From: owner-wg-c-digest@dnso.org (WG-C-DIGEST) To: wg-c-digest@dnso.org Subject: WG-C-DIGEST V1 #76 Reply-To: Sender: owner-wg-c-digest@dnso.org Errors-To: owner-wg-c-digest@dnso.org Precedence: bulk WG-C-DIGEST Monday, April 10 2000 Volume 01 : Number 076 ---------------------------------------------------------------------- Date: Sun, 9 Apr 2000 17:16:10 -0400 From: "Milton Mueller" Subject: Re: [wg-c] S/K principles [Was: Working Group C agenda] In response to Becar: - ----- Original Message ----- > > The problem will be the string, for example .union has a complete different > meaning in France (so in french) than in English !!!!!!! and may be it's an > insult in some other language...so Do we have the goal to invent a global > language???????? This is not really a problem. The principle refers to the "marketing" of the string. If it is marketed to and for labor organizations it would be differentiated. "Dot Com" really doesn't mean anything to anyone by itself. It has acquired "secondary meaning" through its association with web sites. Same could happen with new TLDs. So in that respect, yes, we may be inventing a global semantic. BTW, this means that for the differentiation principle to be effective we *must* give the owners of chartered TLDs the right to determine who can and cannot register within their space, to define the criteria, and to enforce the criteria (bearing in mind of course that they may not choose to enforce any). In response to Dawson: > I do not think that the > registry should be the ones who are tasked with the job of enforcement. NSI > has had always been "blessed" with the fact that it is untouchable. They > are not in the enforcement business - and they never got dragged down that > rat hole. In fact, they do enforce the .edu criteria. More significantly, some registries may WANT to enforce criteria for registration (Jamie Love clearly does). NSI did not want to enforce any because it had no particular incentive to do so for com net and org. Some registries will be commercial, some won't. > The UDRP and the courts should be the ones deciding on the > enforcement based on the guidelines established in the charter. UDRP only affects cybersqatting. So, yes, it is an effective answer to trademark holders concerned about the impact of new TLDs on their marks, but it cannot and was never intended to be used to make registrations conform to particular criteria. Suppose, for example, that a .enum registry for domain to telephone number mapping registered a SLD that was not related to its function. That could not be challenged under the UDRP. > If the > owner of the SLD is in violation of the charter - then the registry would > be ordered by the court to revoke the name of the offender. But, the > registry should not make these decisions on their own. If the UDRP does not > work --- than why even have it ??? > > > 4. These guidelines should not be read to impose overly bureaucratic > > procedures on registries. > > I agree (this should not be a numbered item - this should be at the head of > this list !!) > > > 5. The selection of a gTLD string should not confuse net users, and so > > gTLDs should be clearly differentiated by the string and/or by the > > marketing and functionality associated with the string. > > I agree. > > > 6. A gTLD should not unnecessarily increase opportunities for malicious or > > criminal elements who wish to defraud net users. > ------------------------------ Date: Sun, 9 Apr 2000 21:02:20 -0400 From: "Milton Mueller" Subject: Re: [wg-c] Two Para D&I Summary (PPE) - ----- Original Message ----- > The model for deployment advocated in PPE is very simple: give a TLD to me and my friends. > The model for implementation advocated in PPE is the public resource DNS model, > with shared registries operated on or near their cost-recovery basis, with some > affirmative policy or jurisdictional scope applied by the registry operator. > The policy/scope may be "weak/none" (e.g., identical to NSI registries) or > "strong/specific", (e.g., identical to the North American Aboriginal proposal). Nonsense. Time to call this year-long bluff. If you actually read the proposal, the so-called "public resource" is owned by the registry operator, because the operator determines who can and cannot register names in the .naa name space. I have no problem with this -- it's just a proprietary registry model, put forward by a non-commercial entity or group of entities. What does strike me as incoherent is a description of this as a "shared" registry. Tell us, please, how an SRS standardized interface that allows any accredited registrar in the world to write names into your zone file 24 hours a day without your permission or knowledge is compatible with "strong/specific" policy choices. Did I miss the clause in the ICANN accreditation contract that requires each of them to know which tribal names belong to which groups? Or may be the registrations accepted from several hundred registrars will be held in suspension for a few months while the Indian Councils decide who gets what. In which case, one might well ask, why bother with these registrars at all? Why not just file the request with the Council to begin with? As for "cost recovery," that model made sense, even if it was not universally desirable, when gTLD-MoU proposed that the registry