On Monday 8/25 a Federal judge in San Francisco ruled that the Commerce Department's export controls on encryption products violate the First Amendment's guarantees of free speech. In a 35-page finding [1], U.S. District Judge Marilyn Hall Patel reaffirmed her December 1996 decision [2] against the regulations then administered by the State Department, saying that the newer Commerce Department rules suffer from similar constitutional flaws.

Patel barred the government from "threatening, detaining, prosecuting, discouraging, or otherwise interfering with" anyone "who uses, discusses, or publishes or seeks to use, discuss or publish plaintiff's encryption programs and related materials." Daniel Bernstein, now a math professor at the University of Illinois, filed the lawsuit in 1995 with the help of the Electronic Frontier Foundation.

On 8/28, at the request of Justice Department lawyers, Judge Patel issued a stay of this injunctive relief, but said she will reinstate the part of the injunction that allows Professor Bernstein himself to discuss and publish his own "Snuffle" encryption code.

How significant is this ruling? A concensus is emerging that the judgement was worded too narrowly to have much impact in the short term, but if upheld could be a significant precedent in future cases. The decision doesn't block enforcement of U.S. crypto export rules, but permanently bars the government from trying to stop Bernstein or anyone else from posting or discussing his particular cryptographic program, Snuffle. The judge delivered a mild slap to the Clinton administration for moving encryption export licenses from the State to Commerce Department in December. "The government cannot avoid the constitutional deficiencies of its regulations by rotating oversight of them from department to department," she wrote.

A sampling from the flood of sound bytes captured by the online press after last Monday's ruling:

"The bottom line is that Bernstein wins," said Michael
Froomkin, an associate professor specializing in encryption
at the University of Miami's School of Law. "But I'm not
sure this would apply to a commercial product." [3]

Stewart Baker, the former general counsel at the National
Security Agency and a partner at the law firm Steptoe &
Johnson, concurred that if upheld, the decisions would
likely cause the restrictions to crumble. "I would expect
lower courts to have great difficulty distinguishing PGP's
publication from what Bernstein has done with Snuffle," he
said. "The decisions may so weaken the strength of the
regulatory program that the program collapses." [3]

Officials at the Electronic Frontier Foundation went further,
saying in a statement "The decision knocks out a major part of
the Clinton administration's effort to force companies to de-
sign government surveillance into computers, telephones and
consumer electronics." They went on to call the ruling "a vic-
tory for free speech, academic freedom, human rights and the
prevention of crime." [4]

Here is other coverage of the story, in decreasing order of cluefulness: [5], [6], [7], [8].

[1] http://www.eff.org/pub/Legal/Cases/Bernstein_v_DoS/Legal/970825_decision.images/...
[2] http://www.tbtf.com/archive/1996-12-24.html
[3] http://www.news.com/News/Item/0,4,13745,00.html
[4] http://www.zdnet.com/zdnn/content/zdnn/0826/zdnn0007.html
[5] http://www.wired.com/news/news/politics/story/6397.html
[6] http://www.infoworld.com/cgi-bin/displayStory.pl?970826.eencrypt.htm
[7] http://www.yahoo.com/headlines/970826/tech/stories/encrypt_1.html
[8] http://www.nando.net/newsroom/ntn/info/082697/info1_21782_noframes.html

On 8/25 Marimba submitted its application, distribution, and replication protocols to the World Wide Web Consortium for consideration as standards, with the backing of Novell, Netscape, and Sun [9], [10]. Together these protocols form the base of Marimba's "push" technology; they make it possible to reduce Net traffic by sending only the bits that change over the wires. The next day Novadigm filed suit [11]. This company claims, effectively, to have patented the techniques known in the Unix world as diff and rdist. Can you say "prior art?" Novadigm's domain name was registered in 1994. diff comes from Version 6 UNIX (Bell Labs), 1976; rdist was first released with 4.3BSD (Berkeley CSRG), 1985-1986. Oh, and we'll also need Larry Wall's patch in order to apply the diffs automatically -- around 1985. Thanks to John Robert LoVerso <john at loverso dot southborough dot ma dot us> for the historical details -- but the interpretation is mine alone.

