Tasty Bits from the Technology Front: TBTF for 1998-12-15: Meme pool

Proposed law would set up a Controller to monitor all Net traffic and Web sites

This note [1] was sent to me from India yesterday by Abhay Kushwaha. He forwards a newspaper story describing a draft bill that covers many facets of Internet policy, including provisions to monitor all Net traffic — whether plain-text or encrypted — passing through any Indian ISP. The sender of an encrypted communication would be required to decode it. This bill portends nothing good for the future of the Internet in India.

Kushwaha adds that he wants to hear what other Indians on the TBTF list have to say about this proposal. Please write to him directly at <abhay@kushwaha.com>.

Note added 1998-12-16: Eric Scheid <eric dot scheid at ironclad dot net dot au>, host of the discussion list InternetIndia-L, writes:

[The list] is for discussion of Internet policy in India. The membership is small but strong and I'm sure this issue would prompt much discussion... [For information and instructions on how to join see] http://www.ironclad.net.au/lists/.

Note added 1998-12-18: Udhay Shankar <udhay at pobox dot com> notes one flaw among many in the Indian government plan outlined above:

Another example of spectacular cluelessness strikes me here — how can the onus to decode rest with the sender? If I don't keep a copy of the plaintext, and if I have encrypted to your public key, I cannot decrypt it even if I wanted to.

Note added 1999-01-10: Udhay Shankar <udhay at pobox dot com> has posted an analysis of the proposed law and links to knowledgable commentary about its possible implementation and effectiveness.

[1] http://tbtf.com/resource/India-intercept.html

South Carolina drops out of the antitrust case

On 7 December the attorney general of South Carolina, Charles Condon, announced [2] that the state is withdrawing from the Microsoft antitrust case. Condon said that AOL's proposed acquisition of Netscape proves there is plenty of competition in the Internet space. Microsoft's legal team took advantage of the PR opportunity to hold an hour-long press conference from Washington, which featured a patched-in Bill Gates declaiming via satellite, "It's hard to believe the government is still pushing its case with a straight face." According to ABC News, Microsoft officials acknowledged [3] that the company had contributed $20,000 to the South Carolina Republican Party before last month's election, a sum a state party official described as one of the largest he had ever seen.

[2] http://www.msnbc.com/news/221412.asp#BODY
[3] http://abcnews.go.com/sections/tech/DailyNews/msdoj981207.html

Politically aware online advocacy the US could learn from

A new organization called STAND offers a Web site [4] where UK voters can "adopt" and "feed" their local politicians. Prompted by widely disparaged British government moves to limit the use of strong cryptography in that country, members of the UK new-media community — including Danny O'Brien of Need To Know [5] — have cranked up what looks to be a fairly savvy attempt to educate Members of Parliament to the damage they are about to do to nascent e-commerce. The STAND FAQ [6] gives a good backgrounder on the organizers and its aims. (Example: What does STAND stand for? A: It stands for itself). Readers outside of Britain can visit the links page [7] to bone up on developments in the local crypto debate. I'll thank Malcolm Hutty <malcolm at liberty dot org dot uk> for the tip, even though he is one of the organization's founders.

[4] http://www.stand.org.uk/
[5] http://www.ntk.net/
[6] http://www.stand.org.uk/faq.php3
[7] http://www.stand.org.uk/links.php3

Closely watched case may provide broader basis for patent appeal

Mary Ellen Zurko <mary_ellen_zurko at iris dot com> is the lead submitter on a software patent that will be reviewed by the Supreme Court in the 1999 session. At issue is not whether the patent is valid or invalid on the basis of prior art or any other such legal grounds. What the Supremes will decide is if it's OK to tell the Patent Office that they're just plain wrong. Large institutions that endure for a sufficiently long time breed a belief in their own infallibility. It seems that while you can tell the Patent Office that they acted in an "arbitrary and capricious manner," you can't base an appeal on a contention that they simply blew it. Here is Zurko's description of the situation.

When a patent application I co-authored with eight other colleagues was denied by the patent office (on the grounds of obviousness), then appealed by (then) Digital Equipment Corporation lawyers, I figured that was nothing remarkable (after all, convicted felons always appeal). It turns out not, in more ways than two. According to David Malakoff, who wrote an article in Science Magazine on the Lehman v. Zurko case accepted by the Supreme Court (Volume 282, Number 5394 Issue of 27 Nov 1998, p 1622), while over 100,000 patents are rejected in a year, fewer than 100 end up in appeals court.

