May 19, 2004

Techies wanna do policy

I'm sure we have all heard a techie or standards body tell legislatures, courts, and business groups to keep their mits off of the internet; that such groups are "clueless" and that they will damage some noumenon or other indistinct, but critical, principle of the internet.  Consider, for example, the condemnation of competing DNS roots by ICANN and the IAB.

What makes today so interesting is that two well respected techies have stepped forth and made strong social/economic/business policy statements.

Now it isn't that we techies are not capable of making good social policy statements or that we don't have a right to do so.  Rather the point is that such comments ought not to be given inordinate weight based on a presumption that those who are good at technical matters automatically are experts in economic, business, legal, or social matters.

Let's look first at the Tim Berners-Lee's document New Top Level Domains Considered Harmful

That document argues that new top level domain, particularly .mobi, would hurt the internet.  The logic of the document is essentially this: New top level domains would bring change and change causes discomfort, therefore change must be prevented.

And then let's look at Paul Vixie's note to the IETF today in which he asserted a legally actionable right to block ISP's from establishing a root server on a locally routed instance of address 192.5.5.241.  In essence he is asserting a kind of new legally cognizable proprietary interest over a common, freely offered, service.

Both Berners-Lee and and Vixie have contributed mightily to the internet; they have deservedly accumulated enormous techno-karma.

But no matter how many karma points they may have, they are repeating the behavior that turned the innovative telephone companies of the early 20th century into the stodgy telcos that fought competition and innovation at every turn from the early 1900's up through the 1970's.

I am old enough to remember the days when we in the US rented our telephones from "The Phone Company".  Those were the decades when the biggest new advance from "The Phone Company" seemed to be the Princess Phone and when long distance calls were expensive and often involved an operator.

And there were cases, such as the Hush-A-Phone case, in which The Phone Company fought tooth and nail against even the most benign of innovations.  And it wasn't just the lawyers and marketeers who were behind these anti-innovative efforts; there were legions of engineers who swore up and down on stacks of bibles that things like the Hush-A-Phone would destablise the internet of the era, the voice telephone system.

The engineers in the phone companies weren't naturally anti-innovation.  Rather they rightfully were very proud of what they had built.  And that pride led them to resent and resist further technical change.

We are observing the same kind of resistance to change and ossification of technology occurring in the internet.

What is Tim Berners-Lee's argument except a conclusory assertion that further innovation would be harmful?  And Vixie's statements to the IETF seems to be a kind of territorial marking to warn newcomers that they should neither innovate nor improve customer service if that might erode the F root server's existing user base.

The ability to innovate on the internet is rapidly disappearing.

It is bad enough that the end-to-end principle is being violated by things like Verisign's SiteFinder and by cable operators and ILECs who offer only discriminatory services that are designed to bolster their own offerings at the expense of their competitors.

But there is more beyond the craven moneygrab of things like SiteFinder.  A new wave of resistance to change is developing.  History has shown us that engineers (or those who employ engineers) use seem-like technical arguments as vehicles to promote economic, social, legal, or business goals.

Those who consider the weight of such arguments would do well to ask whether a seemingly technical argument on a business, economic, social, or legal matter is really a technical argument or simply an form of sophistry designed to give the appearance of technical imperative and disinterested evaluation to what is really just another social opinion.

Let me close by once again referring to The First Law of the Internet.  I believe that the First Law provides an appropriate framework in which to balance innovation against the costs of change.

Posted by karl at 4:32 PM

May 13, 2004

The ACLU and the "Justice" Department

I've recently become aware of the ACLU lawsuit over these "National Security Letters" - See http://www.aclu.org/SafeandFree/SafeandFree.cfm?ID=15543&c=262 and http://www.washingtonpost.com/wp-dyn/articles/A22404-2004May12.html

Apparently the "Justice" Department is taking a hard line on the disclosure of any information regarding these "National Security Letters".   And that triggered my discordian thought process to see what can be done within the letter of the law.

Well, thought I - the Patriot Act only says that those who receive such National Security Letters are forbidden to talk about it - which means that those of us who have not received any such letter are in no way encumbered from telling the world that we have not received one.

So, suppose every day I come onto my web site and add a line of the form: "It is now May 13, 2004, 6pm PDT, and as of this time I have not received a National Security Letter."

Then, should I receive a National Security Letter I would be required to stop saying anything on my web site.  And that absence of my daily comentary would be a signal that I have indeed received such a letter.

Of course, the "Justice" Department might demand that I continue publishing my daily (but now false) denial.  So to deal with that situation I take note of the fact that I am perfectly at liberty to make up fanciful stories.  So suppose, that instead of making a daily record of no National Security Letter landing on my doorstep I instead make up a daily story and say "Gee, It's 6pm PST on May 13, 2004 and I received a National Security Letter" today.

