|
|
|
 February
28, 2005: A Rant on Things to Avoid in High School
Rant time. Paul Graham has written another
stunning essay (which was aggregated on Slashdot some time back)
addressed to high school students, and basically addresses the issue
of knowing what to do with the rest of your life. For the most part
he's right on: It's less important to know what your life's work
will be in high school than to just cast about, read lots, work
on projects, try to answer interesting questions, and most of all,
don't waste time.
Among his discussion of what can motivate high schoolers, this
one stood out for me:
Closely related [to
wanting to do things you're not supposed to do] is the desire
to do something audacious. Sixteen year olds aren't supposed to
write novels. So if you try, anything you achieve is on the plus
side of the ledger; if you fail utterly, you're doing no worse
than expectations.
I was a contrarian even in grade school, and certainly by high
school. A friend razzed me all through junior year as I sketched
out the details of my 10" telescope. I estimated that it would
weigh 200 pounds. (The truth was closer to 300.) He drew a cartoon
of me as a mad scientist in front of this amazing fantasy contraption
(which bore little or no resemblance to a telescope) with the underlying
note that "YOU CANNOT BUILD 200-POUND TELESCOPES!"
And novels, heh. I wrote one when I was 15, another when I was
16, and a third when I was 17. I didn't write another until I was
48. I didn't know back then how hard it was, so I just did it. I
didn't do much with my teen-era novels, or the ideas they contained,
because I was taught that writers are beatnicks. I wrote them because
I was good at it. If I didn't do it for a few days, I got seriously
twitchy—something still true to this day.
What Graham does not address in the essay is why kids don't do
this sort of thing anymore, as they did when he and I were high
schoolers. I have two answers to that, both of which I've touched
on here in the past, and both of which are among my nastiest rants:
- Cynicism. Our culture is cynical, and much of that is due to
our morally/spiritually bankrupt intellectual class, especially
liberal arts professors. I used to say they should all be ground
up for dogfood, but that would be animal cruelty. (I mean it either
way you could take it.) There have been a couple of recent articles
in The Atlantic about how it's possible to go through a
complete Harvard liberal arts curriculum without learning anything
at all, and this doesn't surprise me, though it's not quite true:
What students are learning is the Big Existential Lie that
nothing matters, so why break a sweat on anything?
- Overmanagement. The other biggie is that we are exhausting our
children by forcing them to do idiotic things every moment of
their days. Sports is the biggest time-waster of anything we foist
on our children. Some families insist that kids play at least
organized sport, and sometimes two. Sports teach nothing—absolutely
nothing—but that winners can and should stomp on losers.
That's what the whole pitiful, sadistic business is for. In many
sports, there is little or no exercise involved, especially for
spindly little kids who mostly warm the benches while the big
kids get to play. (And when spindly little kids try to
play and do badly, because they're spindly or clumsy or whatever,
they are taunted and laughed at and often physically abused. Ask
me sometime how well I did—or, more to the point, how long
I lasted—on the 8th grade softball team.)
There are other minor issues, including the seductive nature of
video games, but most of it cooks down to our cynical culture and
berserk time-management. Some parents of teen boys try to keep them
busy every minute of their day out of a terror that they'll join
a gang or do drugs. This kind of imposed-from-without discipline
prevents the development of the sort of imposed-from-within discipline
that allows young kids to do remarkable and sometimes unprecedented
things. Kids need a little slack time to wonder a little, and try
things, without parents nagging them that they have to go play football,
or worry that they'll hurt themselves. In all my inquisitive nerd's
life, I have never broken a bone or even injured myself badly enough
to require stitches. I built my own 300-volt DC power supply when
I was 12, to power the tube projects I was creating out of torn-up
TV sets, and I'm still alive to talk about it. (Still have the power
supply in fact, and still use it.)
I think that it's basically this simple: If you teach a child some
basic common sense and allow him or her the use of tools and a little
time to use them, and if you make it clear that creativity
and ambition are valued in your house, and if you then have
the guts to stand back and let them roar, marvelous things will happen.
Maybe not instantly, but the process matters more than perfection
or immediate results. I know it's still possible; the seven-year-old
son of my friend George Ott is always out in the backyard, stitching
inventions together out of boards and rope and other junk, which I
think everybody but his parents and me considers a mostly harmless
nuisance. Not so. It's the last best hope of humanity, and if we succeed
in stomping this impulse out of our children, we might as well hand
the world over to the roaches.
|
February
26, 2005: Earth 1, Psychics 0
Well, I think we've given SilverJade enough time on her predictions
of a catastrophic California earthquake on February 23. (See my
entry for February 14, 2005.) California
is still there—which is good, because Carol and I were still
in Orange County on the 23rd. The note on the site now says "Her
dates and steps are not accurate." Duh! And if California does
suffer a nasty earthquake six months from now, fersure people will
point to this prediction and say, "She was just off by a little."
Sheesh, this has happened again and again and again...and people
are still standing there panting for the next addle-pated prediction.
It annoys me because I'm pretty sure that there's a pony inside the
paranormal bag somewhere, and this sort of thing makes people roll
their eyes and refuse to do any serious research, or even take seriously
the serious research that some people actually have done. (See
Jahn & Dunne's The
Margins of Reality.)
|
February
25, 2005: Live Windows?