The Supreme Court case has nothing to do with the technology in the patent (a method for reducing the amount of trusted code in a secure user interface). It's about whether the U.S. Court of Appeals for the Federal Circuit (the next step after the PTO's internal Board of Appeals) can find that the factual basis for a denial is "clearly in error." The Patent and Trademark software Office wants the appeals court to tell them to reconsider a patent rejection only if it finds the PTO acted in an "arbitrary and capricious" manner. The case is notable for its effect on patent law — several law schools were teaching it when it went to the Federal Court. The Science magazine article has more detail on the case, and I expect NPR commentary to be the best source of insight when the case is heard at the Supreme Court.

Current case law on downloading or linking copyrighted material

This National Law Journal article [8] runs through existing case law relating to fair use, linking, framing, and copying Net content, for example for use on an intranet. Of particular interest is the discussion of "deep links": links directly to content buried within a site, bypassing its top page (which presumably carries advertising). TBTF is strongly biased to use such "rifle-shot" links. The author cites the case of TicketMaster Corp. v. Microsoft Corp.-- and maddeningly does not report the actual outcome — in concluding

With respect to the company's use of hyperlinks on its intranet, legal scholarship on this issue... suggests that hyperlinks will not give rise to liability if the linked sites' home pages are the destinations. Liability appears likely to attach only if deep links are used.

Thanks to Monty Solomon for the NLJ cite.

[8] http://www.ljx.com/cgi-bin/f_cat?prod/...

Advice to newbies in the post-dead-trees age

Jargon Scout [9] is an irregular TBTF feature that aims to give you advance warning — preferably before Wired Magazine picks it up — of jargon that is just about ready to hatch into the Net's language. Anton Sherwood <antons at jps dot net> forwards the ringing phrase STFW, which he says he's seen several times on the newsgroup alt.fan.cecil-adams in response to trivial questions, meaning Search the flinking Web. The term is a cyberspace variant on the paper-based RTFM [10], though a more precisely analogous reading might be Surf the fine Website.

Julian Harris <jharris at clear dot co dot nz> claims to have originated the alternate form STFN. Usage:

- Do you know what the latest version of Crystal Reports is?
- Oh come on, STFN.

The Communications Decency Act

See also TBTF for 1999-02-01, 1998-12-15, 12-07, 10-27, 10-19, 10-12, 09-14, 07-27, 1997-11-17, 06-30, 03-21, more...

CDA II injunction extended

The initial 10-day injunction [11] forbidding enforcement of the Child Online Protection Act (called CDA II by its critics) has been extended by several months. With the agreement of both parties, a judge extended the temporary restraining order until mid-February [12] to give the government time to prepare its case.

[11] http://tbtf.com/archive/1998-12-07.html#s01
[12] http://www.aclu.org/features/f101698a.html

More consolidation of Net security firms

Continuing an agglomeration trend noted recently in TBTF [13], [14], over the last two weeks six security firms announced merger plans [15]. CyberSafe announced it has acquired Canadian encryption firm Sagus Security. Alladin Knowledge Systems purchased eSafe Technologies, an Israeli firm that makes software to block hostile Java applets. British encryption firm Zergo Holdings Plc said it's in merger talks with Baltimore Technologies, an Irish digital certificate company.

[13] http://tbtf.com/archive/1998-03-30.html#s01
[14] http://tbtf.com/archive/1998-02-23.html#s01
[15] http://www.news.com/News/Item/Textonly/0,25,29749,00.html?tbtf

Making a market in bandwidth

John Kristoff <jkristof at interaccess dot com> forwards news of two organizations matching buyers and sellers of bandwidth. Band-X [16], based in London, calls itself an "independent virtual market for trading international wholesale telecom capacity, minutes, or bandwidth." The original market may have been mediated by humans, but Band-X now offers buyers and sellers direct connection to its switch [17]. San Francisco-based RateXchange [18] says it's an "efficient marketplace for sellers and buyers of wholesale telecommunications capacity." Both sites require membership. RateXchange provides more visibility into its market for non-registered guests, listing weekly indicators for minutes and bandwidth [19] and spot prices by country [20].

[16] http://www.band-x.com/
[17] http://www.band-x.com/switched.cfm
[18] http://www.ratexchange.com/
[19] http://www.ratexchange.com/telechart/
[20] http://www.ratexchange.com/spot.html

Web been getting too serious for you lately? These sites offer endless hours of fun for the easily amused.

