> roving_reporter t byfield [email] Thu Mar 1 00:02:47 EST 2001 ICANN watchers The Old Faithful of ICANN-related sites, ICANNwatch, has unveiled its facelift: a Slashdot-like news and comment forum. Good move. Between this and the New Faithful of ICANN-related sites, Bret Fausett's stellar ICANN Blog, the roving_reporter is could just about hang up his hat. Heh--well, maybe not... The new ICANNwatch and the ICANN Blog both serve functions that ICANN itself could and should be doing: providing a useful forum for input and dialog, and providing timely, informative, and neutral news on its activities. Unfortunately, ICANN's failure in this regard has been singularly and consistently spectacular: witness, only most recently, the prompt posting the testimony of Maximum Leader Mike Roberts and CoB Vint Cerf before the House and Senate committees, as opposed to the curious absence of Boardmember Karl Auerbach's Senate testimony. (For a larf, see this rather candid "offtopic" assessment of that inconsistency.) Kudos to Bret and the maintainers of ICANN watch -- in particular, Michael Froomkin -- for their efforts. Thu Feb 22 15:11:14 EST 2001 annus-horribilis.qeii.ca CIRA, Canada's new admin for the .ca ccTLD, has published its requirements regarding who is eligible to register a domain: Canadian citizens, permanent residents, legal representatives, corporations, trusts, partnerships, associations, trade unions, political parties, various kinds of educational institutions, libraries, indian "bands," aboriginals, the government, registrants of trademark and official marks, and Her Majesty Queen Elizabeth II and her successors. Fri Feb 9 15:24:40 EST 2001 Not very NAIS at all An impressive coalition billing itself as the NGO and Academic ICANN Study (NAIS) has written a remarkable open letter to Carl Bildt, who is slated to chair ICANN's At-Large Membership Study Committee (ALMSC). In it, NAIS has made an initial list of data and documents it wants access to: server logs from the systems used in ICANN's and election.com's handling of the election, registration records "of at least some information collected from users," mailing records, voting records, technical specifications, financial records, and "internal and external communication of ICANN officers and staff regarding the election" -- pretty much the whole shootin' match. The letter goes out of its way to stress that NAIS is specifically not requesting personally identifiable data; however, it does say that ICANN should "make all of this election data available not only to the undersigned researchers, but to other research efforts as well." (For ravings on some of these subjects, see roving_reporter journals 1, 2, and 3 passim.) ICANN has also invited former boardmember Pindar Wong and Charles Costello, director of the Carter Center's Democracy Program, to join the ALMSC as vice-chairs. The roving_reporter takes a dim view of these invitations, particularly the latter. The Carter Center's statement on the MAL elections got off to a shaky start by plopping out a real howler: "ICANN 'keeps the Internet running' and assigns high-level names and" -- perish the thought -- "through registrars, e-mail addresses." More substantively, despite the CC's efforts to hedge their language ("reasonably," "subject to further review of the available balloting data," "based on a limited monitoring of registration," etc., etc.), the letter far overshot its ill-defined scope on key points. For example, it duly repeated ICANN's claims about spikes in request rates being the cause of registration server problems -- but failed to note that the server had been throttled down to process no more requests than ICANN's staff could support. (The CC agreed to answer a number of questions from the roving_reporter but, on receiving the questions, fell silent.) Without casting aspersions on Mr. Costello's integrity, we nevertheless doubt that this record suggests his contributions to the ALMSC will be optimal. ICANN has also appointed a former NSF, DoC, and electronics trade-group staffer, Denise Michel, as Executive Director of the ALMSC. All very bottom-up, isn't it? Only fools make predictions. But since the roving_reporter is -- nay, takes pride in being -- profoundly foolish, we predict that ICANN will be rather less than forthcoming with the requested materials. Specifically: ICANN will not release server logs for the machine on which its broken election "front end" ran because they probably didn't archive them; ICANN will not release webserver logs on the pretext that requesting IP numbers (or possibly looked-up DNS entries) can in some cases obviously