Several people wrote to tell me (which I knew) that there are now
many versions of Linux that boot from CD and never even touch the
hard drive. I'm familiar with such "live CDs" and use
Knoppix as a "rescue disk" to get data off unbootable
Windows hard drives. What I was actuallly talking about was something
baked into ROM that would boot the Windows kernel (not necessarily
the whole shebang) and thus prevent a rootkit from hiding itself
by modifying the way the Windows kernel works. As George Ewing pointed
out, some primordial personal computers like the Commodore 64 put
the whole OS in ROM to avoid the need for a hard drive. I can't
imagine, if we can create a Flash drive with 2 GB capacity, that
we couldn't create a relatively cheap board with the Windows kernel
in half a gig of unalternable ROM. Everything else (drivers, GUI,
IP stack) could be installed on the hard drive. It would prevent
Windows piracy (mostly) and might also be faster than booting entirely
from hard disk. I don't mind paying for Windows plus an upgrade
every year or so (it's really very useful, if buggier than we'd
like) and this would be one conceivable way to prevent threats to
the kernel itself.
It's interesting to ponder whether there could be a Live CD version
of Windows that would boot and run but not install. My guess is that
it's technically feasable, but Microsoft would never do it because
the CD could be easily copied and used (in a cheap optical drive)
in place of Windows on a hard drive in the conventional fashion. "Right
Men" like Bill Gates and Steve Ballmer want to stamp out piracy
even if it kills them or Microsoft itself. Linux is getting scarifyingly
good, and while it's not at parity with Windows yet, it could be in
another ten years or perhaps sooner. That's when all the real inteeresting
stuff starts, and if the rootkits thing gets really bad, a rootkit-proof
on-ROM version of Linux could look a great deal better.
|
February
23, 2005: How Many Could We Lose?
Having read The
Great Influenza and the link mentioned in yesterday's
entry, I was thinking a bit on what a worse flu might
do to humanity. There have been SF novels about some mysterious
disease basically wiping out all but a handful of us (see George
R. Stewart's classic Earth
Abides) but nothing I've ever seen about a plague wiping
out a third of us, or half of us, while leaving the human infrastructure
undamaged. It's an intriguing question: How many humans could you
pull from the Earth and still have anything like a society? Could
things still work with only two out of three people alive and functional?
And if so, what sort of society would that become? Reduce that to
one out of two and what would we have?
Such depopulation has happened in isolated areas before, especially
after the Black Death in Europe and the ravages of "the companies."
(See my May 14, 2004 discussion
of Barbara Tuchman's A
Distant Mirror.) Interestingly, the Black Death shook loose
a lot of wealth held by old families and drove the last nail into
the coffin of feudalism, and the Europe that emerged was in some
respects more prosperous (if not necessarily more humane) than what
had been before. My intuition is that human society is not quite
as fragile as we might guess. Such a plague might limit the reach
of national governments for a period of years (they would have to
prioritize with a vengeance) but at the local level, life would
probably go on, and over time society would reassemble in a slightly
different form. Would that form reflect collectivism or individualism?
What would we lose? What would we gain?
That's yet another line in my (ever-longer) list of novels-to-be-written.
I'd better get to work, heh.
|
February
22, 2005: Odd Lots
A few odd lots while I'm working through some difficult issues
here:
- Michael Abrash sent me this
link, which indicates that we may be having some trouble with
flu next year. Pay no attention to that 72% mortality figure.
Things like that are never accurate. Almost never. We hope.
- Michael Covington made a suggestion with a lot of merit: That
Windows (or any OS, really) should boot its kernel from some sort
of read-only medium. To avoid rootkits (See my entry for February
18, 2005) I would certainly be willing to buy Windows with
the kernel on ROM, perhaps as a small add-in board for one of
those right-next-to-the-CPU expansion sockets. I wouldn't mind
if it took a few more seconds to boot. Nor would Microsoft, I
would guess, mind the strong antipiracy protection that this would
provide. Nonetheless, I don't think it will happen—unless
a popular version of Linux began to be distributed this way. (There
are rootkits for Linux too.) Let's watch.
- Carol and I visited a wonderful Old Catholic community this
past Sunday (yes, I'm traveling right now): St.
Matthew's Old Catholic Church, in Orange, California. The
church is in the shadow of a monster freeway interchange called
The Orange Crush, so it's relatively easy to reach it from almost
anywhere in Orange County. St. Matthew's is a member of the Ecumenical
Catholic Communion, and is relatively liberal theologically
without being over-the-top in any sense. The liturgy and music
are very close to those used in the Roman Catholic Church, and
we found the people there tremendously welcoming. I spoke a bit
with Bishop Peter Hickman, the pastor, and found him to be a wonderfully
empathetic man. He's also a spectacular preacher. If you're looking
for an alternative to the Roman Catholic Church in Orange County,
I don't think you're ever going to do better than this.
|
February
18, 2005: Windows Rootkits
Slashdot aggregated a
disturbing piece about "kernel rootkits" for Windows,
which allow the bad guys to create spyware and trojans that melt
into the Windows kernel, take control of how Windows discerns their
presence, and as a result become both undetectable and impossible
to get rid of— except, perhaps, by keeping a ghost image of
your fully configured Windows partition offline somewhere, and rolling
back to your standard configuration on a regular basis.
