Current
Later
Archive
Home
Search
Mail
Stuff |
2000.09.10 : 2000.09.16
Saturday, September 16, 2000
Yo. Web search! We were looking around this afternoon
for a shirt for my upcoming Virus
Bulletin presentation. I wanted a black, long sleeve shirt with a band
collar. We tried both Bloomingdale's and Macy's. Turns out that band collars
are out of style, and were nowhere to be found. "Try the Web," says Randy.
He's so smart. As always, the
Web provideth.
Plurp. My major technical triumph today was figuring out how
to do that bit of Paint Shop Pro coloring that I groused about the
other day. I am such a technostud!
Friday, September 15, 2000
Yo. George
Dyson (Esther
Dyson's brother, Freeman
Dyson's son) is a clever, soft-spoken guy, author of Darwin
Among the Machines and such. He may also have been separated
at birth from American horror writer H.P.
Lovecraft.
 
What do you think?
Yow. Bruce Damer says he's
got half a million people participating in his virtual
worlds, their avatars running around, interacting with each other and
doing silly bot things in big virtual cities like AlphaWorld
and elsewhere.
Yeah, it's kinda fun stuff, but do I really believe this is the Future
Of Mediated Human Interaction? Probably not. I think it misses almost all
of the key social interactions that make face-to-face meetings such a big
deal. Though we do use both telephones and chat rooms and Windows 98, and
they are Real Different from face-to-face discussions. Maybe I'm just prejudiced
that people talking face-to-face is the ultimately best social interaction.
Bruce does cool stuff. Its coolness may be more in its visual appeal
than anything else, but hey, that's OK. Check out, for instance, simple
artificial organisms that learn to walk, swim, etc.
As a kid, did you watch Frankenstein
and say "Heck - I could do that"? Maybe you can win
the AlivePrize for an artificial creature that is judged to be "alive".
Rant. The DoD folks at this
conference draw the following distinction between business and government.
When an adversary comes after a business, the business might lose money.
When an adversary comes after a government, people die. So, they say, governments
have to be way more concerned about security and control than businesses
do. Governments have to "maintain superiority".
But is that right? Suppose you had a federation of states, where each
of the states could opt to leave the federation whenever they wanted to
(or subject to some set of rules that allowed them to leave). Then states,
even though they are geographical monopolies, could move fluidly between
one federation and another. The reasons might be economic, or social, or
whatever. But no life need be threatened, and no person need die.
It seems to me that the root of the problem is that governments are
happy to enforce their monopolies with deadly force. A government would
rather kill people, even "its own" people, rather than allow them to secede
or, worse yet, join another government. And a government, having found
that it cannot reach some peaceful arrangement with an adversary, would
rather kill people in the adversary's country than live with its failure
to reach agreement.
I think that's the fundamental difference in the approach to competition
between (most) businesses and (most) governments. Ultimately, businesses
go about their activities economically; governments are willing to go beyond
that and kill people.
One could argue that governments must do that to maintain their
monopolies. Perhaps. But that's very a different motivation from protecting
the rights of the people.
Thursday, September 14, 2000
Plurp. Oh, now I remember why I hated California
mornings. This
conference is on the seacoast near Santa Cruz. Mornings in this area
are always the same. The sun comes up, but you can't see it because the
world is enshrouded with a thick overcast that, were it a bit nearer, would
simply be fog. All of the warmth is filtered from the sunlight, leaving
only an intense, blinding blue.
I'm not a morning person. Seriously not. As far as I'm concerned,
single-digit hours are entirely in the evening, and dawn is something you
stay up to, got get up before. When I was growing up in central California,
my mother got tired of trying to get me out of bed for school. Instead
of gently cajoling me awake, she reverted to the only thing that actually
worked: she came silently into my bedroom, threw aside the blackout curtain
I had over my bedroom window, and exposed my gritty eyes to ultraviolet
hell.
So, this morning, when the alarm went off, I went to the window and
pulled back the curtain (hotels always have blackout curtains; sometimes
it's hard to know what time it is), a shrieked. Childhood trauma.
Come 10 AM, the fog will burn off and it will be lovely. But until then
there is only pain.
Yow. There are lots of clever people in the
conference today.