be correlated with specific individuals; ICANN will not release MAL-associated financial records, because doing so could reveal ways in which it uses project-specific revenues to cover its general operating costs; ICANN will not release mailing records, because those records consist, in large part, of stacks of returned envelopes -- which of course would reveal personally identifiable information; and ICANN will not release internal correspondence because it would be too damn embarassing. We would, of course, be very pleased to be wrong about this. Mon Feb 5 16:25:13 EST 2001 ICANN forgets to announce Congressional hearings [updated] Bright and early Thursday morning, Billy Tauzin's House Committee on Energy and Commerce will hold hearings, titled Is ICANN's New Generation of Internet Domain Name Selection Process Thwarting Competition?, which "will focus on ICANN's recently announced selections of registry operators for new top level domains." ICANN hasn't seen fit to note this among their announcements. [Crikey, yet another addendum (these things are killing me...): A roving_reader points out that ICANN's Correspondence page even includes a section devoted to the U.S. House Committee on Commerce. As noted in this space, though, that page is the informational equivalent of a Potemkin village. As ever, thanks for the pointer.] Mon Feb 5 16:22:02 EST 2001 Wham, bam, thank you MAL [updated] Several sources present at the ICANN-Studienkreis 2-3 February meeting in Zurich confirm that, in Marc Holitscher's panel discussion on the Membership at Large, Chief Policy Poobah Andrew McLaughlin stated that the MAL "doesn't exist" anymore. Evidently, ICANN staff is flogging the claim that the MAL was just a temporary mechanism for "selecting" regional directors for the board -- and claim, further, that this was the plan from the very beginning, it's all in the documents, &c., &c. If indeed that's been the case all along, then why don't any of the MAL-related resolutions -- from Berlin (27 May 1999), Santiago (26 August 1999), LA (4 November 1999), Cairo (10 March 2000), or Yokohama (16 July 2000) -- mention it? And why did Mclaughlin express concern that the MAL was being "hijacked" at the Berkman Center MAL panel discussion in the Pressing Issues II sessions prior to ICANN's November 2000 meeting in Marina Del Rey? Why is there no mention of it in ICANN's grant proposal to the Markle Foundation? And why hasn't ICANN's MAL site made any mention of the MAL's transience -- not in the schedule, not in the FAQ, not in the rules? And, really, why would the FAQ include this? Do I have to renew my membership every year? Yes. Members will be asked to reconfirm their membership data every year. Reconfirm their membership data. That means "change of address." Not: "sign up again." Every once in a while, the roving_reporter's really gotta call a spade a spade: Andrew McLaughlin, ICANN's self-styled Chief Policy Officer, wouldn't know the truth if it hit him in the ass with a banjo. But it shouldn't be personal: he's just doing his job. For ICANN. And against you. [Addendum: McLaughlin writes (not necessarily in response): To reprise what I said in Zuerich: The At Large Membership does not exist in the form Hans [Klein, of CPSR] wants it to exist. At the moment, it exists in a state of suspension: Under the Bylaws, the membership was constituted for one purpose and one purpose only: to choose 5 Directors in 2000. After that, the At Large Study is supposed to figure out what sort of role and function it should have in the future. But, of course, there haven't been any registrations since July, and no activations since September. So it can't be a surprise to anyone that the membership database lies dormant, pending the outcome of the study. Too clever by half. First and foremost, it should be noted that Hans Klein is far from alone in being dissatisfied with ICANN's handling of the MAL. In fact, the roving_reporter would be curious to know of anyone who is happy with it. Secondly, roving_readers would do well to review, for example, the MAL FAQ -- written, one assumes, by ICANN staff -- to see whether its language supports the picture McLaughlin's presents. If it doesn't, then ICANN has only itself to blame. McLaughlin sees the paucity of explicit statements that the MAL should continue to function post-election as a positive and affirmative justification for icing it. But another explanation -- one that's much more sensible -- is that people assumed it would continue: that MALers would elect directors, that the directors would remain accountable and responsive through continuous input, and