Tuesday, September 30, 2003


Consumer Confidence Drops

The Wall Street Journal reports a slight decline in consumer confidence this month:

Monday, September 29, 2003


The ten most censored stories of the year

Project Censored brings us the 25 most censored stories of the year. Here are the top ten:

#1: The Neoconservative Plan for Global Dominance
#2: Homeland Security Threatens Civil Liberty
#3: US Illegally Removes Pages from Iraq U.N. Report
#4: Rumsfeld's Plan to Provoke Terrorists
#5: The Effort to Make Unions Disappear
#6: Closing Access to Information Technology
#7: Treaty Busting by the United States
#8: US/British Forces Continue Use of Depleted Uranium Weapons Despite Massive Evidence of Negative Health Effects
#9: In Afghanistan: Poverty, Women's Rights, and Civil Disruption Worse than Ever
#10: Africa Faces Threat of New Colonialism
I think they missed a big one. When over 40% of the inhabitants of the United States still believe that Saddam Hussein was DIRECTLY involved in the September 11, 2001 attacks, then the U.S. press is not doing its job whether you call it censorship or not.


AT&T's 'Multiple Personality Disorder' gets worse

In Scott Bradner's recent Network World column, he asks, "Can AT&T Learn to be Stupid?" Bradner points out that on the one hand AT&T will spend US$3 billion over the next few years to build an all-Internet core network. Meanwhile, at the same time, it is building "Vonage-like" Internet voice communications apps that will run over *anybody's* network.

Hmmm. Let's see. If these Internet communications apps will run over anybody's network, why is it so all-fired, three-billion-dollars important for AT&T to have an upgraded network?

For comparison, an arm of Singapore Telecom is buying 61% of Global Crossing's shiny new global fiber network for a mere $250 million. I recently spoke to a Global Crossing employee who sand that after you read all the fine print, Global Crossing's network is valued today at about US$600 million. You can run all the Internet communications apps you want on it. In contrast, AT&T's current network is neither new nor global.

The question is not, "Can AT&T learn to be Stupid?" The question is whether AT&T will decide to be a network connectivity and transport provider or a network applications provider after a lifetime when these two functions were inextricably intertwined. The Internet has rent these two functions apart. The imperative has been clear for seven years. If you want to provide network connectivity and transport, you've gotta carry everybody's apps. And if you want to provide communications apps, they've gotta work on everybody's network. If you do both, each of these distinct businesses will be seen as competition by the customers of the other one. That's the dilemma AT&T faces.


More on 'Steal this Election'

Warren Slocum is the Chief Elections Officer & County Clerk-Recorder-Assessor of San Mateo County. He's concerned about the way verification was designed out of today's voting machines. He's started a verifiable voting blog -- an informed insider's view of what's happening. Not a pretty picture.

Here's a Salon piece on funny business within the IEEE voting machine standards committee.

And here's the Verified Voting website.

Thanks to David W for all these pointers.


Scatt Oddams recommends: WiFi Speed Spray™

Is your wireless connection slooooow? Maybe you need WiFi Speed Spray!
"Usually five or six sprays is all it takes. As your computer sends data, each bit also carries hundreds of invisible WiFi Speed Spray™ "scrubbing" molecules. It works at the speed of light. and even penetrates lead walls (not even Superman can do that!). Within .0025 seconds, the entire path between you and the receiver is cleaned, scrubbed, polished, and sanitized . . . When you notice that things are slowing down, just grab your handy WiFi Speed Spray™ and reapply. It's easy!"
A shpritz of thanks to Steve Crandall for the link.

Thursday, September 25, 2003


Steal the California Recall Election?

In an article in today's New York Times, entitled "Report Raises Electronic Vote Security Issues," James C. DiPaula, [of Maryland's] Department of Management and Budget, recommended to Gov. Robert L. Ehrlich Jr. that the state advance a plan [that] he said, "will correct specific risk factors and ensure reliability of the election process."

[Note: no links to NYT articles because the links are temporary -- David I]

The article also quotes whistle-blowing Johns Hopkins computer scientist Aviel Rubin, who first raised concerns with Maryland's Diebold machines: "It almost seems as though the people writing the Maryland action plan either did not read or did not understand the S.A.I.C. report. What they should say is, `We're going to put these systems on hold until they say that these things are safe to use.' "

The website BlackBoxVoting.org says that "encryption and password upgrades will be made only for the machines destined for Maryland and would not be available for the 33,000 touch-screen machines already in use elsewhere," according to Diebold executive Mark Radke.