A dip into the meme pool

One of the regular sections of Need To Know [21] is titled Memepool. Perpetrator Danny O'Brien informs me that NTK invented the word before this site [22] came onto the scene, but says he isn't going to fight over it, because

there's something a bit sick about trying to impede the propagation of a word like "memepool."

Memepool.com [22] offers one or two strange links per day, each with a single phrase as description or teaser. Here I learned where you can go if you simply must ship a friend a Goliath Bird Eater tarantula [23]. Here I found crackpotologist Donna Kossy's entertaining Kook Museum [24], outgrowth of her book Kooks [25] (unfortunately out of print), outgrowth of Kooks magazine, outgrowth of the Kooks Pages in the early 'zine False Positives. Thanks to Karl Hakkaranen for this cite.

[21] http://www.ntk.net/
[22] http://memepool.com/
[23] http://www.tarantulas.com/goliath.htm
[24] http://www.teleport.com/~dkossy/index.html
[25] http://www.powells.com/cgi-bin/partner?partner_id=23196&cgi=search/search&searchtype=isbn&searchfor=0922915199

Ring tones

The Net is the polar opposite of broadcasting: we can each burble on about our favorite obsession and like-minded people will find us out, collect around our watering hole, and start to gossip. Got a Nokia phone? Bored with its built-in selection of ring tones? Here's one of several pages devoted to downloadable ring-tone resources [26] for the Nokia 8110i. This page is part of Simon Whitaker's Netcetera site [27], which also features a small collection [28] of his favorite .sig's, of which one happens to be mine [29]. Netcetera funnels me two or three visitors a day.

[26] http://www.netcetera.org/8110i/ringtones.html
[27] http://www.netcetera.org/
[28] http://www.netcetera.org/sigs.html
[29] http://dawson.nu/sig.html

Searching Pi

This page [30] claims to let you search for any given string of numbers in the first 50 million digits of Pi. Perhaps you need to be a particular flavor of math wonk to find this service fascinating, but if so then I'm that kind. Experimentation indicates that a randomly selected 7-digit number is likely to occur several times in the first 50M digits, while an 8-digit number has long odds against being found at all. (The script accepts up to 120 digits.) When a string is found you're presented with 20 digits of Pi's context on each side of it.

Subversive thought: of course the page could be a hoax and it would be difficult to tell. Perhaps the script behind it simply waits a variable amount of time depending on how far into the 50M digits it claims to have found a match and then prints your input string embedded in 40 digits of randomness. Since many people know a few digits at the beginning of Pi, the script might correctly store the first 120 digits so a simple test couldn't catch it out. It could remember the last few dozen generated answers so you couldn't easily spot it acting randomly. And so on.

Note added 1998-12-17:

David G Andersen <danderse at cs dot utah dot edu>, creator of Searching Pi [30], writes:
Bwahaha. I wish I'd thought of that — it probably would have been easier than actually writing the search engine (discovering, in the process, a few bugs in the way FreeBSD handles MMAPing 50-megabyte files. Fortunately, they've all been fixed). I'm fairly sure that my script is honest, but then again, I could just be saying that.

[30] http://www2.aros.net/~angio//pi_stuff/piquery.html

The Man Who Loved Only Numbers:
The Story of Paul Erdös and the Search for Mathematical Truth
by Paul Hoffman
Hyperion, 302 pages ($22.95 list; $16.07 at [31])

Reviewed by Alice D. Phalen <adp at world dot std.com>

This is an unusually graceful book that will engage even a minimally numerate reader. Paul Erdös was a mathematical prodigy who did productive work into his 84th year, writing or co-authoring 1,475 academic papers. He tailored his life to support his passion for mathematics: he travelled widely and maintained an even wider correspondence. A typical letter begins, "I am in Australia, tomorrow I leave for Hungary. Let k be the largest integer..." When I finished reading Paul Hoffman's graceful tapestry of anecdote, biographical data, and mathematical discourse, I immediately started to read it again for the sheer delight of Erdös. He was, at once, immensely serious about his work and utterly without self importance. It is true that he serenely depended on colleagues to attend to the logistics of food, shelter, and travel but he gave good value for this dependence. He collaborated with 485 mathematicians. He was generous with time and encouragement. Never wealthy, he was nevertheless as liberal with his money as he was with his time. He made loans to students, often turning aside repayment with the instruction to do as he had done. He donated to charities and causes and sent money to bereaved families of his colleagues. A lifelong celibate — a wife and children would have taken valuable time away from mathematics — he was fond of "epsilons," his word for small children. He was a loyal, compassionate friend; his heart was as great and open as his formidable mind.