so on. That understanding is quite suited to an organization that purports to be "bottom-up"; McLaughlin's pharisaical understanding is not. And for someone who claims to adhere so vigorously to the letter of the by-law, his claim that "the At Large Study is supposed to figure out what sort of role and function it should have in the future" is tendentious. In point of fact, the current by-laws say the study will determine (note the priorities): Whether the ICANN Board should include "At Large" Directors; If so, how many such Directors there should be; How any such "At Large" Directors should be selected, including consideration of at least the following options: selection by an "At Large" membership; appointment by the existing Board; selection or appointment by some other entity or entities; and any combination of those options; If selection by an "At Large" membership is to be used, the processes and procedures by which that selection will take place; and last and also least... What the appropriate structure, role and functions of an "At Large" membership should be. That last question depends on the answers to the previous ones. If the study were to determine that there should be no MAL directors, or that there should be some but they shouldn't be elected, then it's hard to imagine what role or function the MAL would serve, now, isn't it? But perhaps most noteworthy is McLaughlin's pedantic tone: he understands and is satisfied; Klein doesn't and is a malcontent. If that's how ICANN's Chief of Policy regards the Chairman of the Board of CPSR, imagine if you will, roving_worm, how he looks upon you.] Mon Feb 5 01:48:30 EST 2001 bq--0101001101... spells snafu Brace yourself: this one's very meta. The ICANN Blog notes a cryptic resolution passed on 30 January by the Executive Committee of ICANN's board: VeriSign Multilingual Testbed RESOLVED [EC01.6], that until such time as the Board can further consider the matter, the Vice President and General Counsel is authorized temporarily to waive terms of the Registry Agreement between ICANN and Network Solutions, Inc. (also known as VeriSign Global Registry Services) as necessary to permit blocking of initial registration through non-testbed means of second-level domains within the .com, .net, and .org top-level domains intended for use within VeriSign's multilingual testbed. It took the roving_reporter a while to parse this Faustrollian jewel of bureaucratese, but we did it. Roughly translated, it says something like: Until the board figures comes up with a damage-control plan, Louis Touton pretty much has carte blanche to stomp out multilingual-testbed-style domains we don't like. So what's the damage? Well, roving_readers may remember our 20 January item, "Hawaii 2.0," in which we tried to make sense of the multilingual morass. Rising up from that morass, like the creature from the black lagoon, was the specter of RACE -- the Row-based ASCII-Compatible Encoding standard, a kludge for using ASCII to encode non-Latin characters. To organizations whose very existence depends on the current ASCII-only DNS architecture -- NSI (now VGRS), which is desperately squeezing every last cent out of it, and ICANN, whose raison d'être seems to be the surbordination of substantive technical progress to venal intellectual property interests -- RACE is a dream come true. It offers a way to service select Asian markets without muddying their hands by coordinating a coherent restructuring of DNS in a way that would actually meet the needs of a global linguistic environment. And yet RACE is ASCII-"compatible"; that means RACE-"encoded" domains are ASCII. As VGRS's own Multilingual DNS FAQ states, "domain.com for a given language will be stored as bq--gde6djht.com." Since RACE describes an algorithm, the converse is true: bq--gde6djht will be rendered as "domain" in a given language. Which means that someone really clever might register in ASCII domains that, when RACEd, correspond to, say, famous names or trademarks in a given language. And they might do so outside of the VGRS/ICANN "testbed" -- as one would register any other domain (and for a lot less than the $199 VGRS is charging). If so, then those domains aren't provisional: they're the real thing. If indeed what we surmise sparked the Exec Comm's open-ended mandate for VP Louis Touton, such a hack -- beyond being seriously slick -- could pose immense problems for ICANN, its UDRP, and so-called "cybersquatting" in general. Even the literal one-to-one.com correspondence of trademarks to domains has proven to be profoundly controversial. "Substring" issues (e.g., verizonreallysucks.com) have