Danny
Hillis, once of Thinking
Machines fame, currently building Really
Big Clocks;
Bill Cheswick
of Bell Labs, at least until next
week when he becomes CTO of a startup
that uses his cool Internet
mapping technology (you can even buy
the maps);
Taher
Elgamal the security guru, now also in his
own startup; Greg
Papadopolous, CTO of Sun, which was
a startup a while ago; Eric
Brewer of UC Berkeley and also
founder of
Inktomi (are we detecting
a trend?);
David Ackley of
U.
New Mexico, proselytizer of artificial life;
Esther
Dyson, currently chairing ICANN,
among other things.
Rant. Late this afternoon, we were discussing the security of
the world's computing systems. The panel moderator pointed out that, to
make the world's systems secure, we need to deploy A1 systems, have pervasive
bulk encryption, a global PKI, software without bugs and people that do
what policy says they should do. Everyone snickered, but I got to thinking.
Yeah, all that stuff's impossible. Everyone agrees, but nearly everyone
ignores that, and figures we all ought to fix some bugs, install that crypto
thing somewhere, and cross our fingers. From the point of view of some
given business, that probably makes sense. Businesses deal with risks as
business issues. How often does it happen? How much does it cost? How much
does it cost to avoid? Their adversaries are almost always in it for the
money, which constrains the attacks and makes it easier to catch the bad
guys. Businesses pretty much don't fall over and die because of risks.
(Though sometimes they do. Banks go under because of large, highly correlated
runs on their assets. Small businesses get wiped out in earthquakes and
floods.) Even when they do, the world goes on.
Governments, and especially the "defense" parts of governments, seem
fundamentally different. Their adversaries are not lone criminals but other
governments, often with loads of money to spend on being bad guys and having
ideological rather than monetary incentives. And it's plausible that the
bad guys don't run bozo attacks constantly, so their adversaries can learn
what they attacks look like and prepare for them. Rather, the bad guys
are likely to save their time, money and deviousness for One Big Attack,
at just the right moment (probably during a war) when it will devastate
their adversary. And that's hard to prepare for.
Hence the government (and especially DoD) fascination with security
and secure systems.
Everyone who's looked at this problem seriously believes that this kind
of impenetrable security will never really exist, not pervasively, not
even commonly. What if we take that notion seriously?
Businesses will continue to do pretty much what they do today: judge
risks and trade them off against costs. And they'll continue to survive.
Governments, on the other hand, have a problem. They will engage each
other in devastating cyber-attacks, and the defender won't have much to
work with. (This has been demonstrated several times in the last few years,
both with DoD tiger teams and with random teenagers.) Some will survive,
whether by staying beneath notice or by treaties and alliances. Others
will be targets of dedicated, well-funded adversaries and will get wiped
out.
Business, in response to this risk, will seek to move their assets to
countries that are less likely to be attacked, and will diversify their
assets among several countries just in case. Some governments will survive.
Others will go down. Business, in the large, will go on.
This also suggests that large, central governments that control large
regions or large populations may not be optimal in the future networked
world. The possibility that taking out a single organization could take
out all of the US, or all of Russia, or all of China, seems awfully fragile.
Perhaps the longer term evolutionarily stable point is further Balkanization
of the world, with territory-based governments possessing less power and
transnational businesses possessing more. This is similar to, but perhaps
not as radical as, William
Gibson's speculation in Neuromancer,
et al.
Wednesday, September 13, 2000
Rant. Just as I suspected, it's hard to write
every day!
Plurp. On the way to pick up our under-powered Ford at the Hertz
place in SFO (or, more accurately,
"sort of vaguely near SFO", as the City Parents have apparently decided
to turn a simple airport into the Aviation Mall
of America), we passed no fewer than three shiny new Jaguar sedans
which we could have rented instead. Could have, that is, if we were dot-commies,
Sergeants of Industry, Masters of Small Parts of the Universe, that kind
of thing. And I surmise there's a lot of that kind of thing out here these
days.
Yow. There was a whale today, a small one, cavorting in the bay
below my room at the Seascape Resort
where I am for a couple of days at a conference.
She surfaced several times, spouting occasionally, before gliding beneath
the silver water. A cluster of birds circled above for several minutes,
hoping to make a meal of something left behind. Then the water was quiet,
the birds soared randomly away, and it was as if nothing had ever happened.
There was a certain magic to it.
Yak. I had a fun conversation with David
Ackley from U. New Mexico tonight
at the pre-conference
bonfire. He's the only longhair here (so far, anyhow), and we spent a pleasant
time intertwining ideas of how software will come to resemble biological
ecologies more than engineered bridges. Great fun. I met Esther
Dyson tonight, along with George
Dyson and some guy named Peter
Schwartz.