I don't know Windows internals well enough to understand why rootkits
are even possible. Certainly this sounds like a terrific reason
not to run as administrator all the time, though I'm far
from sure that that's complete protection against such things..
More on the technical side is here.
This is a truly ugly business, and if anything begins running people
out of Windows territory to the Mac or Linux, this may be it. More
as I learn it.
Note that I may be posting irregularly or sparsely the next ten or
twelve days. I'll do my best, but sometimes things happen.
|
February
17, 2005: More XP Wi-Fi Weirdness
Reader Bill Dillon had the same problem that I had with Wi-Fi and
Windows XP Service Pack 2 (see my entry for February
11, 2005) and he solved it by rolling back SP2 and then reinstalling
it. It's unclear to both of us why this should matter at all, but
there you have it. The problem I may have with the brand-new PC
that I described in the earlier entry is that I'm not sure you can
roll back SP2 when SP2 wasn't an upgrade to an earlier version of
XP. I'll have to look into this, and if I have further useful adventures
you'll hear about them.
|
February
16, 2005: The Romance of Vacuum Tubes
 I'm
a sucker for tube-based electronics, not because I think it's better
somehow than solid state (it's clearly not) but simply because it
has a certain mad-scientist aura that little plastic squares full
of silicon simply lack. Tubes (and their associated discrete components)
are big enough to grip with your fingers, unlike surface-mount components,
some of which are barely big enough to distinguish from dandruff.
And they look cool. Years back I built an all-tube high-voltage
power supply that contains three gas-filled regulator tubes as well
as the canonical 5U4 rectifier. Gas inside a tube will glow when
electrons are moving around inside of it, and while with most tubes
this is a bad thing, with gas regulators it's what makes them work—and
each variety of gas regulator tube glows a different color, depending
on the mixture of gases inside it. (I've tried to capture this in
a photograph and it just hasn't worked.)
I've begun posting information on playing around with tubes (see my
12V "space charge" tube page here)
and hope to do more in the future. For a look at some of the variety
of form in the tube universe, check out the National
Valve Museum. (Tubes are called "valves" in the UK.)
Once you're in the "exhibits" section, browse the "construction
and technology" pages to get a sense for the breadth of design
of these things. Now that I have a workshop again, I have various
tube projects underway, and I'll post photos here as things progress.
|
February
14, 2005: Odd Lots
A few odd lots while the snow melts on this beautiful Colorado
Springs day:
- All the psychics who missed the Christmas Tsunami are trying
to play catch up. Here's
one that's pretty specific: An earthquake for February 23,
2005. Lots of maps, calculations (including references to C structs,
egad) and plenty of objective stuff that we can wave at poor SilverJade
if she misses. I'll post something soon after 2/23; we can probably
give her a day's grace.
- And while we're talking psychic powers, several people sent
me a Slashdot link about a
random number generator that supposedly predicts the future.
This is about the Global
Consciousness Project, spearheaded by Drs. Robert Jahn and
Roger Nelson. The random number generators used by the project
are based on "quantum indeterminate electronic noise,"
and there is some technical information (though not enough) here.
It would be interesting to build one of these. At some point they
should publish circuits, or even sell PC boards. The more people
are watching these things, the more defensible the Project's position
would be.
After a quick scan, I still have a lot of doubts. I admit that
I'm in a little deep, but my understanding of electronic noise
is that it's not a quantum phenomenon, but if any physicists
out there can clarify what they're talking about, I'd appreciate
it. I have some respect for Dr. Robert Jahn based on his 1989
book The
Margins of Reality, which summarized his experiments on
psychokinesis and seemed to be reasonably good science. If I
get a better sense for what these guys are doing, I'll post
a more detailed entry here.
- Many people took time to explain the date problem with Pepys
Diary. (See my entry for February 6, 2005.)
There was an English tradition at that time to begin the year
on March 25, and end it the following March 24. So dates between
January 1 and March 24 could be in two years, and are often given
with the two years separated by a slash, as 1728/29. The custom
was officially dropped in 1752 and the January 1 New Year made
legal across the board, so English dates between 1/1 and 3/24
prior to 1752 are often given in that dual form. Thanks to all
who wrote with clarifications.
|
February
13, 2005: Delphi Is Ten
Tomorrow marks the 10th anniversary of Borland's release of Delphi
1.0. How could it be? I admit, I haven't done much programming lately,
but if it hadn't been for Delphi, I don't think I would have done
much at all, especially under Windows. The Delphi Programming
Explorer, a book I did with Jim Mischel and Don Taylor, was
one of the first out on Delphi, and was the #1 moneymaker for Coriolis
Group Books in 1995. And although Delphi remains a strong contender
in parts of Europe, here in the US the C bigots have won, even though
C and (especially) clib are at the root of a great many security
issues with modern software. Nothing makes a C bigot howl like any
suggestion that C is to blame for anything, and behind all that
howling I sense a lot of very uneasy consciences.
But I'm tired of fighting that battle. I may learn C# someday,
but for the moment I'm working on other things, and when I need
to gen up a simple utility, I still turn to Delphi 6. I'll do no
howling, but I will re-state my constant premise that if Delphi
were to replace C in the programming universe, things like buffer
overflow exploits would become almost extinct. (Yes, you can write
exploitable code in Delphi, but you have to work at it. When
you use clib, such exploits happen by default, unless you consciously
opt out of using certain fundamental library routines.)