proven even more troublesome. But the idea that trademarks encompass strings encoded with one standard but parsed with another -- for example, that "Disney" also covers 64 case-insensitive binary expressions such as "010001000110100101110011011011100110010101111001" -- might be a tad too high-concept for even the most zealous proponents of "intellectual property" claims. But if a trademark registered in ASCII doesn't apply to binary, then why should it apply to RACE -- or to any other non-obvious encoding scheme? Now, the architects of RACE did anticipate this problem. Paul Hoffman, in his Internet Draft draft-ietf-idn-race-02, notes: All post-converted name parts that contain internationalized characters begin with the string "bq--". (Of course, because host name parts are case-insensitive, this might also be represented as "Bq--" or "bQ--" or "BQ--".) The string "bq--" was chosen because it is extremely unlikely to exist in host parts before this specification was produced. As a historical note, in late August 2000, none of the second-level host name parts in any of the .com, .edu, .net, and .org top-level domains began with "bq--"; there are many tens of thousands of other strings of three characters followed by a hyphen that have this property and could be used instead. The string "bq--" will change to other strings with the same properties in future versions of this draft. [emphasis added] But Hoffman's subsequent remarks implicitly acknowledge that this scheme is basically braindead, when he recommends that zone admins had better not name any hosts in ways that might confuddle RACE-based resolvers: Note that a zone administrator might still choose to use "bq--" at the beginning of a host name part even if that part does not contain internationalized characters. Zone administrators SHOULD NOT create host part names that begin with "bq--" unless those names are post-converted names. Creating host part names that begin with "bq--" but that are not post-converted names may cause two distinct problems. Some display systems, after converting the post-converted name part back to an internationalized name part, might display the name parts in a possibly-confusing fashion to users. More seriously, some resolvers, after converting the post-converted name part back to an internationalized name part, might reject the host name if it contains illegal characters. ICANN, the "technical coordinating body" of the internet, approved this standard as the way to meet the needs of East Asia. And now that the flaw in this scheme is revealing itself, the Executive Committee of ICANN's board has effectively suspended its agreement with NSI and thereby granted its Veep free rein to put in a fix. [Addendum: The roving_reporter's surmise was correct. According to the draft Minutes of [MINC's] Registration Policy WG Meeting of 30 January, "19000 names [were] registered as pure ASCII"; the registry deleted them, but they were "re-registered seconds later." As always, thanks to the diligent people who turn up these gems of documentation.] Fri Feb 2 01:57:01 EST 2001 IANA_dysfunction.mp3 The roving_reporter is very pleased to report that, through dogged persistence and the kindness of strangers, he has obtained a videotape of the lost MDR2K IANA-ccTLD meeting (see our coverage here and here). The Berkman posse was unable to videotape the session due to an unexplained last-minute switcheroo of venue; but an observer caught the drama -- and dramatic it was -- on tape. A hearty thanks to all who helped us obtain a copy. If roving_readers would like to hear ccTLD luminaries such as ISOC NZ's Peter Dengate Thrush, nic.vi's Peter de Blanc, and Nominet MD Willie Black giving the ICANN staffers' sophistry a thorough trashing, the audio is now available in MP3 format here (four files, 23MB total). Share and enjoy. [Addendum: One roving_reader has suggested making the four MP3s available via HTTP as well to facilitate streaming. Done. And another roving_reader with a big heart and bigger pipes has mirrored them at internet.com both as separate MP3 files (1, 2, 3, 4) and as a single RealAudio stream with minimized background hum. Hats off to both of you.] [Addendum 2: Simon Higgs has kindly provided another mirror here Hats off to him as well.] Wed Jan 31 07:20:47 EST 2001 Your gTLD application fees at work Precisely in accordance with prophecy, yesterday ICANN's Executive Committee held a teleconference in which the numero-uno item was "Authorizing repayment of loan" -- that's terse for the outstanding $150,000 loan from Cisco. Unlike MCI/WorldCom, 3Com, and Deutsche Telekom -- all of which have