be owned by registrars themselves. Because as owners the registrars would have the knowledge, the incentive, and the authority to determine what true costs were and reduce wholesale rates to cost-recovery levels. If the registry is not a consortium of registrars -- and your .naa proposal clearly is not -- the only way to get "cost recovery" is through government rate regulation, as in the NSI contract. So perhaps you can clarify this for us. ------------------------------ Date: Sun, 09 Apr 2000 23:56:37 -0400 From: Eric Brunner Subject: Re: [wg-c] Two Para D&I Summary (PPE) [...] > If you actually read the proposal, the so-called "public resource" is owned > by the registry operator, because the operator determines who can and cannot > register names in the .naa name space. The proposal has been read quite a few times, and rather closely, but by serious federal indian law types, not to mention several hundred fairly careful indians. Does application of policy concerning the disposition of an object both transform the object into property and establish title to the object? We don't think so. However, feel free to argue the converse. Heck, ICANN can make cyber-precident up as it goes along, so your theory is possible, though I'd avoid it on several principles. > I have no problem with this -- it's just a proprietary registry model, put > forward by a non-commercial entity or group of entities. What does strike me > as incoherent is a description of this as a "shared" registry. I don't think anyone has a problem with any one person, or even a group of persons, arguing for a creative reading of a position paper they oppose. We don't actually have a process for abuse less flamboyant than Jim Flemming's or Jeff Williams', so problematic advocacy is a fairly low risk proposition which you are free to pursue. If, to make your best arguement, you need to argue that delegation of some capacity to engage in affirmative policy ("meaning" in S/K) to the operator of the registry constitutes a "proprietary registry" (new term), feel free to do so. You are of course free to refrain from making the same observation when advocating for S/K, I doubt anyone will actually notice if only because it is clear that S/K doesn't actually "mean" anything but a waste of time. The incoherence you've put your finger on has been there since Position Paper E was first published -- policy binds on all registrants, registrars, and operators of a policed registry. If the registrar (your example) is in fact accredited for a specific registry policy, then it has write access to a registry zone file. It is conceivable that expertise sufficient to function as a competent registrar for .NAA exists in Outer Mongolia, or even Syracuse. It is unlikely that every registrar ICANN accredits will be equally competent in every policed domain, but I suppose the case you really are trying to make is the one you've been pushing for a year -- no policy is the best policy. If PPE were for a policy specific to the sexual preferences of penguins, you'd make just as compelling a case against that by arguing that Indians don't know much about penguins, and don't have many opportunities to make scholarly, if personal, inquiries into the sexual preferences of Antarctic fowl. > Tell us, please, how an SRS standardized interface that allows any > accredited registrar in the world to write names into your zone file 24 > hours a day without your permission or knowledge is compatible with > "strong/specific" policy choices. Did I miss the clause in the ICANN > accreditation contract that requires each of them to know which tribal names > belong to which groups? It is likely that being responsible for portions of the namespace which have non-trivial associated policy will require that applicants to be registry operators, registrars, or critics do their own homework. I don't care that you don't like indians, the bigger problem is you don't like limits, the topicality of which is indians today, anything else tomorrow. > Or may be the registrations accepted from several hundred registrars will be > held in suspension for a few months while the Indian Councils decide who > gets what. In which case, one might well ask, why bother with these > registrars at all? Why not just file the request with the Council to begin > with? It would be nice if you'd actually read a) the Position Paper, or b) if critical thinking on your own is too difficult, asking people who already have done the work. There will be edge cases, but that goes with attempting to delimit anything. > As for "cost recovery," that model made sense, even if it was not > universally desirable, when gTLD-MoU proposed that the registry be owned by > registrars themselves. Because as owners the registrars would have the > knowledge, the incentive, and the authority to determine what true costs > were and reduce wholesale rates to cost-recovery levels. > > If the registry is not a consortium of registrars -- and your .naa proposal > clearly is not -- the only way to get "cost recovery" is through government > rate regulation, as in the NSI contract. So perhaps you can clarify this for > us. This is a little far afield, maybe you should try and take a clear swing at why you don't like policy, and take a deep breath and in a seperate text, do in cost recovry or at least