I asked Peter what he did and he said "I help companies understand the
future." Yeah right, I think, so do I. "That's a pretty broad brush, Peter.
In what way do you do this?" "In all ways." Oh boy, I think, remembering
that, earlier, he had casually mentioned his recent purchase of a Mercedes
330 convertible, "the second one in California; it's a new model". I figure
he's the stuff of Dogbert
the Consultant. "I also write books
about the future of technology. And movies. I contributed to War
Games."
I'm still extracting pieces of my foot from my mouth.
Plop. Today's Web conundrum is how to produce graphics for my
many soon-to-be Web sites. I downloaded a trial version of Paint
Shop Pro, which can easily do all sorts of fancy transformations on
images - swirling them this way and that, colorizing them, etc. etc. etc.
But I can't figure out how to do the simplest possible thing - make one
particular pixel a particular color. Sheesh. I hate computers. Where are
my crayons? I may have to break down and RTFM. I hate that.
Tuesday, September 12, 2000
Plurp. So what kinds of things could I talk about? Lessee
...
-
Lunchtalk. Alien food symbols. The alien colony at the Medical Center.
The Chernobyl Center for Genetic Diversity. Patents.
-
Web search. You've played the game, now read about us playing the
game.
-
Bumper sticker. Both real and imagined.
-
The apartment. Oh god. The apartment. That's more of a rant than
anything else.
Plurp. Then there's the issue of the name of the thing. Dave
already stole the wonderfully minimalist "log". Some people try for cleverness,
like "usr/bin/girl/". Hmm.
There's the school of mysterious names, like "syrup.org".
So what could I use?
-
Spew
-
YAWL (yet another web log)
-
This Is Not A Web Log. (Yeah, right.)
-
I Am What I Type
-
You Are What You Type
-
Plurp. (I kinda like that. Just made it up yesterday. Only two Google hits.)
Yow. Ooh - yay! Google now has advanced
search. 'Bout time!
Monday, September 11, 2000
Plurp. "This Might Be A Weblog". I wonder
if I really want to start a weblog. Oh sure, it's tempting. My friends
Dave
and Ian both have weblogs now, and
I am quite addicted. I haven't gotten into the billions
other of weblogs. I've read a few of them once or twice, but they're
not (yet?) regular reading for me. I have several problems with the idea
of me doing a weblog.
-
I am, somehow very deeply, a Late Adopter. I just got my
own domain, ya know? I just broke down and got a cellphone a few weeks
ago. And I haven't even started using a Palm Pilot yet. I mean Really Late.
So I can't possibly do a weblog. Too new. Too trendy.
-
I have no time to do even what
I do at work, much less what I do at home. Weblogs take time. Lots
of time, it seems. In addition to time, weblogs are a commitment. Day after
day. Or the readership evaporates. (And if we're not writing for our readership,
what are we writing for?) I'm very ambivalent, very bipolar, about commitment.
I am, at the same time, both fanatic about it and frightened of it. Is
there an XML schema for neuroses?
-
I'm not one of those people who spends hours a day cruising the Web for
weird things that you haven't already seen. I'm just not destined to be
a great source of interesting new links.
-
I am a very private person. (And that's a long essay in itself.) Do I really
want the Vast Unwashed knowing what I think is important? (Hmm. Would anyone
read a weblog of entirely trivial and generic events? Surely there must
be a webring of these by now. Perhaps even a bot that writes them.) I'm
certainly not going to JenniCam
my life. No way!
But, perhaps more than that, I fundamentally doubt that anyone would care
what I think from day to day, would believe that my way of viewing the
world is worth their time, would make the effort to come by my log to see
what meager insights might be littered there, their bits decomposing. I
have an image of myself, up at 3 AM, clicking "reload" on my own weblog,
watching a little hitometer (which I don't even know how to do yet, sigh)
click up from 0000002 to 0000003 to 0000004, knowing that, if I come back
a month later, another look will just have clicked it over to 0000005.
I sometimes (well, more often than I want to admit!) contribute to Dave's
log. Ian says "Oh, yeah. Another of Steve's rants. Too long. Skipped
most of it." It's even easier not to click on it in the first place. Will
anyone care? And yet, it is intriguing.
So I'm plurping some text here, in this not-yet-weblog, to find out
if I'm really serious about it. We'll see.
 |
