Rants & Raves
Money Talks
Thanks for a superb interview with one of the truly great hucksters of our time ("The Future of Money," Wired 4.10, page 140). The reactionary infotopianism Wired promotes won't be complete until it embraces Walter Wriston's redefinition of democracy: fund managers manipulating the flow of stateless money to punish governments that try to stop people (citizens, in pre-information-age jargon) from getting completely shafted by money vultures like Wriston. And we should see this as a "plebiscite on policy"?
Forget what the word plebiscite means. Forget all the popular sovereignty garbage of old-time democratic governments that thought people ought to have an equal say in decisions that affect their lives. No, let's embrace Wriston's idea of "economic democracy" in which only capital fund managers get a vote. Such pure spirits will make those pesky governments accountable.
Oh, and let's be sure to forget the long, sordid history of Wriston's venerable Citibank. Even today, Citibank is embroiled in a federal investigation concerning money laundering for drug cartels. Wriston's democracy doublespeak is a real advance for the Wired view of the world.
William Pietz
pietzw@delphi.com
Money Talks
I enjoyed the interview with Walter Wriston. He certainly is nothing like the stuffy New York banker I imagined. I only hope that my mind will be half as alive as his when I'm 77.
Jack Lavelle
lavelle@primenet.com
Horton Heard Us
In response to the Net Surf titled "Horton Hears a Marketing Opportunity" (Wired 4.08, page 175), with its plea to make the site less promotional:
We've redone our Web site and made it more fun.
We've added a lot, but we're still not quite done.
We'll keep on improving and adding new looks.
(But oh, by the way, we still want to sell books.)
Please take a new look and we think you will say,
"The Cat in the Hat has some great games to play!"
So all you Seuss lovers can now give a cheer:
"We heard you in Seussville - loud and clear."
Randi Benton
Random House New Media
rbenton@randomhouse.com
Embedded Thoughts
I enjoyed "The Embedded Internet" (Wired 4.10, page 98) and wanted to offer a slightly different perspective. The first weakness in the plan to use embedded systems to leverage the general populace onto the Net is the Net itself. Think of the choice: you can pick up a phone - knowing you'll get a dial tone - and mash 911, or you can press a button and get a "Failed to make modem connection" error message.
The second weakness is embedded systems, not to mention the hubris and blatant arrogance of embedded systems vendors. I have an automated engine room on my ship, fully certified for unattended operation of a 33,000-horsepower propulsion plant, and a 5-megawatt electrical plant. Sure, it works (sort of). And yeah, it's reliable (kind of). But it's nothing I'd spend my own money on. The smallest denomination of currency with which our automation vendor is familiar runs to six figures, and its shortest waiting period for a field change is half a solar cycle.
A final point on security. I'm wired to the Net. If a vandal breaks in and turns my microwave on and off for kicks, no big deal. But how far do I trust the Net? Is it fine to let the power company read my meter, or will it really be my meter it reads?
Again, I loved the piece. But the embedded system vendors I've dealt with are low characters with questionable morals. Princely sums for paltry technology! Put bluntly, I'd trust Microsoft before I'd trust an embedded systems vendor!
Chuck Becker
104262.731@compuserve.com
The true killer application of the embedded Internet will address a problem much less compelling than 911 service. In the future, all clocks will be online, polling every few seconds to determine the correct time. The hassles of resetting clocks for Daylight Savings Time, checking that the VCR has the correct time before programming a taping session, et cetera, will be eliminated and
poof! another of life's minor but pesky inconveniences disappears.
Ray Ross
lalanovich@aol.com
Was it Yogi Berra who said "Prediction is hard, especially when it's about the future"?
I remember a diorama from the World's Fair that offered a vision of life in the 1980s, when people would ride around inexpensively and luxuriously in dirigibles. This was before the Hindenburg disaster in 1937. Anyone who went to grade school in the '60s remembers the push for all-electric homes because in a decade or so nuclear power would be so cheap we'd pay just a flat monthly rate ...
As a technology writer, it stuns me that embedded technology has become so mainstream that it is covered in Wired. Two years ago the Embedded Systems Conference seemed more like a high school reunion than a major trade show.
1996 was the first year that embedded Internet or Java was even mentioned at the Embedded Systems Conference. But it's not a fad; it's here to stay. Maybe Java won't be the app that makes it into all the Net appliances, and maybe the Net will look completely different by the time it controls my home door locks, but the technology seems unavoidable.