write a clear critical question. Try and distinguish between delegation -- a conduct and temporally limited right to operate (of a registry), and ownership (of a registry). I suggest the first term is a better choice, but if you insist on the latter that is your right. Remember, it isn't the cuteness that counts, its the content. This isn't the S/K mumble-game, this is title/access/policy/scope/cost. Take your best swing, and do try not to write for the audience, it only detracts from your limited arguements -- if anyone cares to read Meuller v Brunner they can sort out the issues better without horsing around with indirect phrasing. Cheers, Eric P.S. Don't forget to get your own D&I in for PPB. I want you to loose on the merits, not because you forgot to respond to a request by the NC via our own co-chair. You don't have to expose it to public criticism, Jon will accept a response made in a sotto voice. ------------------------------ Date: Mon, 10 Apr 2000 00:39:02 -0400 From: Jonathan Weinberg Subject: RE: [wg-c] S/K principles [Was: Working Group C agenda] There weren't too many comments over the weekend on the principles. In response to Jean-Michel, I'm reluctant to fiddle with the language of what's now number 4 because the language seemed to satisfy a broad range of people when it was first proposed, and I don't want to reopen old controversies. Are other people dissatisfied with this language? I've been talking offline with Tan Tin Wee and James Seng (who's co-chair of the IETF internationalized domain name wg). I gather that the IETF WG is on track to issue its own informational RFC by July; that WG will not be addressing governance issues. I've had the thought that we might add a sixth principle that reads something like this: 6. As a general matter, new TLDs should be designed to serve non-English-speaking as well as English-speaking communities. As application for a new TLD, thus, may contemplate that the proposed TLD string will have its primary semantic meaning in a language other than English. The DNSO should promptly charter a working group to develop policy regarding internationalized domain names using non-ASCII characters. Is this agreeable to the working group? If we're to have a week for a formal consensus call, and still get the result to the Names Council before the NC's own vote, then I need to issue the consensus call *tomorrow*. Do folks want any further changes before then? Speak now . . . Jon - ------------------- 1. The initial rollout of six to ten new gTLDs, followed by an evaluation period, should include both open, unrestricted TLDs and chartered TLDs with more limited scope. 2. An application for a chartered TLD should explain what meaning will be imputed to the proposed TLD string, and how the applicant contemplates that the new TLD will be perceived by the relevant population of net users. 3. An application for a chartered TLD should explain the mechanism(s) for charter enforcement. The simplest possible enforcement mechanisms is registrant self-selection: that is, if .AUTOMOBILES is set up for entities in some way associated with the automobile world, the applicant might simply rely on its understanding that other entities will have no interest in registering in that TLD. Alternatively, an applicant might choose a more elaborate, registry-driven enforcement mechanism. 4. The selection of a TLD string should not confuse net users, and so TLDs should be clearly differentiated by the string and/or by the marketing and functionality associated with the string. 5. New TLDs are important to meet the needs of an expanding Internet community. They should serve both commercial and noncommercial goals. The authorization process for new TLDs should not be used as a means of protecting existing service providers from competition. 6. As a general matter, new TLDs should be designed to serve non-English-speaking as well as English-speaking communities. As application for a new TLD, thus, may contemplate that the proposed TLD string will have its primary semantic meaning in a language other than English. The DNSO should promptly charter a working group to develop policy regarding internationalized domain names using non-ASCII characters. ------------------------------ Date: Mon, 10 Apr 2000 00:52:51 -0400 From: Jonathan Weinberg Subject: Re: [wg-c] Two Para D&I Summary (PPE) Without meaning for a moment to wade into the substance of the Brunner-Mueller debate -- Eric's correct that I passed along, about a week-and-a-half ago, that >One of the NC members has asked for a >short, user- friendly summary of the various models for deploying and >implementing new gTLDs that folks described in their position papers. I'd >be grateful if each of the position-paper authors could send me a short >summary (no more than a paragraph or two) of the scenario for deploying new >gTLDs contemplated in that author's position paper, so that I can collate >them and transmit the resulting short paper to the NC. > > Thanks. > My thanks to Eric for his submission, and I urge the other authors to submit theirs; anyone who doesn't send one in runs the risk of having me (mis)characterize his views. Assuming that an author does submit a summary, I'm going to let that author be the final judge of whether his summary is internally coherent and accurately reflects the