obligingly deferred repayment until summer or fall -- Cisco wanted its money back by 2 February. ICANN continues to present these loans, which were executed between July and October of 1999, as "one year unsecured loans" (though the link they offer as if to substantiate that claim is to a boilerplate loan agreement); as noted in this space, ICANN's public explanations of the terms of these loans differed remarkably from the disclosures it made to the IRS in its application for 501(c)(3) status. Wed Jan 31 07:19:00 EST 2001 Wrights and rongs [updated] The end of the month is as good a time as any for the roving_reporter to set an example for our fine friends at ICANN by owning up. We were way off with our speculations about Don Heath, which one very well-informed roving_reader dismissed as a "conspiracy theory." We chafed at that description; it chafed back; we lost. Then again, other well-informed sources noted that ICANNauts were publicly scoffing at our report about Carl Bildt -- that someone so prominent would never take a job at ICANN. Heh. In every case, we hasten to thank our sources and correspondents: it's you what keep us honest. [Addendum: The roving_reporter seems to have been wrong about being wrong about Heath, which is to say he was right about Heath: rumor has it that ISOC is nominating him for a seat on ICANN's board.] Mon Jan 29 16:08:36 EST 2001 I'm blogging as fast as I can Well, not really...but no way can the roving_reporter keep up with Bret Fausett's ICANN blog, which has quickly become the preeminent source of timely news about the activities of ICANN and its alphabeticalistical infrastructure-cum-suburban sprawl of working groups, supporting orgs, committees, constituencies, and divers other bureaucratic chinoiserie. Fausett is doing what ICANN, what with its soporific mantra about "openness" and "transparency," should have been doing all along: maintaining a fast and accurate log of their doings, peppered with links to primary sources. Had ICANN done so, though, they likely wouldn't have lasted as long as they have -- which may make clear why they've chosen instead to dole out nanospoonfuls of legalistic boilerplate in the form of press releases, spiced up with the very occasional morsel of "correspondence." And an alarming amount of that on Friday afternoons, no less. Note that ICANN is now hiring a "Public Affairs and Communications Director." And a "Policy Analyst," too. Given that each of these jobs pays $40-80K per year, that's an interesting move because ICANN's budget for FY 2000-2001 line-itemed $1,611,000 for "personnel" -- only $11K more than the $1,600,000 line-itemed for "executive and staff compensation" in the FY 1999-2000 budget. With associated costs, these new positions could run ICANN a cool $150-250K/year easy. Now, where did ICANN get that kind of money after they drew up the budget on 6 June? Surely not from the ccTLDs, whose admirable recalcitrance reduced ICANN's revenues by a substantial chunk of $1,355,000. And certainly not from gTLD registrar contributions, which have flatlined at $535K/Q (1999 listing are here, 2000 FQ1 is here). The roving_reporter bets it came, in accordance with prophesy, from the surplus revenues drawn from new gTLD application fee. If that's correct, then ICANN's willingness to commit some of that windfall to an ongoing expense may suggest that new rounds of gTLD applications will be coming in the foreseeable future. And it also suggests that their public budgets, like their public announcements, are but a veneer. If only the excellent Mr. Fausett had access to the financial gruesomes as well. Not that we'd really wish such a fate on him, mind you... Sun Jan 21 11:56:30 EST 2001 Poor guy versus multinational cult Dutch columnist and anti-$cientology activist Karin Spaink has published some remarkable accounts of the appeals trial of Scientology versus [Zenon] Panoussis -- her partner's appeal of a Swedish court's 1998 ruling in the cult's favor. That ruling awarded the Co$ a measly US$2,000 in "damages" incurred when Panoussis publicly posted some Co$ copyrighted "religious" twaddle on the net. Unfortunately, the court also awarded the Co$ a more formidable US$150,000 in legal fees; Panoussis has been all but penniless since, because the Co$ has garnished his wages in -- in our opinion -- its ongoing effort to destroy him. The Co$ had asked the court to force Panoussis to pony up $2 million in legal fees. The spectacular battles on the net between the Co$ and its opponents established the framework for many subsequent abuses of "intellectual property" claims to silence critics of corporate and political