[31] http://www.powells.com/cgi-bin/partner?partner_id=23196&cgi=search/search&searchtype=isbn&searchfor=0786863625

You connect the dots; you pick up the pieces

In this season when many news outlets are advancing projections for the Net in 1999, TBTF offers you the chance to show off your own predictive acumen. This quantitative test, scored like the Economist's recent science quiz [32], is included inline below to encourage maximum participation by email. I'll accept emailed answers until midnight EST on 31 December 1998. Early next year I'll publish a "survey says" consensus of your predictions: The Sense of TBTF for 1999. And in December 1999 I'll publish the names and (if they wish) email addresses of the top ten scorers as rated by history.

Please copy the questions below into an email message and send them, with your answers, to dawson dot tbtf at gmail dot com. For my sanity please begin your email's subject line with [QUIZ].

Thanks to Bob Treitman for suggesting such a contest and to Jon Waldron for significant help with its form and content.

[32] http://tbtf.com/archive/1998-11-11.html#s11

The questions

  1. Predict the closing stock price in $US of Amazon.com on
     1 December 1999.

  2. Predict the profit (loss) for Apple Computer in $US in
     the July - September quarter of 1999.

  3. How many Linux systems will be in use worldwide on 1
     December 1999? (Hint: the consensus range for early
     1998 was 5-7 million.)

  4. Name five independent software companies that will be
     bought out by competitors other than Microsoft between
     1 January and 1 December 1999.

  5. Name five independent software companies that will be
     bought out by Microsoft between 1 January and 1 December
     1999.

  6. Name the top five Web sites, as ranked by number of hits
     per day, for the month of November 1999.

  7. Predict the percentage of US households having Internet
     access on 1 December 1999.

  8. Of the US households having Net access, what percentage
     of them will cover the last mile via:

       - analog modem?
       - cable?
       - xDSL?
       - satellite?
       - wireless?

  9. What is the strongest cryptography (key length in bits)
     that will be exportable without a license from the U.S.
     on 1 December 1999? From Britain? Israel? Australia?
     Finland?

 Essay question:

 10. What will be the status of the Microsoft antitrust trial
     on 1 December 1999? Who will win? If the DoJ wins, what
     remedy will be ordered? Will the case be appealed? Will
     it have reached the Supreme Court? Will they have ruled?
     How, and by what split? Which Justices will vote each
     way?

 Extra credit / tie breaker:

 11. Predict the title of Bill Gates's new book.

Scoring

For questions that turn on market statistics, I will develop a consensus best estimate or range from available sources (and cite the sources). Opinions of the judge — that's me — will be final.

10 points if within 2.5%; 5% => 8 points; 10% => 6; 15% => 4; 20% => 2.
10 points if within $0.5M; $1M => 5 points; $2M => 2.
10 points if within 5% of the best estimate, 5 points if within the consensus range.
2 points each.
2 points each.
2 points each.
10 points if within 2%, 5 points if within 4%.
2 points for each one within 5%.
2 points each.
Up to 10 points.
Up to 10 points based on humor, self-reference, and/or the obscurity of literary allusions. Examples (please blame these on Jon Waldron, and don't submit them):
    The Road Farther Ahead
    The Prince
    Where Lawyers Stay Up Late

If you would like to review a book for TBTF, please send me a proposal or a review (200-400 words). Reviews should not have been published elsewhere; you retain all rights; remuneration is limited to net.fame.

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

TBTF home and archive at http://tbtf.com/ . To subscribe send the
the message "subscribe" to tbtf-request@world.std.com. TBTF is Copy-
right 1994-1998 by Keith Dawson, <dawson dot tbtf at gmail dot com>. Commercial
use prohibited. For non-commercial purposes please forward, post,
and link as you see fit.
_______________________________________________
Keith Dawson    dawson dot tbtf at gmail dot com
Layer of ash separates morning and evening milk.


TBTF HOME	CURRENT ISSUE	TBTF LOG	TABLE OF CONTENTS	TBTF THREADS	SEARCH TBTF

TBTF for 1998-12-15: Meme pool

Contents

India moving toward Net interception

Then there were nineteen

British crypto debate spawns Net activism

The Zurko patent

How much trouble can you get into by linking?

Jargon Scout: STFW

Quick bits

Quick surf

TBTF Book Review: The Man Who Loved Only Numbers

TBTF readers' prognostications for 1999

Notes

Sources