views expressed in his paper. Jon ------------------------------ Date: Sun, 09 Apr 2000 23:48:38 -0700 From: Simon Higgs Subject: Re: [wg-c] Two Para D&I Summary (PPE) At 03:01 PM 4/9/00 -0400, Eric Brunner wrote: >Title: Two Para D&I Summary (PPE) > >The model for implementation advocated in PPE is the public resource DNS >model, >with shared registries operated on or near their cost-recovery basis, with >some >affirmative policy or jurisdictional scope applied by the registry operator. >The policy/scope may be "weak/none" (e.g., identical to NSI registries) or >"strong/specific", (e.g., identical to the North American Aboriginal >proposal). Could you please provide a definition of "shared registries" (as opposed to "registrars")? How does cost-recovery not become price-fixing (illegal in the most countries including the US)? Best Regards, Simon - -- The bandwidth is out there... ------------------------------ Date: Mon, 10 Apr 2000 00:06:22 -0700 From: Dave Crocker Subject: Re: [wg-c] Two Para D&I Summary (PPE) At 11:48 PM 4/9/00 -0700, Simon Higgs wrote: >How does cost-recovery not become price-fixing (illegal in the most >countries including the US)? (here we go again.) constraints on pricing are often entirely legal. there is no such thing as a completely free market system, in the real world. not even hong kong. d/ =-=-=-=-= Dave Crocker Brandenburg Consulting Tel: +1.408.246.8253, Fax: +1.408.273.6464 675 Spruce Drive, Sunnyvale, CA 94086 USA ------------------------------ Date: Mon, 10 Apr 2000 00:24:01 -0700 From: Kent Crispin Subject: [wg-c] deployment and implementation summary position paper D Position paper D divides deployment and implementation of gTLDs in two phases -- the short term startup phase, and the long term. Short Term: In the short term PPD advocates selection of 5-9 names for new open tlds. In PPD there is no proprietary connection between names and registries whatsoever, and indeed, names can move from registry to registry, so, therefore, the selection of names is not terribly critical, and could be done by ICANN's legal staff, using their best judgement concerning the legal impacts of particular name choices. Simultaneously, ICANN should produce an RFP for candidates for 3-6 registry operators. These proposals should describe in convincing detail how the registry operator would implement efficient transfer of registry data for a particular TLD from one registry to another. ICANN will work with the proposals to come up with a set of 3-6 operators for registries that can meet the requirements. Contracts would be let; the operators would implement their technology, and they would each be arbitrarily delegated one or more of the approved TLDs, though the delegation could be moved for load balancing or other purposes. All these registries would be shared, and all ICANN-approved registrars would be authorized to register in the new TLDs. Long Term: The long term view primarily is concerned with a selection process for new names -- the registry selection process would stay much the same. In the long term, PPD advocates that parties interested in the delegation of a new TLD would petition the Names Council for formation of a Working Group to develop a proposal for that TLD. This is a procedural step -- normally the NC would approve any reasonable request for formation of a WG. The WG would produce the concrete documents specifying the rationale for the TLD; its charter (if any); its sponsoring authority (if any); special contracts with registrars that might be required; dispute resolution mechanisms; and so forth. (The sponsoring authority could be the party proposing the TLD, of course). The WG would provide the mechanism for public input into the creation of the TLD; the documents would be given to the NC; posted for public comment; and either returned to the WG or passed on to ICANN for ICANN's consideration. ICANN would approve or disapprove; and, if approved, the TLD would be delegated to one of the current registries. As more TLDs are approved, more contracts for registry operators could be let. Note that the WG process is intended to be general: in exceptional circumstances a particular registry could be delegated a particular TLD. But a very good argument would have to be made that the single registry could meet the stability requirements specified in PPD. - -- Kent Crispin "Do good, and you'll be kent@songbird.com lonesome." -- Mark Twain ------------------------------ Date: Mon, 10 Apr 2000 01:03:40 -0700 From: "Christopher Ambler" Subject: RE: [wg-c] Two Para D&I Summary (PPE) That's nice, Dave, but you didn't answer the question. - -- Christopher Ambler chris@the.web - -----Original Message----- From: owner-wg-c@dnso.org [mailto:owner-wg-c@dnso.org]On Behalf Of Dave Crocker Sent: Monday, April 10, 2000 12:06 AM To: Simon Higgs Cc: wg-c@dnso.org Subject: Re: [wg-c] Two Para D&I Summary (PPE) At 11:48 PM 4/9/00 -0700, Simon Higgs wrote: >How does cost-recovery not become price-fixing (illegal in the most >countries including the US)? (here we go again.) constraints on pricing are often entirely legal. there is no such thing as a completely free market system, in the real world. not even hong kong. d/ =-=-=-=-= Dave Crocker Brandenburg Consulting Tel: +1.408.246.8253, Fax: +1.408.273.6464 675 Spruce Drive, Sunnyvale, CA 94086 USA ------------------------------ Date: Mon, 10 Apr 2000 01:27:03 -0700 (PDT) From: "William X. Walsh" Subject: Re: [wg-c] Two Para D&I Summary (PPE) - -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 10-Apr-2000 Dave Crocker wrote: > At 11:48 PM 4/9/00 -0700, Simon Higgs wrote: >>How does cost-recovery not become price-fixing (illegal in the most >>countries including the US)? > > (here we go again.) > > constraints on pricing are often entirely legal. And under what conditions? Those conditions are VERY specific, and you have a real threshold to meet under the law. And this protection alone may be the only protection we have from your ilk to force this type of model onto every registry. - - -- William X. Walsh http://userfriendly.com/ GPG/PGP Key at http://userfriendly.com/wwalsh.gpg - -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.1c (Mandrake Linux) Comment: Userfriendly Networks http://www.userfriendly.com/ iD8DBQE48ZBW8zLmV94Pz+IRAjVcAJ0eywGUGDWyhvKwtemVrx2M4z+ScQCfTSGd dfAtVyNUvXgIc9DgxB6yasI= =uAhz - -----END PGP SIGNATURE----- ------------------------------ Date: Mon, 10 Apr 2000 07:49:47 -0400 From: Eric Brunner Subject: Re: [wg-c] Two Para D&I Summary (PPE) Simon, > Could you please provide a definition of "shared registries" (as opposed to > "registrars")? At least two independant functionally distinct mechanisms exist to support multiple reader, mediated writer access, the NSI-SRS mechanism, discussed on another list Rick and I are on, and the SRSP mechanism, which is a more general mechanism. Neither of these however attempts the problem(s) of registry operator failure in the presence of two or more temporally colocated or disjoint-sequential registry operators. Both allow for a single point of failure. You'll want to read Position Paper D (its not long, and if WG-C had a single test for admission, it would be having read at least PPD twice), then ask any question you think prudent or useful. We assume that when .NAA becomes operational, that within some reasonable period of time some other registry, e.g., .MUS, will also be operational, and that the disjoint operators will adopt a risk-adverse internal data model and practice, e.g., common format (or convertable format), routine data transport, routine replication and routine escrow. > How does cost-recovery not become price-fixing (illegal in the most > countries including the US)? If this is a different issue than the one Chris Ambler then it deserves a better presentation, if it isn't, then it is already asked and answered. Before getting hijacked off into one of a long series of ratholes, S/K only the latest, the discussion of pricing was vigorous and interesting. Please see the archives, you'll want to look for discussion in particular of why a price floor set at NSI's present marginal cost might preserve NSI's monopoly. You may want to introduce yourself to the rest of WG-C as well, not everyone will recall the exchange between Tony Rutkowski and myself on the subject of the IAB Technical Comment on the Unique DNS Root, or realize that you are the CTO & Bottle Washer of the SuperRoot Consortium, a happy collection of folks who've simply no business pissing under the ICANN tent having put up canvas of their own somewhere else. I'm not sure which is worse, having Jim Flemming go on about his cure for everything, Jeff Williams go on about having more supporters than the Dallas Cowboys, or the ICANN-optional crowd treating ICANN working groups lacking a test for good-faith participation as ritual urinals. Cheers, Eric ------------------------------ Date: Mon, 10 Apr 2000 09:09:11 -0400 From: "Kevin J. Connolly" Subject: RE: [wg-c] Two Para D&I Summary (PPE) Price-fixing is illegal when it is an aspect of a combination in restraint of trade. Antitrust laws prohibit monopolization, not competitive schemae. The test is whether the price has been fixed at a level greater than would obtain under free competition. Since free competition results in the market price becoming equal to the marginal cost of producing the object of commerce, an agreement which fixes the price of an object at the zero-profit level would not, under most antitrust analyses, be open to objection. Of course, there could be other ways in which the schema could come under attack. For example, if another good or service were tied to the zero-profit item (we make $0 profit on domain name registrations so long as you use Evil ISP to host your virtual domain at a price of $100/month + traffic and storage surcharges, and to make sure you don't make an end run around us, our people are the technical and zone contacts and we secure the zone file with Guardian) then I daresay the business model is illegal. Moreover, an agreement to operate registry services on a strict cost-recovery basis does not fix the price of registrations to the consumer. In fact, it doesn't even fix the price to the registrar, since a registry that operates more efficiently than others will, presumably, lower its prices (otherwise how would it still be operating on a cost recovery basis?). Kevin J. Connolly The opinions expressed are those of the author, not of Robinson Silverman Pearce Aronsohn & Berman LLP This note is not legal advice. If it were, it would come with an invoice. As usual, please disregard the trailer which follows. >>> "Christopher