activities. Sat Jan 20 13:34:50 EST 2001 Hawaii 2.0 Total Telecom reports that ICANN Maximum Leader Mike Roberts, in a talk at the twenty-third Pacific Telecommunications Council conference, broached two of ICANN's more delicate subjects: "multilingual" DNS -- that is, domain names encoded in ways more suited to orthographies around the world than ASCII's C-ration approach (i.e., A-Z, 0-9, and "-") -- and ICANN's relationship to the ccTLDs. Those two subjects are becoming intertwined like lovers in Oshima's Realm of the Senses, but not for any particularly good reason. It's mostly attributable to ICANN's clix Americana worldview, wherein all things un-American get lumped together under "miscellaneous" ("the signpost of ignorance" as Marcel Mauss put it) -- as expressed, for example, in the organizational dichotomy within ICANN between the gTLD Registry "Constituency" (one member: VeriSign Global Registry Services, née NSI) and the ccTLD Registries Constituency (approximately 244 members). The roving_reporter has dealt with ICANN vis-à-vis the ccTLDs in detail, but he has strenuously avoided plunging into the kaleidoscopic world of multilingual DNS. Oh well -- here goes... In January 2000, Nunames, the commercially oriented ccTLD registrar for Niue's ".nu", announced a multilingual test, but -- a minor detail -- didn't specify which character encoding standard their extended system would use. There are many, e.g., DOS Codepage 850, Macintosh, ISO-Latin-1, Unicode, Unicode/UTF-8. (At the time, we noted in our sadly neglected Bitbucket column on Telepolis that we were "eager to see a California judge of Latin descent rule on whether Ámázóñ.nu is a trademark infringement"). Although Nunames' experiment didn't go far, a few similar, scattered efforts followed. Since those halcyon days, though, the situation has become more complex -- specifically, it has become more overtly political. In November, with a nod from ICANN, Verisign Global Registry Services (VGRS) -- the M&Aed registry half of ex-NSI, as distinct from the registrar half of the split (see ICANN's summary) -- announced a "testbed" involving Chinese, Japanese, and Korean characters. This was significant, because it omitted other Asian languages (Thai and Khmer, for example) whose orthography is covered in Unicode/UTF-8. Instead of doing something sensible and truly international -- like working fairly and squarely within the IETF's developing Internationalized Domain Name (IDN) framework, or collaborating with the remarkably international Multilingual Internet Names Consortium (MINC) -- VGRS jumped the gun by adopting the backwards-compatible RACE (Row-based ASCII-Compatible Encoding) kludge, which takes Ginsu knives to the bitstream: slicing, dicing, and otherwise mungeing 7-bit ASCII into UTF-compatible encodings. Why did VGRS do this? Well, money^W I mean marketing strategies -- no doubt based (generally) on the West's centuries-old delusional dream of "opening" the fabled markets of Asia to their warez^Hs, and also based (specifically) on the perceived need to respond "at internet speed" to certain trends revealed in their statistical analyses of domain registrations (ZDNet Asia's coverage of VGRS's surprising stats can be found here). But more important than the origins of VGRS's actions are the consequences; and, as roving_readers surely know by now, in the N-dimensional looking-glass world of the politics of DNS, tend to be refractory indeed. The most obvious immediate result (brought to our attention by Brock Meeks) was a sudden surge of new domains that when rendered in ASCII read as bq--3bp25d3prp5yabi.com, bq--3b2tkw2qkvdf5fy.org, bq--3bixctvloc4q.com, bq--gdpkjl7nxxk4rkx42wr3s.net, and so on. The "bq--" prefix marks these domains as "multilingual." Whether registered trademark-holders like BBQ-World of Reno, Nevada, which could pull a QVC, will appreciate the complexity of the situation is another story; and British Quizzes could have something to say about it, too (but only according to the ludicrous logic that fetishizes "consumer confusion"). The more pressing result was that the VGRS's multilingual initiative offended China's Middle Kingdom sensibilities In a Big Way and elicited a ballistic response. Hu Qiheng, the director of the China Internet Network Information Center (CNNIC), said: [T]he U.S. government has no right to authorize any company to manage Chinese domain names with Chinese characters. A company shouldn't be allowed to provide Chinese domain name registration services in China without the approval of the Chinese government. While the roving_reporter isn't especially sympathetic