[9] http://192.215.107.71/wire/news/aug/0826marimba.html
[10] http://www.infoworld.com/cgi-bin/displayStory.pl?970827.enovadigm.htm
[11] http://www.novadigm.com/aB5.htm

The law firm of Oppedahl & Larson, which holds the domain name patents.com, has filed a lawsuit [12] against three companies and the ISPs that provide them with Web space and name service. The suit claims that the three sites -- prowebsite.com [13], codeteam.com [14], and advancedconcepts.com (whose home page is currently blank) -- used the words of their trademarked name, "Oppedahl" and "Larson," in their meta tags on Web pages in order to draw traffic to their sites from search engines. None of the 11 URLs listed in the suit now displays any such words in meta tags or elsewhere (most in fact return Error 404 messages). The law firm discovered the offenses from Alta Vista searches (example at [15]), and has preserved the source code for each of the 11 claims on their site (see for example [16]).

After Declan McCullagh spread word of this case on his fight-censorship mailing list, Gant Redmon wrote of his own earlier experience in the same arena.

As counsel to Axent Technologies, I have already tagged a com-
petitor for putting our name in its meta file. It drew people
looking for us to our competitor. The rule in trademark law is
that you have violated a person's trademark when you use that
trademark to cause confusion in the marketplace. It says no-
thing about seeing the mark. What they are doing is intentional
and wrong. I was thrilled to shut down their deceptive activity.
It has nothing to do with freedom of expression. It has a lot
to do with being slimy.

[12] http://www.ljx.com/internet/
[13] http://www.prowebsite.com/
[14] http://www.codeteam.com/
[15] http://www.patents.com/ac/eb.sht
[16] http://www.patents.com/ac/ed.txt

The large US collectors and resellers of personal data on American citizens permit each person to buy a copy of his/her own record from time to time. (This is big of them, don't you think?) In California they are forbidden by law to charge for this personal service but in other states they do. One of the large data-collection companies, Experian (formerly TRW), set up a Web page from which people could pay using a credit card to view their own data. Sounds friendly so far. But they got the CGI code slightly wrong and at least four people viewed the private record of some other individual. Experian withdrew the online service within a few days. Dan Gillmor, computing editor at the SJ Mercury News, outlines this snafu and other recent encroachments on personal privacy in an article entitled Sale of personal data will grow until we rebel [17]. He writes:

The combination of the Net and increasingly powerful computers
and software properly feeds public angst about the availability
of personal information to all comers.

Two years ago Ram Avrahami sued U.S. News & World Report under the laws of the state of Virginia for selling his name to Smithsonian magazine [18]. He lost, partly because he didn't prove to the judge's satisfaction that his name had value, and partly because he had used a variant of his name to find out who had sold his personal data. Now Avrahami plans to found an advocacy group, to be called The Named, to assert the individual's right to control the use of his/her personal data. This effort lines up with the American Civil Liberties Union's launch last month of its Take Back Your Data campaign [19].

[17] http://www.sjmercury.com/business/gillmor/docs/dg082497.htm
[18] http://www.tbtf.com/archive/1995-10-24.html
[19] http://www.aclu.org/action/tbyd.html

The Internet is still expanding, but by one measure its period of exponential growth may be over. Tony Rutkowski of Net Wizards has published a twice-yearly snapshot of Internet growth [20] since 1990. Though the current graph [21] looks exponential to my eye, Rutkowski says that close analysis reveals since January 1996 a heeling over of the growth curve from exponential to linear. A good least-squares fit puts the growth rate at 52% per year -- 18,339 new hosts per day -- during this period. Rutkowski attributes the slowdown to the increasing tendency to hide large numbers of hosts behind firewalls, to outsource services, and to share common hosts. Rutkowski writes:

As the Internet scaled through its seventh order of magnitude,
it was apparent that the growth could not remain exponential
indefinitely.