"Elsewhere" (according to BlackBoxVoting) includes California, where 14 counties will be using Diebold machines in a few days for the recall race.

Recently the ACLU tried to halt the Calif Recall because of problems with paper ballots. This would be MUCH worse, if true.

An article in the Seattle Times on Diebold machines in King County indicates that officials are taking the Diebold situation much more seriously than the Maryland happytalk spinmeisters. It also contains some very interesting detail about alleged Diebold security vulnerabilities -- and an allegation that another whistle blower, Bev Harris, had her site summarily taken off the 'net "after a Diebold attorney said she was violating the company's copyright by posting a link to a New Zealand site that contained 15,000 pieces of Diebold e-mail."


Hard Times in Telcoland, cont'd

Verizon just revised its forecast in the usual direction -- down. And it cut its plan for capital expenditures by US$500 million. That won't make telecom equipment suppliers happy. And it filed for a US$10 billion self-dilution. That won't make VZ shareholders happy.

Then IDT announced that it is offering a US$40 UNE-P based all-u-can-call package in four states, with others to follow. That won't make Verizon, Sprint and AT&T happy; IDT is undercutting them by 20%.

Meanwhile, in a great example of how customers can run their own network services, Dartmouth University is making long distance calling free for its faculty and students. That won't make any telco happy.

Warren Buffet has sold off half his stake in Level3 since June. Telecom reporter (and friend) Scott Moritz says:
"After a brief period of stability, wholesale phone capacity prices once again appear to be in free fall. Here's why: A stampede into what was one of telecom's few remaining high-margin businesses has reignited a damaging price war. It wasn't supposed to be like this. After a three-year plunge in business communications spending resulted in a raft of bankruptcy filings, the industry was finally beginning to breathe easier, its bloodiest battles behind it.
[But w]holesale rivals are back from the brink, and so-called service integrators have entered its turf. It seems some failed foes have returned with a vengeance: Restructured outfits such as WilTel (WTEL:Nasdaq) and ICG have emerged from bankruptcy, wielding a sharp knife on capacity prices. Not to be outdone, AT&T (T:NYSE) , Qwest (Q:NYSE) and Sprint (FON:NYSE) have been willing to take lower terms to keep customers."
At least some customers should be happy.

I think the last year has been a telecom echo-bubble. The longer term trend for telephone companies is down. Here's why: as technology improves, we need them less and less.

The over-crowded long haul market is a different story. Even companies that are all-optical and all-IP will suffer until the U.S. gets a broadband access policy that's shaped more by technological progress than by dying incumbents.


More on 'Steal This Election'

Today's Washington post says, Md. Plans Vote System Fixes After Criticisms (washingtonpost.com)

Spinmeisters at work:
An independent review released yesterday found 328 security weaknesses, 26 of them critical, in the computerized voting system Maryland has just purchased, flaws that could leave elections open to tampering or allow software glitches to go undetected . . . 'Because of this report, Maryland voters will have one of the safest election environments in the nation,' said Gov. Robert L. Ehrlich Jr. (R), who last month ordered the review by computer security experts Science Application International Corp . . . [Voting machine maker] Diebold executive Mark Radke [said] "The electorate throughout the entire country should be very comfortable with the security of our system."
Whew! I am so relieved that they found (all of the?) 328 security weaknesses. The system that created these weaknesses is now fixed, and I don't have to worry anymore.

The report was a reaction to "an explosive" report by Johns Hopkins computer scientist Aviel Rubin after Rubin found Diebold's source code that had "mistakenly been left on an open Internet site." Rubin "excoriated Diebold's software designers, who had built passwords such as 1111 into the machines, and said he would have flunked them in basic computer security classes."

The "328 security weaknesses" report was written by government contractor SAIC. The Washington Post quoted from it as follows:
"The system, as implemented in policy, procedure, and technology, is at high risk of compromise . . . Unencrypted information could be intercepted and released prematurely or altered . . . failure to [train poll workers in computer security [WTF?!? -- David I]] makes it significantly more likely that an intruder's actions will not be detected . . . exploitation of any of the resultant security holes could lead to voting results being released too soon, altered or destroyed."
The SAIC report was "heavily redacted" -- that means that they blacked out the really good parts so only the "good guys" at Diebold and SAIC could see them.