My company (on the Web at www.systronix.com/) designs and sells embedded programming tools, though at the moment our hardware and software don't support network connectivity. Why?
Because most factories don't network their manufacturing systems. It's too expensive and difficult to set up and maintain. Plus, there's little to be gained by connecting a cheese cooker to a deep freezer. Different tools and machines (if they are computerized at all) don't speak the same language, so why connect them? I believe that will change, not overnight, of course, but pretty darn fast.
Who knows what the future holds? All I know is that it's sure to be interesting. I hope my new Internet-ready toaster is more crash-resistant than my PC.
Bruce Boyes
bboyes@systronix.com
Snow Job
I loved "Machine Translation" (Wired 4.10, page 84) but had to respond to the comment that "an Eskimo has lots of words for snow; I have one." We have many words for snow, especially those of us who live in ski country. These come to mind: powder, packing, mashed potatoes, sierra cement, champagne powder, crud, drifting, blizzard, whiteout, squeaky, consolidated ... the list goes on.
Mike McKenna
mikem@sdc.tiac.net
Copy Cops
Copyright hysteria has taken over. I recently went to the University of Waterloo photocopy shop to make a copy of the Rock and Roll Hall of Fame and Museum article (Wired 4.09, page 47) - which was, ironically,
a satire on the silliness of some aspects of copyright and trademark law - but the operator turned me away. He refused to copy it on the grounds that (and I quote) "the postcard pictured cannot be determined to be a real postcard in existence or not."
Information may want to be free, but a million satraps are scared by the idea.
Jeffrey Shallit
shallit@graceland.uwaterloo.ca
No Such Thing as Freedom in Singapore
I just happened to be downloading Netscape 3.0 when I stumbled upon your Web page and found "Disneyland with the Death Penalty" (Wired 1.4, page 51).
I am a Singaporean and want to commend William Gibson's precise understanding of local terms, culture, and places. I cannot agree with him more about the direction this country is heading. The problem with today's political climate is that many Singaporeans don't realize that "Big Brother" is everywhere. And the local media just praises the government, printing things like "We are the world's best in this and in that." We are economically successful, but we are also politically immature and have been bred in an environment that teaches silence.
Thanks again for the article, and I thank the Internet for freedom of speech - it may not be around for long in Singapore.
Edwin Lam
fhlam@pacific.net.sg
Singapore has never pretended to be a model for cultural exchange (Electric Word, Wired 4.11, page 42). Just as the US stands as the champion of freedom, Singapore is guided by its forefathers' dreams of peace and prosperity. I'm glad this nation is not as totalitarian as we are frequently viewed or I wouldn't be able to purchase your excellent magazine on the local newsstand.
Philip Tan Boon Yew
firehzd@pacific.net.sg
What McLuhan Said
Kevin Kelly's interview with Derrick de Kerckhove ("What Would McLuhan Say?" Wired 4.10, page 148) reminds me that talking about Marshall McLuhan is like herding cats.
Twenty years ago, I studied under Wired's patron saint and came to appreciate his interest in the future of religion. Updating and replacing this concern, the perplexed soul might well ask: In what ways does the Net refract the Clear White Light?
Emile Lefort
Luxembourg
Netheads vs. Bellheads vs. Wired
Networking protocol design is not about culture; it's about science. (See "Netheads vs. Bellheads," Wired 4.10, page 144.) I am not a Bellhead or a Nethead. I'm both. I ran a BBS for three years, a Fido-style mail network, a Net site, and I have worked for ISPs and telephone companies.
In the networking industry, ATM gets a lot of respect. A friend of mine, who is so deeply involved in protocol design that his bathroom walls are lined with header information, called ATM "the perfect protocol."
Steve G. Steinberg is right that the variable-length headers are much more efficient and adaptable. However, a cursory glance at any ATM book will explain the reasoning behind the fixed-length cells. ATM is not designed for audio, video, or data traffic. It is designed to handle all of them. Real-time multimedia and telnet have very different requirements. Data that should be circuit-switched should be circuit-switched; data that should be packet-switched should be packet-switched. ATM allows for this.
If we are going to have an information infrastructure that handles audio, video, and text data - real-time or not, datagram or stream, with all the traffic decentralized and many-to-many - we need a protocol that can handle it all. POTS can't. ISDN can't.