It was interesting that I saw nothing on the main Borland site about
the 10th anniversary; all I got was a simple email from David Intersimone.
Delphi thus gets almost no honor even in its own house. I'll continue
to use it and hope it survives, but at this point I've begun to worry.
|
February
12, 2005: An It Harm Not? Yeah, Right.
It's not every day that you realize that St. Augustine agrees with
the born-again pagans. But today was that day, when I again happened
across Augustine's epigram: "Love, and do what you will."
Wiccans and pagans and others of that sort have a saying too: "An
it harm not, do what thou wilt." I laughed out loud over that
one. (It's always stated in something like Middle English, where
"an" means "if".)
But it brought me to a more serious topic: Does law trump love,
or does love trump law? Most of the great divides in the world today,
in religion as in secular life, have roots in that question. Conservatives
feel that love, to be genuine, must operate within a system of predefined
(and ideally Divine) law. Liberals feel that law is invalid unless,
in all things, it respects the requirements of love. And here I
am in the middle, watching them duke it out and trying to decide
which side I'm on. It's not as simple as it sounds.
I've heard it said that love is self-validating, and law is not.
Those who create law can make it say anything they want, including
selfish and destructive things, all the way up to idiocies like
prohibiting pigeons from gathering in a park, or rounding off pi
to an even value of 3. Love, on the other hand, has some boundary
conditions that are universally understood. The kicker is that law
can be made as precise as it needs to be, whereas the definition
of love is always a little fuzzy. The question I have for the Wiccans
(and what I think what Augustine had in mind in stating his epigram)
is this: How do you know that you will do no harm?
All actions have consequences. Even the so-called victimless crimes
are not always without victims. If doing drugs makes you nuts or
incompetent, somebody is going to have to take care of you. Quite
apart from the risk of STDs, casual sex can leave behind psychological
trauma that may not surface for years afterward. ("Casual sex"
is rarely as casual on both sides of the transaction.) My point
here is that love can fool itself into thinking that no harm is
being done, when in fact it requires great care to live without
harm to anyone or anything.
As a liberal religionist, I'm more than willing to state that love
trumps law. But love without deception is much harder to maintain
than law without corruption. Law has to constantly examine itself
to be sure of its motives, and love has to constantly monitor itself
to be sure of its honesty. Divine love (which is what Augustine
was talking about) implies the Great Law: Do unto others as you
would have them do unto you. Don't fool yourself into thinking that
this is easy.
In other words: Love, and do what you will. Dare you!
|
February
11, 2005: The Dialog That Contradicts Itself
 Windows
XP Service Pack 2 threw me a curve the other day, and I can only
assume it's a bug of some kind. Note the screen cap at left, and
read it carefully: Up top it says "Not connected," and
in the middle of the same item says, "You are currently connected
to this network." I've certainly never seen Windows contradict
itself within a single pane before. I guess there's always a first
time. The truth was that the PC had not connected to the
network. Associated with the AP, perhaps, (my router gave it an
IP) but not connected to Windows networking in any meaningful way.
(For the curious: The unprotected network listed at the bottom is
my next door neighbor's.)
The circumstances behind the message were peculiar. I was here
at home configuring a new system for someone else, and the first
time the Wi-Fi client in the new PC connected to my WRT54G router,
everything was fine. I then changed the router's SSID to reflect
the router where the PC would be installed ("yttrium")
and got the message in the screen capture above. When I changed
the router SSID back to my own SSID, it connected fine. Puzzled,
I swapped in my old BEFW11S4 Wireless-B router, and it connected
fine...because the SSID of both of my routers is the same. When
I changed the SSID on the Wireless-B router to "yttrium",
it again refused to connect.
It seems like the copy of Windows in the new PC somehow fixated on
the first SSID that it saw (which was mine) and after that first connection
wouldn't connect to any other SSID. I still haven't figured out how
to fix this, and may have to do a system restore (basically, reinstall
Windows from the hidden partition using the PC vendor's recovery software)
before taking it to its new home. If you have any suggestions, I'm
listening—but it only hardens my conviction that Service Pack
2 is a very nasty thing. Beware.
|
February
10, 2005: Louis "Bud" Maday, Requiescat in Pace
The past wasn't always better than the present—but sometimes,
and in some ways and places, it was. I learned from several people
in the past few days that Louis
"Bud" Maday died this past Monday, and I threw a prayer
his way, for he was a small but important part of my childhood.
More than that, he was emblematic of a way of life that few of us
ever appreciated until it was gone.
I've mentioned Bud a couple of times on this site, in my entry
for June 9, 2004, and also in my larger essay on Hi-Flier
kites, which in my own personal history always came from Talcott
Hardware. Founded by Bud's parents Lou and Julia about 1950 and
owned by Bud for many years after their passing, it's now owned
and still operated by Bud's daughter Julie. I hope Julie can continue
to make it work, but if she can't, Lord knows it won't be her fault.