Thanks to Jim Warren for pointing me to this article, and to Tony Cafferty, who originally posted a pointed pointer on FOI-L.


Either I am impolite, or . . .

. . . VoIP delay is still a huge problem. I just completed a phone call where I was on my Sprint PCS (mobile) phone and the other party was using Vonage. We spent the whole conversation stepping on the beginnings of each others' sentences, then apologizing and saying, "you go." I constantly had to remind myself not to interrupt, and so I would wait until I thought he was done, then I'd try to speak just as he was saying something else. We almost had to use the old half-duplex radio protocol, over. Alright, I've completed my thought, over. The delay is too long, over. I hope the telecommunications revolution isn't, over. Roll, over.


More on Cloudshield

Andrew Odlyzko writes that application-level price discrimination
" . . . is really equivalent to letting the carrier be vertically integrated and controlling everything. After all, if you control pricing, you control everything."
But Odlyzko doubts that looking inside every packet (a la Cloudshield) will provide a feasible basis for price discrimination. He continues:
"The only place that there is some argument for fine-scale control (and so where tools [like Cloudshield] might possibly take hold) is at the edges. It really boggles the mind to see anyone seriously imagine that (i) a backbone could really effectively examine/classify/price/control the millions of flows it handles and (ii) that even if this was technically feasible, it could be done in what is necessarily a small commodity business where the access carriers would have no interest in allowing the backbone carriers to do this."
He comments,
"I don't think this is all that scary. Silly is more like it."
Well I dunno. I'm aware of at least one research program on how to use the data that Cloudshield makes accessible, and it is funded by "the U.S. government," (further details withheld (from me)). Such research might decrease the silly factor in the near future.


Riverbend's Blog

Riverbend is an Iraqi blog that's at least as well-written, in as genuine a voice, as Salam Pax's blog. (Thanks to PEA for the link!) I've just added it to my linkroll. Here's a short excerpt:
"The whole neighborhood knows about S. who lives exactly two streets away. He’s what is called a ‘merchant’ or ‘tajir’. He likes to call himself a ‘businessman’. For the last six years, S. has worked with the Ministry of Oil, importing spare parts for oil tankers under the surveillance and guidelines of the “Food for Oil Program”. In early March, all contracts were put ‘on hold’ in expectation of the war. Thousands of contracts with international companies were either cancelled or postponed.

"S. was in a frenzy: he had a shipment of engines coming in from a certain country and they were ‘waiting on the border’. Everywhere he went, he chain-smoked one cigarette after another and talked of ‘letters of credit’, ‘comm. numbers’, and nasty truck drivers who were getting impatient.

"After the war, the CPA [Coalition Provisional Authority, that's Paul Bremmer & Co. -- David I] decided that certain contracts would be approved. The contracts that had priority over the rest were the contracts that were going to get the oil pumping again. S. was lucky- his engines were going to find their way through… hopefully.

"Unfortunately, every time he tried to get the go-ahead to bring in the engines, he was sent from person to person until he found himself, and his engines, tangled up in a bureaucratic mess in-between the CPA, the Ministry of Oil and the UNOPS. By the time things were somewhat sorted out, and he was communicating directly with the Ministry of Oil, he was given a ‘tip’. He was told that he shouldn’t bother doing anything if he wasn’t known to KBR. If KBR didn’t approve of him, or recommend him, he needn’t bother with anything.

"For a week, the whole neighborhood was discussing the KBR. Who were they? What did they do? We all had our own speculations… E. said it was probably some sort of committee like the CPA, but in charge of the contracts or reconstruction of the oil infrastructure. I expected it was probably another company- but where was it from? Was it Russian? Was it French? It didn’t matter so long as it wasn’t Halliburton or Bechtel. It was a fresh new name or, at least, a fresh new set of initials. Well, it was ‘fresh’ for a whole half-hour until curiosity got the better of me and I looked it up on the internet.