IP can't. ATM can - and will significantly reduce the costs of high-speed networking. Wired should stick to politics and leave hacking to the IEEE.
Tom Cross
decius@ninja.techwood.org
Having played significant roles introducing and implementing advanced network technology at Sprint between 1991 and 1995, we couldn't resist responding to "Netheads vs. Bellheads." The Netheads are wrong. Bandwidth isn't cheap. (Note: This belief is rooted in the fact that the federal government subsidized much of the global Internet through 1994.)
We have worked with several generations of SprintLink engineers. They are all terrific at what they do and are as critical to Sprint's Internet operation as you suggest. But have those engineers ever had to make a business case that justifies additional network capital/network bandwidth? In preparing that case, they may find that the cost of that "cheap bandwidth" generates only minimal (if any) profit margin given the Internet's fixed, flat monthly price approach to service. The Bellheads have been justifying additional bandwidth for decades, which is why they rarely say "bandwidth is cheap."
The Netheads have yet to face "real" performance requirements. Internet service providers will not provide network performance guarantees (or even objectives). If users experience delays or connectivity problems, they try a ping or trace route. If they feel the results are unacceptable, they report the problem to their service provides. In turn, their service provider first explains that pings and trace routes are not reliable ways to measure performance because they are "low-priority packets" and the network is "very dynamic." And if that doesn't work, it explains that the performance problem is not with the network but with one of the many networks that the user must connect through. And if that doesn't work, it hides behind the fact that the Internet technology is very new and will sometimes break under the stress.
To the contrary, the Bellheads have been there ... starting with the telephone Grade of Service that is published in tariffs. Currently, ATM users are contracting for end-to-end performance guarantees that include maximum network delay thresholds and high network reliability requirements. If the service does not meet the requirements, the service provider refunds some portion of the price. We have not heard of ISPs backing up their service with money.
No matter how you measure it, the Internet is much smaller than the public carrier networks. And scaling up will create extraordinary stress on the current Internet. The Bellheads have successfully addressed this issue, creating a network of enormous complexity that presents a simple common interface to billions of users worldwide.
Internet technology supports a terrific communication system and has contributed more to computer interoperability and electronic information exchange than any other computer technology. As a stimulus to electronic commerce and intercompany communications, the Internet has no peer technology. But the jury is still out on the Internet technology's ability to meet future bandwidth and network performance requirements.
The jury is also out on ATM. Will users have to sacrifice some of the interoperability that the Internet provides in return for performance guarantees? And ultimately, how will the costs of each service compare?
The debate between the Netheads and the Bellheads is healthy and important for the advancement of network technology. But it's time to get beyond the emotional and religious fighting and
on to a reality based on facts, figures, performance, and deployment.
Tim Clifford and Bob Doyle
tclifford@linguateq.comandbdoyle@linguateq.com
Of Mice and Pac-Men
"Spawn of Atari" was beautifully written and very accurate (Wired 4.10, page 166). I worked for Atari Corp. just after its heyday - from 1987 to 1992 under the Tramiels. I was 21 and had left college because
I dreamed of working for Atari, not realizing at the time how much the company had already changed. The Atari 400 and 800 shaped my future, and I knew instantly what it was I wanted to do with my life.
Thanks for summing up the history of Fuji U.,
as my coworkers and friends often referred to Atari.
It's an important part of Silicon Valley's history.
Jerome Strach III
jerome@pixar.com
Computer Firsts
Colossus was a very important computer for the Allies during World War II ("Resurrecting Colossus," Wired 4.10, page 78). And Howard Aiken's Mark I, finished in 1944, was a major achievement. But neither was the "first" computer. Konrad Zuse (1910-1995) created the first functioning, freely programmable, and fully automatic computer. The Z1, a test model, was completed in 1938
in Germany. The Z2, a functioning electromechanical computer, was completed in 1940. The Z3, the first freely programmable computer using binary arithmetic, was operational in 1941. The Nazi r�gime never recognized the importance of Zuse's invention. Zuse had applied to the German Ordinance Department regarding an encoding and decoding machine, but he was rebuffed as it was thought that Enigma was absolutely undecipherable.
Destroyed in air raids in 1943 and 1944, the Z3 was reconstructed in 1960 for the Deutsches Museum in Munich. A rebuilt model of the Z1 is in the Berlin Museum f�r Verkehr und Technik. For more information on Konrad Zuse, read his autobiography, The Computer - My Life.