It's difficult to understand Talcott Hardware without understanding
a little about about a remarkable corner in Chicago: The
intersection of Talcott Road and Canfield Road. I've been around
the country some, and recently I realized what a rare thing that
corner was: A perfect microcosm of the massive big-box "commercial
strips" that we all now depend on for shopping. At least up
until the mid-1960s, at one intersection just three blocks from
where we lived were:
- Talcott Hardware
- Certified Groceries
- Perlin Drugs
- Louie's (and later Bill's) Barber Shop
- The Pink Poodle Beauty Salon
- Pikul's Restaurant
- A Citgo gas station with a very cold Coke machine
- A small men's clothing/tailor shop, which I think may have been
called The Sheik
- A bakery that smelled like cookies even when the cookies were
all sold out
- A medical office where our dentist and several doctors were
plus one or two other small shops, the nature of which I've forgotten,
though one might have been a women's clothing shop, and another
a gift shop. Note that this mix includes in miniature almost every
category of shopping we now drive endless miles for: Food, prescriptions,
hardware (and hobbies; Bud's store was part hobby shop) baked goods,
gas, clothes, a restaurant, hair care, and medical care. My parents
and our neighbors made good use of it, and lord knows I spent plenty
of time (and money) there, either on kites, model airplane kits,
and pipe fittings at Talcott Hardware, or ice cream cones at Perlin
Drugs.
Nearly all of the shops are now gone; Talcott Hardware is perhaps
the only business there that hasn't changed ownership since the
corner was built about 1950. Bud knew hardware like no one else
I've ever met, and significant to geeky kids like me, he was willing
to talk to us about it. When I was 13 and designing my first big
telescope, I asked Bud if he could put 6" worth of threads
on a length of 2" pipe, thinking of pipe threads like threads
on a bolt—and he very patiently explained to me that pipe threads
were tapered and thus could only go for an inch or so, and even
pulled a 2" pipe elbow out of a bin for illustration. He special-ordered
odd pipe fittings for me (I still have the 2" cross he sold
me, still working as the main joint between the two axes on my big
equatorial mount) and fixed screens for my mother. And it wasn't
until I was old enough to drive to farther places that I came to
understand that few hardware stores sold model airplane kits and
engines, or refills of sodium ferrocyanide for our chemistry sets.
Like, um, none.
Back in the mid-1960s, Bud Maday was a cornerstone of a certain
way of life, which seems in retrospect to have been living small,
but in a sort of gentle and unappreciated equilibrium with our surroundings.
Today we're willing to frantically drive fifteen miles to endure
limitless abuse and neglect in big-box stores where everybody works
for minimum wage and nobody knows anything. The older I get, the
less I enjoy wandering around lost at Lowe's, looking in vain for
a couple of odd bolts that Bud could have laid hands on in fifteen
seconds. The comfortable and close-in life we lived around Canfield
and Talcott was something we gave up willingly, for a couple of
bucks' discount and dizzying product choice. We came upon it accidentally,
we lost it without realizing it, and I wonder if anything like it
can ever happen again.
In the meantime: Thanks, Bud, and Godspeed. You will not be forgotten.
|
February
8, 2005: Hoarfrost
 We
have a phenomenon here in Colorado Springs that I haven't experienced
before: Hoarfrost.
Yesterday morning we woke up to its second occurrence this winter.
It happens when the temperature is above freezing when fog rolls
in, and then, as the temperature drops below freezing, the moisture
in the fog begins to crystallize out everywhere. When we woke up
yesterday it was 22 degrees, and the fog was still so thick I couldn't
see Villegreen St. off the back deck, and it's only about 200 feet
away.
The above photo (of an ornamental cap on our back deck railing)
gives an impression of some of hoarfrost's peculiarities: It grows
whiskers on things, and the longer the fog remains with below-freezing
temperatures, the longer the whiskers grow. The fog hung around
all day yesterday, growing whiskers on everything. The trees looked
snow-covered after awhile, but it didn't look quite right for snow,
and I failed to get any photos that did the effect justice. However,
you can get a general idea of below-freezing foggy weather from
this photo, looking across the street
from our front porch. I'm still not completely clear on the physics
of why the fog didn't precipitate out as snow when the temps got
cold, though I've seen notes that fog can be supercooled to far
below freezing without precipitation.
However it happens, it was beautiful and strange, and I like
that sort of thing!
|
February
7, 2005: Battling the Spam Zombies
Yes, I've learned a little about the sea-change in the spam-zombie
problem that I described in my February 5,
2005 entry, thanks largely to long, thoughtful notes from Rick
Widmer and Darrin Chandler. (Many thanks, guys!) I can probably
cook it all down to a few bullet points:
- The "trick" of sending spam through the ISP's SMTP
servers is no trick at all—and certainly not a technical
one. (The only trick for the zombie is figuring out whose network
they're on, which isn't much of a trick.) They used to do it that
way at the dawn of spam time, and the ISPs used to catch them
and cut them off. They're doing it now because they have no choice:
With Port 25 blocked by nearly all major ISPs, there's no other
way to get the spam out.
- Detecting spam zombies is extremely difficult and getting worse.
The newest spam relay engines don't listen on any ports at all.
Instead, they periodically visit an obscure Web site or IRC channel
somewhere and download instructions. So port scanning will only
catch the older zombies, and soon will catch none at all.