"KBR stands for Kellogg, Brown and Root . . . "

Wednesday, September 24, 2003


Cloudshield -- Scary at Layers Two through Eight

Cloudshield is a packet analyzer that can look at the contents of packets at line speeds apparently up to OC-48 -- and maybe faster if we look behind the black curtain. From Cloudshield's website:
"The Internet will choke under its own success if intelligence continues to be relegated only to the edge of the network. The notion that networks should remain 'dumb' and simply perform transport is outdated. Deploying certain application functions closer to the network core, instead of solely at the edge, relieves pressure on downstream access devices and applications, and allows the network to be more efficient, manageable, resilient and secure.
. . .
"CloudShield develops innovative solutions for carriers, service providers, large enterprises and government organizations to reduce the cost of operating networks, improve efficiency, and provide more granular control over content. CloudShield's Packet Server and RAVE software development environment offer a platform for 'Cloud-based' network solutions that can inspect and manage packet contents (layer 2 through 7) at multi-gigabit network speeds."
I can get over the fact that Cloudshield is ignorantly trash-talking the Stupid Network.

Worse, Cloudshield is FUDding with the tired pseudo-spectre of Internet collapse.

Even more egregiously, Cloudshield is trying to implement security in the wrong place. If "good guys" could know the packet contents of every packet on the network, what would they use this knowledge for? They still won't be able to distinguish an innocent "Let's take the 7AM flight tomorrow," from a guilty one. And they won't detect the criminal actions of Kenny Boy Lay or Gary "Global Double Crossing" Winnick. Or find Weapons of Mass Destruction where there aren't any. But they *will* be able to tell audio files from text files, and then they'll be able to differentially block, or charge different rates for the two types of traffic. Will Cloudshield networks block unknown media types? If so, at what cost to innovation? Will certain keywords (or, more abstractly, certain ideas) trigger different kinds of actions on Cloudshield networks? If so, at what cost to our Rights to Free Speech, Privacy, Freedom of Religion, Freedom of the Press? Bad guys will find a work-around, but the rest of us might be sorely stifled. I don't see one word about the Bill of Rights on the Cloudshield web site.


More on Zipf volitility

Ben Hyde (bhyde at pobox dot com) writes:
The is a little data about zipf volatility for incomes, but it is almost useless because it deals with quintiles so there isn't enough resolution to really think clearly about it.

The most interesting data about zipf volatility is from firm size data. Firms are power-law distributed along most metrics [1] (income, size, employees, etc). There is also data on year to year variation in their rankings.

So the data shows that grow is largely independent of size; i.e. there doesn't appear to be tendency for firms of size A to do better or worse than firms of size B.

But, there is a significant variance of growth rates. Small firms have extremely variable growth rates, and huge firms have little variation.
I.e. large firms tend to grow with the economy and there is little variation from that, while small firms suffer huge variations from one time frame to the the the next. The distribution of the variance is quite sharp, a double exponential.

I think this is another example of the scale free nature of these things - i.e. there isn't a typical firm, blog, website. You can't sample your neighbors to get an accurate model of the behavior of the entire set.

[1] http://linkage.rockefeller.edu/wli/zipf/axtell01.pdf

Saturday, September 20, 2003


Girls Gone Wild For Dean

Girls Gone Wild For Dean: "we're women who are wild about Gov. Howard Dean, M.D. The only thing we'll ever flash is our voter registration cards. "

Dan oogled this link first.


Scatt Oddams found this great cartoon . . .

. . . on RIAA copyright enforcement. Scatt thanks Kevin Werbach for the pointer.


Steal this Election?

The Electronic Frontier Foundation is publicizing what, apparently, is an effort by voting machine makers to railroad a voting machine operating standard through the IEEE without any requirement to independently verify the vote count. Further, the EFF says that they have reports of procedural irregularities in governance and membership on the relevant committees aimed at keeping out IEEE members who aren't good ol boys.

This comes on the heels of an August 2003 report of heavy handed behavior by the voting machine industry, which said, in part:
"Dr. Rebecca Mercuri, a leading expert in voting machine security, had her conference credentials revoked by the president of the International Association of Clerks, Records, Election Officials, and Treasurers (IACREOT), Marianne Rickenbach. The annual IACREOT Conference and Trade Show, [] showcases election systems to elections officials . . .