Thomas Corbi
groklib@watson.ibm.com
Linguistics 101
It plays to the worst kind of nerd stereotype that when Wired runs an article on language ("Dejpu'bogh Hov rur qablIj!" Wired 4.08, page 84), the piece focuses on Klingon and Elvish.
One of my many questions about this very questionable piece arose from Gavin Edwards's comment (in the fourth paragraph) that there are "5,000 human languages spoken today." Later, he refers to "the tens of thousands of languages that humankind has spoken through the ages." Tens of thousands? This would mean that there are at least 15,000 attested dead languages. True? Not even close. With more than 5 billion loyal users, human language has an installed base that puts other software to shame. Next time you run a piece about language, get someone - like a linguist - who really understands the material.
Gabriel Cheifetz
gabecheif@aol.com
Hacking Language
Wired seems to be laboring under the misconception that the people who break into computers are hackers and that what they are doing is hacking. Please remember that it is hackers, not these lowlifes, who made it possible for there to be an Internet for you to publish articles about. Take the time to refer to crackers as crackers. If you can't be bothered to correct this mistake, you will be showing, once more, that Corporate America is incapa-ble of respecting the people who make everything possible.
Peter Seebach
seebs@solon.com
Seeing Red
As a scientist and fan of your technology-worshipping magazine, I cringe whenever I read an egregious scientific error in Wired. Although I found the "Tired/Wired 100" (Wired 4.09, page 148) to be thoroughly enjoyable, the very first entry refers to Earth's smaller neighbor Mars as "the red giant." Mars is indeed the Red Planet, but it is no giant.
On another astronomical note, the recent interest in EUV lithography has an analog in the world of astronomical research. The EUV wavelength region recently became the final region of the electromagnetic spectrum to be subjected to astronomical observation. This was made possible by one of NASA's smaller, cheaper satellites, the Extreme Ultraviolet Explorer (EUVE). The UC Berkeley scientists behind the EUVE are pioneering new low-cost techniques for spacecraft support that may be the wave of the future.
David H. Cohen
cohen@duff.astro.wisc.edu
Lost and Found
It is the real story that gets lost in "Lost in Translation" (Wired 4.09, page 130). The Web drives the explosive use of digital text across national and cultural borders. Digital text requires character encodings. Out of the babel of incompatible and single-language character codes of the '60s, '70s, and '80s now rises The Unicode Standard as a comprehensive attempt to put both major and minor languages and scripts on equal footing. Rather than truncating or mutilating people's heritage "for the sake of computers' convenience," as the article charges, Unicode reflects a strong commitment to eventually and fully support all languages.
In a tale reminiscent of C. P. Snow's classic Two Cultures and the Scientific Revolution, the disconnect happens when technical realities intrude. Jose Manuel Tesoro charges that Unicode's support of Chinese is incomplete and arbitrarily limited to 20,000 of 75,000 Chinese characters, and that it is now "too late for the remaining 40,000 or so rejected Chinese characters." In its first version, Unicode supported only 65,355 characters; version 2.0 will recognize 1,114,112. This is not a simple, straightforward process - it takes many years to create a usable character set from scholarly listings, old dictionaries, and other sources. But the process of adding characters to Unicode is ongoing.
It always makes good copy to blame the evil technoids for trampling human culture underfoot, but the real story here is the developers, managers, and engineers who are dedicated and committed to breaking out of the confines of '70s and '80s ASCII-centric technology. Unicode is a real and practical step in that direction, and it retains the potential to grow along with the needs of its users.
Asmus Freytag
The Unicode Consortium
asmus_freytag@unicode.org
RoboFarm
What's the agricultural revolution all about if not getting smarter? ("Weed-Whacking on Smart Farms," Wired 4.10, page 160.)
It will be a long time before a robot outthinks an illiterate, poorly trained lackwit farmhand when it comes to differentiating tomato seedlings from weeds. But give him a VR helmet and gloves that let him control a simple digging robot in the field from the comfort of an air-conditioned factory, and you've got cheaper produce and better working conditions for the laborer.
John Wiley
jtwiley@ingr.com
Undo
Grim News: The photo of Nicholas Grimshaw's Western Morning News building ("Tightrope," Wired 4.11, page 188) should have been credited to Jo Reid and John Peck.
Send your Rants & Raves to:
E-mail: rants@Wired.com
Snail mail: Wired, PO Box 191826
San Francisco, CA 94109-9866