- The only way to catch zombies now is to monitor mail volume
through the ISPs' SMTP servers. It's still a bit of a puzzle why
ISPs aren't spending more effort on this. Some ISPs throttle outbound
email volume from dynamic IP accounts right now, and more are
likely to do it in the future.
- Some ISPs have hindered the zombies a little by requiring mail
clients to log in to their SMTP servers with a password. This
is a good idea, but the response by the zombies is obvious: Dig
around in any detectable email clients on the PC for logins and
passwords.
- There is only one more possible spammer trick in the bag: When
you've got 10,000 spam zombies by the tail, you don't have to
send half a million messages from any single zombie. You can make
it a parallel process and send 500 copies of the message from
each zombie, if 500 messages per day becomes the ISP touchstone
for spamming. We haven't seen this happening yet, but it's almost
inevitable.
- I had one idea, but I don't know how tenable it is: Building
something into Windows that "registers" an email client,
and can somehow reliably tell whether an email stream is coming
from Outlook, Thunderbird, Poco Mail—or something else that
hasn't been registered. Mail would be passed on to the TCP/IP
stack only when it comes from a registered mail client. Could
a system like that be designed and implemented well enough to
prevent a zombie from spoofing the system and passing itself off
as Outlook? (Do I lose points if I doubt it?)
- Finally, SpamHaus
generated the
report that triggered all this uproar, and they are already
protesting that they were misquoted by an interviewing journalist—read
to the end of the link above—but they got the publicity,
which I suspect is what they wanted. I do feel that they are still
dead wrong in assuming that ISPs will not respond to spammer traffic
through their SMTP servers. It's going to happen, and it's going
to happen reasonably soon. When it does, the spam zombie game
will mostly be over.
|
February
6, 2005: Pepys the Blo...I Mean, Web Diarist
Rich Rostrom pointed me to a
very clever project: Presenting Pepys' Diary as...a Web
diary. (Ok, a blog. I still hate that word.) I read the whole Diary
when I was an undergrad, and enjoyed it, mostly because Pepys was
a typical (if not necessarily ordinary) educated Londoner of the
17th century, and his Diary gave us a good look at what big city
life was like in the 1660s. He was obsessed with women and the arts
and all manner of diversion, and wasn't especially religious. Sounds
like Seattle! My guess is that if Pepys were to be brought back
to life today, he'd require a month's course on using modern technology
(especially Movable Type) but after that he'd fade into big city
liberal culture and be indistinguishable from any of his culture-homogenized
colleagues.
My only puzzle is that in the posted entries, the year is given as
1661/1662. I don't have my college paperback edition of the Diary
anymore, but I don't recall any calendric confusion as to the years
during which he wrote. I'll have to take a closer look at that.
|
February
5, 2005: Port 25 Blocking Is Working!
Suddenly the ether is vibrating with a
new warning that email as we know it is doomed, because spammers
have figured out how to make their legions of compromised home PCs
send spam through the ISPs' SMTP servers.
Huh? This is a bad thing? Am I missing something somewhere?
As I read it, this can only mean that port
25 blocking has been more successful than anyone wants to admit
at curbing the activities of spam zombies, which have since their
inception included their own SMTP servers. Block port 25, and spam
zombie SMTP servers can't connect to SMTP servers elsewhere on the
Net. Their only recourse, then, is to use the ISP's SMTP servers.
But that allows the ISPs to monitor mail volume, and cut off residential
customers whose accounts are sending, oh, half a million messages
per day. This sounds more like the end of spamming as we know it
than the end of email. I suspect I'm not understanding what the
new trick is (and the non-technical articles aren't helping) but
I'll research it a little more and report back in a few days if
I learn anything.
In the meantime, this is a good time to wonder why we seem completely
unable to do anything about spam zombies on residential broadband
customer machines. Back when I had a cable modem account at my in-laws'
in Chicago, I used Black Ice Defender as a firewall, and it logged
several HTTP port probes per day from the ISP. They didn't want
people running Web servers, and the probes allowed them to spot
people who weren't following the rules. Ok, that's cool. Now, if
the zombies are following Internet protocols, they will respond
to a port probe. I can only assume that we've reverse-engineered
the zombie software to discover what port or ports they're using
and what, (if any) custom handshakes they're expecting. (If not
that, then let's at least pretend to be spammers, buy a utility,
and rent a list of zombies so that we can capture the connection
packets and figure out the protocol.) Why aren't the ISPs running
zombie probes, and suspending accounts of customers whose machines
are infected?
I've heard estimates of 100,000 new zombie infections per week,
and if that's really true, the ISPs should be mounting Manhattan-class
projects to develop software to identify zombie infections on their
networks. But I see nothing about this on Google.
Again, what am I missing?
|
February
4, 2005: Separated at the...Barber
  Every
so often, somebody sends me a note (accompanied by a newspaper clipping)
telling me that my ringer is working at the Chicago Sun Times,
being the radio and TV critic. This has been going on now for eight
or ten years. This is the only photo I've found of him online; he
actually looked a little more like me a couple of years ago, when
we both had the same "big glasses" I wear in the front-page
photo I've posted since 2000. But I'll leave it for you to decide
how much I look like Robert Feder of the Sun Times. You can
stop sending me clippings, heh.
I do need a new publicity photo, though. That one is now almost
seven years old, and I do really believe in truth in labeling. The
boy ain't 45 anymore.