Mercuri believes that her credentials were revoked because of her position in favor of voter-verified paper ballots for computerized election systems. 'I guess in a very troubling way it makes sense that an organization like IACREOT, that supports paperless computerized voting systems, which are secret by their very design, would not want computer experts who disagree with that position at their meetings.'
. . .
David Chaum, the inventor of eCash and a member of Mercuri's 'voter-verified paper ballot' group, had his credentials revoked on the first day of the conference. On the second day his credentials were partially restored. Chaum was allowed to visit the exhibitors hall, but not attend the IACREOT meetings."
Dr. Mercuri maintains a page on e-voting problems.

The voting machine industry, the folks who are reported to have kicked the activist experts out of their trade show and railroaded the standard through the IEEE, propose that we trust them to count our votes -- without anybody watching, without anybody able to watch, with no capability even to determine whether the software on a particular voting machine is working as it should be. Brrrrrr.

Friday, September 19, 2003


Quote of Note

"I'm not doing this to save money. I'm doing this because the music industry is not giving me what I want."
Steve Vaughan, M.D. on his own music file downloading activities, in the New York Times, September 19, 2003 p. C2.


Quote of Note

"In the Middle Ages, when people were convinced there were witches, they certainly found them."
Hans Blix, recently retired UN senior weapons inspector, on American and British allegations of Iraqi weapons of mass destruction, in the New York Times, September 19, 2003 p. A13.


Library Snooping Powers *Never* Used -- Really?

According to the New York Times, September 19, 2003 p. A20, the U.S. Justice Department says it has used its "newfound counterterrorism powers to demand records from libraries and elsewhere," granted by the USA-PATRIOT act exactly zero times.

I thought that the Justice Department needed all those draconian anti-free-speech Patriot Powers to fight terrorists. Either they're not fighting terrorists or they don't need the new powers. Which is it, Ashcroft?

Thursday, September 18, 2003


RIAA in Abundance-land

Tim Oren's Due Diligence blog has one of the smartest critiques of how the recording industry is blowing its Internet opportunity.
"It's amazing that it's even a question that artificial scarcity is a stupid idea. How much more obvious should that be, when the customers already know there's an alternative, and have it in their hands?
Tim clearly specifies some of the ways the Recording Industry can make a post-scarcity buck:
" . . . [N]o more stacks of polycarbonate discs, access to a world of choice instantly, removal of most of the risks of format obsolescence. Even from the selling side there are advantages: no more inventory holding costs, more packaging flexibility, no more takebacks on unsold product . . . Abundance presents its own opportunities to create customer value. Google, just to take an example, owns an insignificant amount of content, but is valuable because of the overwhelming abundance of it on the Internet. The music fan in a world of abundance has problems as well. How do I organize my songs? Is there new music I should know about? Are any of my favorite artists coming to town, and how do I get tickets? Are there new acts that I might like? Where can I find like-minded fans to hang out with? How do I make sure I can get the music I want anytime, anywhere, on any gadget, and don't lose it? All of these are opportunities to create value, and potential ways to extract revenue and rebuild a business model in a post-scarcity world."
Tim has "this quaint idea that the idea of a business is to add value. And that the value is judged by the customer, not the seller." Golly, what a strange thought.

Tim cites a previous rant on a similar subject by Kevin Marks, who's also wicked smart and seriously articulate.


Word Pirates

Dan Gillmor and David Weinberger have launched Word Pirates, a site designed to identify when "short-sighted, self-interested, sticky-fingered people have been stealing our words."

The site invites visitors to contribute their own misused words, and the words are pouring in. Today some 50 words were submitted, and it is only 11:30 AM!

Yesterday I added competition and foreign.


Today's Optimization, Tomorrow's Bottleneck

One of Mile O'Dell's aphorisms that has stood the test of time is, "Today's optimization is tomorrow's bottleneck."

This BBC article, Goodbye to a flat rate for broadband?, says:
"'The last four or five years has been about building this infrastructure of a high-speed network, providing a dumb access,' says Milind Gadekar, vice-president of P-Cube, whose monitoring system is being tested by service providers in Europe. Now, he says, service providers need to make their networks 'intelligent' so they can identify users and the applications used."
Just what I want -- a network that identifies the application I'm using!

I can imagine

Nicholas Negroponte has been talking about how some bits are more valuable than others for years. For example, we'd pay a lot for the "Is it malignant?" bit or the "Got the promotion?" bit. Is it far-fetched to imagine that an application aware network operator would charge a premium for such bits?