when it comes to China's internet-related policies (though the issues are more complex than most net-heads like to admit), he can only imagine the orgy of xenophobic raving that would greet any unilaterally declared Chinese plan to sell and regulate American things according to Chinese cultural attitudes (like software, say). A certain circumspection in these matters is warranted, because in addition to being national these issues are also cultural. Thus, CNNIC's director also noted -- correctly -- that [r]elated Chinese departments have protested to the Internet Corporation for Assigned Names and Numbers (ICANN) that Chinese-character domain names are quite different from the ASCII (English) ones, since they have unique...cultural and historic implications Anyone old enough to grok the roving_reporter's labors of love will surely remember that ideas like "branding," which of late have been imbued with pretensions approaching theological dimensions, are really only a few years old -- and, as such, necessarily are culturally and historically specific. The very existence of ICANN's UDRP is itself a clear testament to the fact that the ASCII-compatible world can barely manage to sort these questions out (hell, cities like Dallas and even towns like Salinas can't); but, this reign of confusion notwithstanding, the idea that Asian-language (i.e., Asian-cultural and -historical) domains should be subject to the keystone kourt of ICANN's UDRP globaloney as interpreted by, say, the likes of NAF arbitrar^Htor James Carmody (track record here)... Uh-huh. China's reaction to VGRS's initiative (which seems to have mellowed slightly) suggested that, beyond the multilateral model offered by the organizing ccTLD movement, ICANN's ham-handed endorsement of VGRS's multilingual initiative might end up precipitating a new "sphere of influence" approach to the politics of the namespace. ICANN's dark mantra about what will happen if it fails is puffery: what better way to guarantee that gummints get deeply involved in DNS than by poking them in the eye? That ICANN would expostulate thusly even as it set about trying to undermine the ccTLDs en masse (see our coverage here and here) with babble about "trilateral relationships among governments, their ccTLD organizations, and ICANN"...it doesn't look good. But, in keeping with our rather cheerier Ginsu-kinfi motif -- there's more! -- VGRS's RACE-based kludge also seems to have ruffled some feathers at the IETF. On 13 December, AT&T Labs' John Klensin published an IETF Internet Draft (draft-klensin-i18n-newclass-00) boldly -- and, in the roving_reporter's opinion, brilliantly -- suggesting that the best solution to multilingual DNS might be to consign ASCII-constrained cruft to the junkheap of history by declaring a new ("IN") class of data in DNS records which is UTF-8-multilingualized from the get-go. Klensin's draft is quite radical, and a must-read; but, for roving_readers too lazy to klick on a link, the upshot of his draft is that the installed base of pre-"new-class" software and hardware (including, potentially, ICANN's fetishized root) could be reduced to a legacy system. So now, perhaps, roving_readers will see why the roving_reporter has been so chary of treating these subjects. But back to Generalissimo Roberts, whose remarks in Hawaii seem to confirm the roving_reporter's assessment of the Chinese debacle: "[ICANN alone] deciding which version of Chinese will be used to register domains is an easy way to get fired." (By whom? one wonders.) Roberts also went on to suggest -- unsurprisingly -- that ICANN hopes to impose the UDRP on the ccTLDs, when, according to Total Telecom, he "hinted" about the "possibility of getting our system of registration into other countries." If Total Telecom's reportage is accurate, he seems at times to have taken a rather defensive tone: Also, every citizen on the Net feels they have been empowered to challenge what [ICANN does], and I think we should be able to get on with our jobs without organized undermining of what we're doing. It's good to finally hear a candid assessment from ICANN about just how much "bottom-up" support it really has. As the ICANN Blog points out, ICANN staff -- possibly including Roberts -- will be back in Hawaii in less than two weeks for a luau with the ccTLDs. A hui hou, d00d. The above material is Copyright © 2000 by t. byfield. The r_r began as a semi-collaborative nym on the <nettime> list, where it worked well; but the pseudonym precluded comments, and there was more to report than was good for the list, so now it -- or a mutation of it -- has resurfaced on TBTF. [ top]