These data speak only to the number of directly reachable computers on the Internet, and don't reflect the numbers of users, domain names, or Web servers. Growth in Web servers continues on an exponential trend, currently at an annual rate of 256%, according to the same report.

[20] http://www.nw.com/zone/WWW/report.html
[21] http://www.nw.com/zone/hosts.gif

LGMs are little green men. The first pulsar -- now understood to be a rapidly rotating neutron star -- was discovered by Jocelyn Bell in 1967 at Jodrell Bank observatory in England. (Her supervisor Anthony Hewish shared the 1974 Nobel Prize for the discovery, but Bell was not recognized.) The scientists withheld publication for 7 months because they had no explanation for the regular, beacon-like signal. They joked among themselves that its cause must be little green men.

The sober search for extraterrestrial intelligence, called SETI, is now approaching its 38th year. The radio telescope in Green Bank, West Virginia that Frank Drake first employed in the hunt in 1961 is again on the job; and it's being joined by the 300-meter instrument [22] built into a tropical valley in Arecibo, Puerto Rico. Drake's modest Project Ozma inflamed the imagination of Carl Sagan, then a young scientist, who five years later co-authored Intelligent Life In the Universe [23] with the Russian astronomer I. S. Shklovskii. (Originally Sagan signed on as editor of the English translation of Shklovskii's book Universe, Life, Mind, but his notes and additions grew so copious -- finally equalling the original in bulk -- that Sagan was promoted to co-author.)

For a short while the US government funded SETI research, but all current US projects are privately supported. Ongoing work includes the SETI Institute's Project Phoenix [24] and the UC Berkeley-based SERENDIP [25]. Working with data from the latter effort, a group called SETI @ Home [26] is developing software to harness the spare cycles of hundreds of thousands of Internet-connected computers to look for a needle of signal in a haystack the size of the galaxy. This FAQ explains how it will work [28]. Data will start flowing to specially developed screensavers on networked computers next spring. By the time 50,000 PCs are involved, the scope of the search will rival all current SETI projects. The SETI @ home page [29] ends with a plea:

Aliens: If you're reading this, you can save us a lot of
trouble with one simple email!

[22] http://www.naic.edu/
[23] http://www.amazon.com/exec/obidos/ISBN=0816279136/tbtfA/
[24] http://www.seti-inst.edu/
[25] http://sag-www.ssl.berkeley.edu/serendip/
[26] http://setiathome.ssl.berkeley.edu/
[28] http://setiathome.ssl.berkeley.edu/faq.html

Today's TBTF title comes from the Edward Lear nonsense poem The Jumblies, which begins

The Lear verse forms the lead-in to chapter 29 of Intelligent Life in the Universe [23], which walks the reader through the now-classic estimation of the number of technological civilizations in our galaxy. (Result: using guesses that looked reasonable in 1966, somewhere between 50,000 and 1 million, separated on average by a few hundred to 1,000 light-years.) To Sagan, apparently Lear's rhyme encapsulated the folly of casting fragile living bodies into the void when radio signals would serve nearly as well.

This morning TBTF welcomed its first subscriber from Kenya -- the second on the African continent, that I'm aware of, outside of South Africa. (My brother reads TBTF in Cameroon. He used to dial a US number to read mail, but there is now a local ISP in Yaounde.) By the time you read this the direct email subscriber base will likely exceed 4,000, in 72 countries [30].

[30] http://www.tbtf.com/growth.html

For a complete list of TBTF's (mostly email) sources, see http://www.tbtf.com/sources.html.

fight-censorship-- mail fight-censorship-announce-request@vorlon.mit.edu without subject and with message: subscribe . Web home at http://www.eff.org/~declan/fc/.

TBTF home and archive at http://www.tbtf.com/ . To subscribe send
the message "subscribe" to tbtf-request@world.std.com. TBTF is
Copyright 1994-1997 by Keith Dawson, <dawson at world dot std dot com>. Com-
mercial use prohibited. For non-commercial purposes please forward,
post, and link as you see fit.
_______________________________________________
Keith Dawson    dawson at world dot std dot com
Layer of ash separates morning and evening milk.