I now have a link to a
nice Metric-English calculator in my Firefox bookmark bar. Works
well, and I'm amazed at how often I use it.
Oh, and I'll stop here before I make a complete mockery of my time-honored
odd lots format: Brook Monroe pointed me at this.
Completely useless, but what a wonderful hack! Hey, I have microwave
diodes too, but my 1N21s are basically the size of the whole damned
mechanism. (Wi-Fi detectors will become useful when they tell us
the SSID of nearby hotspots, and not before.)
And so (as Pepys—the first blogger—used to say) to bed.
|
February
3, 2005: The Great Influenza, Concluded
Concluding my review of John Barry's The
Great Influenza, which began here.
As I said at the beginning of this review, I like it when a book
tells me more than I had expected it to. In his 500-odd pages of
text and notes, the author put forth four themes, of which three
were interesting:
- He explained how American medical education was utterly transformed
between 1890 and 1910. Before that, medical doctors were trained
a lot like barbers. They didn't even need a high school diploma
to be admitted to med school. Basically, anyone who could pay
tuition was admitted, and at many schools, students never saw
a patient or even examined a cadaver, but were only lectured to,
often by physicians who found practice distasteful and chose to
teach instead. In some respects this didn't matter, because we
knew so little about medicine before 1900 that treatment basically
cooked down to a handful of folk remedies and the inevitable,
"Stay in bed and drink lots of fluids." The Johns Hopkins
medical school changed all that, and (having been shamed by an
upstart institution in a hick town like Baltimore) the Ivies and
other major universities got with the program and began the reforms
that made medicine a true profession by 1925.
- He explained how we basically lost all our constitutional freedoms
when we entered World War I, and that a great deal of what we
today call paranoid or intrusive government was born about that
time. (The FBI, for example, was formed in 1910. Zoning laws and
the Federal income tax also first appeared about that time.) The
author clearly hates Woodrow Wilson, and it was obvious enough
in the text that I feel compelled to look for another view on
the man. The story that Barry didn't tell is what it took
to get our freedoms back, and that is a story I'd like
to have heard.
- The third useful theme was, of course, the Great Influenza itself,
on which the bulk of the book is focused. Apart from killing 700,000
Americans and countless people in the rest of the world (though
perhaps not the 100,000,000 that the author claims) H1N1 galvanized
medical research and turned big city public health departments
from political sinecures to real, functioning organizations. Barry
clearly points the finger at military logistics for enabling the
virus to spread as quickly as it did (and blames most of that
on Wilson and his immediate reports) but I object to that emphasis,
since H1N1 traveled just as quickly to parts of the world where
American troops had not been for years, if ever. Had it not broken
loose in overcrowded military training camps, it would have broken
out in the hearts of big cities, in movie thaters, in overcrowded
slum housing, and many other places. The big lesson here: Even
in 1918, there was constant and relatively rapid travel to almost
every part of the world. It took about three months for H1N1 to
reach the farthest corners of the globe in 1918. It would take
less than a week today—with all that that implies.
The fourth theme was that the researchers who struggled to find
the causes of the Great Influenza were dedicated, energetic, brilliant,
frustrated, and very human. Well, so were the people who invented
radio, and the people who invented electrical power distribution
(Edison! Tesla! Westinghouse!) and a lot of other scientific and
technological breakthroughs. The flu researchers were no different
from other cadres of driven, brilliant men, except that they
completely failed to do what they set out to do. Neither the
cause of nor any preventive or curative treatment for influenza
came out of the research of that era. We were literally decades
identifying the virus and developing vaccines after the Great Flu
came and went. I admire their energy and their persistence, but
the hundred-odd pages that Barry spends on them really cast no light
on anything. 25 pages would have been more than enough.
That, however, is a quibble. Overall, this was the most thought-provoking
book I've read since Jared Diamond's seminal Guns,
Germs, and Steel. Highly recommended, especially if you're young
and vigorous and believe your health will carry you through. Sometimes
health isn't enough. In fact, sometimes the glow of health can kill
you, in eighteen hours or less.
Read it.
|
February
2, 2005: The Great Influenza, Continued
Continuing my review of John Barry's The
Great Influenza, which began here.
The book wastes a considerable amount of time early on explaining
the parentage of the Johns Hopkins institution, along with the lives
of the toffs who got together to create it as a challenge to the
superior medical schools of Europe. In general, Barry spends more
time than he should on the lives of the researchers who chased the
cause of The Great Influenza (which was a virus we now call H1N1)
especially since it was pretty much a futile effort at the time.
Indeed, he spends a huge amount of time describing the career
frustrations of flu researcher Paul Lewis, which might have been
more interesting in a biography of the man himself, or a detailed
history of 20th century medical research. I'd hazard that about
a fifth of the book is a distraction from Barry's primary mission,
which is to give us a understanding of the almost incomprehensible
havoc that the virus left in its wake.
Although Barry does an excellent job describing the situation of
the common people, especially in hard-hit big East Coast cities
like Philadelphia, his best—and most frightening—material
explains the virus itself and how it kills. One of the great mysteries
at the time was why the disease killed so many of the young and
healthy, and killed them in as little as ten or twelve hours. Flu
routinely kills the very old and the very young, but in general
healthy people from 20-35 shake it off with little more than a bad
cough, fever, and some weakness.
