies and $50,000 in revenue, that means that
every copy that is distributed by whatever means is worth $0.05 to me.
I just have to get used to the fact that most of my readers won't be
paying me, and hope that the number who do pay me will suffice.
Thursday, 22 February
I'm spending some time this morning cleaning out my inbox. It had
gotten up to more than 60 messages, all of which require some action,
if only a quick reply. Most of them are administrative stuff from
O'Reilly, requests for me to do interviews, and so on. Some are new or
for which I thank you and apologize for taking so long to process.
Quite a few are comments on what I wrote yesterday, like this one.
From: Jan Swijsen
To: Robert Thompson
Date: 21 February 2007 17:05:38
> I just have to get used to the fact that most of my readers won't
> be paying me, and hope that the number who do pay me will suffice.
On the other hand foreign readers
will get new books easier. None of your books are found in the book
stores around here. For a start 90% of books, even computer books, sold
here are translation, only very successful books or books from famous
authors or highly trend following books are available in the original
language. Sure one can order a specific book but then the cost rise
considerably (50 to 100%) so why do that when another one is available.
If your books become available on
line I might well download one (or more). And a payment of 5€
would be forthcoming. 5€ would be about 10% of the normal price I
would have to pay to get one of your books here.
That's interesting. Our books have been translated into numerous
languages, including French, German, Italian, Spanish, Russian, Polish,
Japanese, Chinese, Korean, and several I don't recognize. For major
markets, we get half or less our usual royalty rate on sales of
translated editions. For minor markets, O'Reilly sells the translation
rights for a lump sum and splits the proceeds 50/50 with us. That
usually amounts to pocket change. Our share is often under $50, and the
largest payment I remember getting was something like $250.
I'm currently exploring the details of PDF publishing. I can produce
PDFs directly from the OOo word processor, which I think have all the
features I need. For obvious reasons, I need to lock up the PDF to
forbid modification of the original text and images. Otherwise, I want
the PDF to grant the user all functionality, including copying and
printing text, and so on. I'm not a hypocrite, so there will be no DRM
whatsoever, although I plan to stamp each purchased copy with the
purchaser's name on the title page, with a note that explains to others
who have a copy of that PDF how to buy their own copy.
I need to do some thinking about formatting, fonts, and so on. Reading
most PDFs on screen is annoying. You scroll down the page as you read
the left column and then have to scroll back up the page to read the
top of the right column. There's lots of wasted motion, kind of like
reading a typical blog. The obvious solution to that problem is to
format the PDF landscape, so that one entire page of the PDF fits on
the screen without scrolling. Of course, that raises issues about
aspect ratios, both of the screen itself--4:3 or 16:9?--but of the
effective aspect ratio if the reader is not displaying the PDF full
I'll also be looking into payment methods. I'm convinced that the
reason more people don't pay for freely distributable electronic
books has nothing to do with readers not wanting to pay the few bucks
the author is asking. It's more likely that it's just too much trouble
to get the money to the author. If I make it easy to pay, I believe the
percentage of readers who will pay may be double or more what it would
For example, if I find an electronic book I want on an author's web
site and that author asks me to send a check for $5, I'm
probably not going to bother. It's too much hassle to get the
checkbook from Barbara, write a check and mail it to him, and then wait
to get the book. On the other hand, if he has a PayPal link, I'll click
on it to send him the $5 and download the book immediately.
Similarly, if I find that book on a torrent and download it, I'm
probably not going to bother sending him a check if that's the only
payment option. But if that "illegal" copy has a PayPal link embedded
in it, I'll click on it and send the guy his five bucks or whatever.
For that matter, I'll probably put up a "conscience jar" on the web
site for the book. I wonder why so few musicians who publish their own
CDs do that. Barbara buys a lot of CDs directly from the musicians, and
every time I visit one of their web sites I look for such a link. I
seldom find one. All they need to do is put up a simple link that says
if you've downloaded one of our CDs and not paid for it, please click
here to send us a few bucks via PayPal.
In the wake of a $1.52 billion judgment against it, Microsoft is
probably reconsidering its support for software patents. The judgment
is absurd, of course. Microsoft had already paid a $16 million
licensing fee to Fraunhofer, the recognized owner of the MP3 patents,
which were based on work done by Fraunhofer and Bell Labs. Microsoft
thought, reasonably enough, that it had complied with all licensing
requirements, and yet it still ended up on the wrong end of a $1.52
billion judgment. What's wrong with this picture?
Of course, it's hard to have any sympathy for Microsoft, which
continues to wield its patent portfolio as a club against Linux and
other OSS. Just this week, Ballmer has been muttering about Linux
violating Microsoft patents, although of course he refuses to say which
patents those might be. He might have learned a lesson from the fiasco
SCO faces after making similar unfounded threats, but apparently
Ballmer is a slow learner.
Perhaps Microsoft will come to its senses and throw its significant
lobbying clout into the battle against software patents. Surely
Microsoft must have realized by now that software patents benefit no
one other than patent trolls. Microsoft has already been the victim of
patent trolls to the tune of billions of dollars. How many more
billions will it have to pay before it realizes that software patents
are a bad idea, period?
Saturday, 24 February
1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007 by Robert Bruce