Then, should I actually receive such a letter, I would be precluded from saying that I did, and thus would be required to stop my daily storytelling, which, in turn, would announce, albeit indirectly but nevertheless quite clearly, that I did, in fact receive the dreaded letter.

Sherlock Holmes said it all:

"Is there any other point to which you would wish to draw my attention?"

"To the curious incident of the dog in the night-time."

"The dog did nothing in the night-time."

"That was the curious incident," remarked Sherlock Holmes.

Posted by karl at 5:50 PM

May 11, 2004

Keepin' the ol hog clean

I see today that ICANN held a special board meeting to adopt one resolution.

One meeting, one worldwide phone bridge, one resolution.  It must have been an important resolution.

Let's take a look at that resolution, or more particularly, let's look at the "whereas" clauses that establish the foundation upon which this resolution is based.  Any errors in those foundations have the effect of weakening the resolution itself.

The third whereas says:

Whereas, the competitive registrar marketplace introduced by ICANN in 1999 has been successful in driving down prices to consumers and businesses for gTLD domain registrations.

Now I find that rather a strange assertion.  In fact it is my strong belief that ICANN's policies over the years have created a price support system that keeps domain name prices to consumers artificially high and has cost domain name consumers hundreds of millions of dollars over the last five years.

ICANN has established registry price floors that serve no purpose except to guarantee registry profits and to limit the range of competition among registrars.  ICANN has established expensive processes that provide no value to domain name consumers and which have utterly no role in support of ICANN's chartered purpose, the 24x7 reliable operation of the internet's domain name and addressing systems.  ICANN has established a body or rules that are a system in restraint of free trade that protects the incumbent domain name registries and registrars from competition from new innovative providers.

This whereas clause is a deception.  It is a fraud on the public.  It is very sad that ICANN's board still fails to recognize how badly ICANN has screwed up by creating a domain name "marketplace" that more resembles a Soviet Five Year Plan than an exercise in the free interplay of economic forces.

ICANN's claim that it has introduced competition and that consumers are now paying domain name prices that truely reflect a competitive market is nothing but hogwash.

One of original claims made for the existance of ICANN was that it would bring competition into the domain name system.  There is no doubt that ICANN has introduced limited registrar-registrar competition and that prices today are less than were paid to Network Solutions back in the early 1990's.  But along the way ICANN has created a heavy regulatory system that adds to registrar costs and neither encourages nor permits registry cost savings.  The path to full competition and fully competitive prices is a long one.  ICANN took one small step, sat down, and claimed victory.

There is substantial reason to believe that without ICANN's regulatory system the competitive price for long-term registrations could be far less than one dollar per year.  Even if that estimate is off a factor of 5x the numbers are still far less than paid by today's domain name customer.

It is very sad that ICANN continues to pass off its price support system for domain name prices as if it were some sort of victory for competitive economics.  It is even more sad that the United States Department of Commerce, which one would assume has expertise in matters of real competition, chugs down ICANN's claims and asks for more.

Posted by karl at 4:27 PM

May 3, 2004

Manipulating Google's Gmail for fun and profit

Don't you just hate it how companies are using us, our knowledge, and our relationships to make money for themselves?  Consider Google's "Gmail" - when I send e-mail to a friend or associate who has an account on Gmail, Google's machines scan the text that I wrote and use that to present paid advertisements.  In effect, Google is data mining my own personal knowledge and relationship with my correspondents.

What's in it for me?  Nothing.  In fact, I run the risk of sending something that triggers an advertisement that really ticks off my correspondent on Gmail.

But there may be ways to turn things around.

Suppose we to establish a company that processes outgoing e-mail.

Suppose further that, just as search companies do today, we sell words and phrases.  For example, we might sell the word "pasta" to some company.

To induce users to send their outgoing e-mail we would pay them, yes actually pay them money, for the privilege of processing their outgoing e-mail.  (We'd put on rate limiters to keep spammers away and impose other caps to limit amounts paid to senders who are trying to artificially drive up the amounts we pay them.)

We then add text and other materials to the outgoing mail.  For email that's going to normal ISP based users, we'd do nothing more than what is typically done today - a header or footer containing links to advertisements.

But for things going to Gmail we could do several things.  One thing is that we could add text designed to trigger certain Gmail advertisements (presumably ones purchased by the same people who bought the word from us in the first place.)  Another thing is that we could try to rephrase the sender's text so that it is less likely to trigger Gmail advertisements by those who are not our customers.

This would turn into a competition between our ability to add and modify e-mails in ways that appropriately tickle Gmail's sensors and Gmail's ability to detect when it being manipulated.  It's sort of like the battle between spammers and spam filters.

The net result would be yet another step along the road of transforming the once useful system e-mail into nothing but bulk advertising.  But at least this time we users might be able to take a cut of the action.

Posted by karl at 12:56 PM