This time it was different. It took over seventy years for medical
science to figure it out, but the answer lies in a chilling, five-page
blow-by-blow description of the battle between an aggressive pulmonary
virus like H1N1 and the human immune system. The immune system is
complex, and it calibrates its response to infection in terms of
how serious the infection is, and especially how rapidly it proceeds.
The more aggressive a pathogen, the more aggressive the immune system's
response. The immune system takes special care of the lungs, because
they are in intimate and uninterrupted contact with the outside
world. This care works so well in healthy people that the lungs
are normally sterile.
In response to infection, certain white blood cells release proteins
called cytokines, some of which are used in coordinating the immune
response, but many of which are directly toxic to pathogens. What
happened with H1N1 is that, unlike most pulmonary viruses, it didn't
slowly infiltrate the lungs from their upper reaches downward. It
attacked virtually the entirety of the lungs at the same time, which
in turn triggered a response from the immune system that modern
researchers call a "cytokine storm." The immune system
is not an intelligent entity; evolution has made it exquisitely
versatile, but it still operates automatically. The massive quantities
of cytokines released in the immune response basically destroyed
the virus in the lungs, and destroyed the lungs themselves in
the process. The delicate tissues of the alveoli just can't
handle that sort of toxic warfare, and what cells were not destroyed
by the virus were destroyed by the immune response, in a sort of
fatal collateral damage that evolution (which doesn't often see
anything like H1N1) has not had a chance to moderate.
This is why young, vigorous people died in huge numbers, and quickly:
The stronger their immune systems, the more aggressively their own
bodies waged war in their lungs against H1N1—and the more rapidly
their lungs collapsed into a mass of blood, cell fragments, and
pus.
At that point, I had to put the book down, take a deep breath,
and go for a walk.
The condition triggered in young people by H1N1 in 1918 has come
to be called ARDS: Acute Respiratory Distress Syndrome. Even today,
with all the high-tech care that a big-city intensive care unit
can bring to bear, the mortality rate from ARDS is 40-60%. Without
that sort of high-tech care, ARDS is still 100% fatal, just as it
was in 1918. That gives us something unsettling to think about,
in case anything as aggressive as H1N1 rises from the seething,
ever-changing viral vortex we still say is..only the flu.
I'll sum up tomorrow.
|
February 1, 2005: The Great Influenza,
Continued
Continuing my review of John Barry's The
Great Influenza, which began here.
It's impossible to understand the Great Influenza without understanding
a little bit about the American role in WWI, and especially Woodrow
Wilson. Wilson was an idealist, and not an especially practical
man. He resisted European calls to join WWI for a long time, but
after it came out that Germany was promising Mexico a huge chunk
of the American Southwest if it would go to war against the US,
Wilson went what we would consider a little nuts. For him the Great
War became a crusade, and he would let absolutely nothing stand
in the way of the war effort.
Wilson basically set aside the Constitution, and instructed the
Federal Government to take over the US economy almost entirely—and
prosecute anyone who published opinions not favorable to the War.
Apart from the Europhiles on the East Coast, the War was bitterly
unpopular with Americans, but newspapers were intimidated into publishing
nothing questioning the effort, or even factual news that did not
indicate that US forces were winning. Wilson became obsessed with
American morale, and gave the Attorney General the power to arrest
and imprison anyone who did anything that would negatively affect
morale. The Librarian of Congress was to secretly report anyone
who requested books on a long list deemed subversive. The FBI created
a volunteer spy force called the American Protective League, with
the mission of secretly listening for "sedition" and looking
for "slackers and food hoarders" and people who didn't
buy—or didn't buy enough—Liberty Bonds. The APL harrassed
German immigrants and members of the International Workers of the
World, sometimes descending to the tactics of a lynch mob. Wilson
created the Committe on Public Information, the purpose of which
was to feed the newspapers information on the War that they were
required to publish unedited. The Federal Government, whipped on
by an increasingly paranoid Wilson, used the CPI to control public
opinion about the War. What the CPI did was lie, lie, and do almost
nothing but lie.
And we gripe about the Patriot Act, heh.
The upshot was that Americans lost any means to hear the truth
about anything. When the Great Influenza broke out in September
1918, little or nothing on it was published in the press, for fear
of affecting morale. The flu began in massive, overcrowded Army
encampments filled with new draftees from every part of the nation.
Little or nothing was said. It was only the flu. The newspapers
could not publish the names or numbers of the thousands of soldiers
who were dying, sometime in less than a day, literally drowning
in blood that filled their lungs and poured from their ears, nose,
and even eyes. Americans were not informed about the flu at first,
and learned nothing except rumors and furtive gossip.
The advice in the press did not align with what people in the big
cities began to see toward the end of September 1918. Hospitals
filled, people died first by the hundreds, and then by the thousands.
Still, the official government line was that it was "only influenza."
And the official government line was the only line there was.
This is why so little was published about the Great Influenza while
it was happening—to do so was essentially illegal. And so a
great deal of hard information was lost. Then again, once the Great
Flu kicked into high gear in October 1918, one could reasonably
wonder whether Americans had any desire at all to remember what
was happening all around them.
More tomorrow.
|
|