It is certainly plausible that an application aware network operator might charge differently for, say, telephony than for, say, streaming audio. This is fine as long as we don't mind that some third party is encouraging certain applications and discouraging others (with motivations that might not be transparent). It is also fine if we already know what all the applications are on our network, and if the encouragement/discouragement process doesn't make it harder to add new ones. In other words, NOT fine.

Andrew Odlyzko has a well-thought-out paper that points out that sellers often have strong motives to do "price discrimination," that is, to charge different prices to different users at different times (e.g., airline yield management, or even end-of-season sales). He observes that economists think price discrimination improves economic efficiency, but that customers often hate it.

We need to accept that price discrimination happens, and focus on where it happens, for what and to whom. According to the End-to-End Principle, if you have a choice to put a function at the edge of the network or in the middle, you should put it at the edge. Price discrimination in the middle of the network is a risk to new app discovery and to free speech. We should keep the network stupid -- and put the "for what" and "to whom" of price discrimination at the edge.

Wednesday, September 17, 2003


More on Terror Laws and Common Crimes

Arnold Kling in Corante: The Bottom Line cites Marginal Revolution in an article called Transparency and Patriot-ism:
First, the credit card companies caved into government pressure and refused to process gambling related transactions. Initially, gamblers shrugged this off and routed their transactions through PayPal but a U.S District Attorney accused PayPal of violating the USA Patriot Act and to avoid charges PayPal was forced to pony up 10 million dollars. (Why am I not surprised that a law intended to go after terrorists has been used to most affect against peaceful gamblers?).
Kling comments, "Any law that is passed in the name of stopping terrorism should be written so that it only applies to terrorism. How hard is that to do?"

My comment: Yes! But it *will* be hard, which is part of the problem with these new laws. And the law and order people (i.e., the police, the prosecutors and the courts) are going to have to *want* to do it, which is the other part.


Free Wi-Fi Hot Spot Access on September 25

Intel is sponsoring a "One Unwired Day" on September 25 when you can connect at public hot spots without paying. Boingo, T-Mobile and Wayport are all participating. I think this is an absolutely great way to "grow the pie." I hope they don't screw it up by requiring a registration, then spamming the unconverted. If it remains a pure connect-and-enjoy experience, it'll be a great event.

If you miss it, or if you find the registration process too cumbersome, you can come over to my house, where every day is Free Wi-Fi Day.

Tuesday, September 16, 2003


Terror laws and common crimes

Salon.com: New terror laws used vs. common criminals: "A North Carolina county prosecutor charged a man accused of running a methamphetamine lab with breaking a new state law barring the manufacture of chemical weapons. If convicted, Martin Dwayne Miller could get 12 years to life in prison for a crime that usually brings about six months.
Prosecutor Jerry Wilson says he isn't abusing the law, which defines chemical weapons of mass destruction as 'any substance that is designed or has the capability to cause death or serious injury' and contains toxic chemicals. "

Hmmmm. Any cigarette manufacturers in North Carolina? Pesticide makers? Air polluters?

Thanks to Dan Gillmor for this link.


Howard Levy, harmonica genius

One of the themes of this blog (and my life) is that stuff can be awesomely wonderful, maybe world class, even when it lives on the long flat tail of Zipf's Law.

Howard Levy has been a musical hero of mine ever since I went, "Who *is* this guy," when I heard him play on a Steve Goodman record 20 or so years ago. I am delighted that Pete Kaminski and David Weinberger have caught Howard's Magic.

Howard's "Molinaro-Levy Project Live" recording (with pianist Anthony Molinaro) has suddenly vaulted into my personal top ten. It bubbles with magical readings of Gershwin, Miles and Ellington, plus originals by both musicians. This record is so "unpopular" it is not even listed on Amazon -- you have to get it at Nineteen-Eight records.

Visit Nineteen-Eight today -- it's about time!


Most Influential People in Communications?

For a couple more weeks, you can still vote for ten of the 20 finalist candidates for Total Telecom's Most Influential Person in Communications. The interim results at this moment show that Bill Gates is tied for the lead with Ivan Seidenberg (Verizon), Jorma Olllila (Nokia), Serge Tchuruk (Alcatel), and Li Ka-Shing (Hutchison Whampoa). Currently Dave Dorman and Mike Powell come in at #19 and #20, respectively.

Several weeks ago I held my nose and voted for a couple of TotalTele's sorry list of supposed influencers. To me, the most influential people in telecom start with Bell, Marconi, Elisha Gray, and the other early pioneers. Then there's a swatch of Internet pioneers (Baran, Kleinrock, Roberts, Cerf and their contempos) and a bunch of enabling technology inventors like Carver Mead and Bob Metcalfe. Bill Gates is in there somewhere, but saying that bean-counter herders like Seidenberg, Ollila and Li Ka-Shing are influential is like saying the tail is the key influencer of where the dog is going.

Wednesday, September 10, 2003


We're all journalists now

I was aggrieved by one paragraph in Jane Black's recent VOIP article in Business Week that *used to* say:
The Internet has changed all that. Since information now travels digitally -- a sequence of 1s and 0s -- no distinction remains between a voice call, an e-mail, or a video stream, and it costs no more if that information goes cross-town or cross-country.
I didn't think that digital transport has anything to do with it. After all, the telephone network uses digital transport, and this changes nothing. The thing that changes the Internet is the end-to-end principle (aka the stupid network), Moore's Law, and freedom from the legacy of obsolete regulations.

I wrote to Jane and suggested that it would be a great chance to educate Business Week's readers about end-to-end. She agreed, asked me to craft a better paragraph, and used my words verbatim. (She says that she usually doesn't do this, and it is certainly not Business Week's policy, but in this case, she agreed with my issue and felt that she could "own" it and stand behind it as if it were her own.)

The paragraph now reads:
The Internet has changed all that because it makes no distinction between a voice call, an e-mail, or a video stream. Digital technology improvements mean that it costs no more if that information goes cross-town or cross-country. Moreover, because the Internet is new, it isn't hobbled by regulations that were established when long distance meant higher costs.
Thanks to the Internet, articles can be corrected after they're released. I s'pose this could be a blessing or a curse. In this case, it seems to have worked out well.

Monday, September 08, 2003


Quote of Note

"[S]omeday, the phrase Internet telephony will sound as archaic as 'horseless carriage' sounds today."
Vint Cerf," in BW Online | September 8, 2003 | Time to Rewrite the Rules of Telecom by Jane Black.

Thursday, September 04, 2003


Zipf volatility

There's so much work that's good but isn't popular. That's why my first and second posts to this weblog were about non-blockbuster movies.

Clay Shirky has been pounding Zipf's Law as if the world were made of nails. Zipf's law states' roughly, that, "the 100th most popular web site would get 1/100th of the traffic of the most popular site (Yahoo), and the 2,000,000th most popular web site would see 1/2,000,000th of Yahoo's traffic." He's right, of course -- hits on websites and blogs are distributed much the same as viewership of TV shows -- a few sites and blogs and shows get very high ratings, and most don't. And he's right that there are, to a first approximation:
"three tiers [that] drive media strategy. The bottom tier -- the millions of sites with very little traffic -- will be composed almost entirely of labors of love, or expensive subscription-driven sites with very selective offerings. The middle tier -- say, the 1000th to 100,000th most popular sites -- will use its closer connection with its users' interests while banding together into networks, web rings and so on in order to aggregate their reach. The top tier, the Yahoos and Geocities of the world, may be victimized by its own success, managing increasing problems of supply and demand."

But I think there's a key property of works -- especially on-line works -- that Clay doesn't cover. I call it "Zipf volatility." Yahoo and Google are permanently popular; they have low Zipf volatility. But my hypothesis is that there's a middle tier of blogs with high Zipf volatility, where a well expressed idea or a funny story or a new factoid can rapidly catapult a blog from #100,000 to #1000, or in rare cases even to #10, in a matter of hours.

I am not sure how you'd test this idea experimentally (comments appreciated), and I am afraid that if you take 100 blogs, say between #100 and #200, and look at their delta-rank over a one week period, they might not look any different than the blogs between #20,000 and #20,100. Despite this caution, I strongly suspect that blog rank (and web site rank, to a lesser extent) has a burstiness that is not characteristic of other media, that permits new ideas (and new sites and blogs) to bubble up and subside, to move more readily than other media along the x-axis of Zipf's Law.

I could be convinced otherwise, but it'd take a good experiment.

This page is powered by Blogger. Isn't yours?