Ambler" 04/10/00 04:03AM >>> That's nice, Dave, but you didn't answer the question. - -- Christopher Ambler chris@the.web - -----Original Message----- From: owner-wg-c@dnso.org [mailto:owner-wg-c@dnso.org]On Behalf Of Dave Crocker Sent: Monday, April 10, 2000 12:06 AM To: Simon Higgs Cc: wg-c@dnso.org Subject: Re: [wg-c] Two Para D&I Summary (PPE) At 11:48 PM 4/9/00 -0700, Simon Higgs wrote: >How does cost-recovery not become price-fixing (illegal in the most >countries including the US)? (here we go again.) constraints on pricing are often entirely legal. there is no such thing as a completely free market system, in the real world. not even hong kong. d/ =-=-=-=-= Dave Crocker Brandenburg Consulting Tel: +1.408.246.8253, Fax: +1.408.273.6464 675 Spruce Drive, Sunnyvale, CA 94086 USA ********************************************************************** The information contained in this electronic message is confidential and is or may be protected by the attorney-client privilege, the work product doctrine, joint defense privileges, trade secret protections, and/or other applicable protections from disclosure. If the reader of this message is not the intended recipient, you are hereby notified that any use, dissemination, distribution or reproduction of this com- munication is strictly prohibited. If you have received this communi- cation in error, please immediately notify us by calling our Help Desk at 212-541-2000 ext.3314, or by e-mail to helpdesk@rspab.com ********************************************************************** ------------------------------ Date: Mon, 10 Apr 2000 17:13:22 +0200 From: "Philip Sheppard" Subject: Re: [wg-c] S/K principles Hello WG C, sorry that I was away last week and could not contribute to the good discussions on the S/Kprinciples. Much commentary was good though often went into greater detail than is necessary. The principles as written can not be directly applied as part of an agreement. They express principle (hence their title) - and are offered for another group to turn into a form in which they might apply. Their objective is the establishment of consumer/net user confidence. They are intended as "Criteria for assessing a gTLD application from a registry operator". Thus Brett Fauset's clarification last week is correct. Languages On multilingualism, this point is to my mind unnecessary. We went to pains to ensure that the principles did not imply Latin1 or English exclusivity. They were written to encompass wide variation. Making a new principle that supports multi lingual names diminishes the application of the rest of them to multi-lingual names. Justification I am also concerned about adding too much detail or justification. Better to state a clean list of principles and add justification later if it is deemed necessary. Guidelines or principles This is perhaps just semantics but it seems to me that a set of detailed guidelines may be drawn up from such a set of broad brush principles. Our task should be the broad-brush principles. Initial roll-out Mixing in the consensus statement on 6-10 is confusing. This is not a "Criteria for assessing a gTLD application from a registry operator". It is already in the report anyway and these principles should be for the next 6-1000! Charter and open This is not a black and white choice. Even dot com has some defining characteristics. It is not dot edu, not dot mil and not dot tv. The essence is competition. Strong competition will be provided by names that add value to the name space. A dot com2 will provide poor competition to dot com. The point of the principles is that even a new open gTLD should, as some commented last week, have a defining characteristic. We tried to capture this in the original principles by "3. Differentiation – the selection of a gTLD string should not confuse net users and so gTLDs should be clearly differentiated by the string and/or by the marketing and functionality associated with the string." This is not, as some have suggested, a call for only charter gTLDs. It intentionally leaves it to a registry to be as chartered or as open as they please, so long as they are different to all that has gone before them. Given that the relevant "gone before" is dot com, net and org defining a new open gTLD is pretty simple. So, let me offer this revision of the S/K principles based on the 6 April Weinberg iteration which usefully tightened the phrasing and reduced the number of principles by consolidating some of the original ideas. I have added two points regarding next steps (which are not really principles but seems to be what the WGC thinks is a good idea, as I do). - ---------------------------------------------------------------------------- - - Criteria for assessing a gTLD application from a registry operator, subject to current technical constraints and evolving technical opportunities, should be based on all the following principles : 1. Meaning: A TLD should explain what meaning will be imputed to the proposed TLD string, and how the applicant contemplates that the new TLD will be perceived by the relevant population of net users. 2. Enforcement: An application for a TLD should explain the mechanism for charter enforcement where relevant. 3. Differentiation: The selection of a TLD string should not confuse net users, and so TLDs should be clearly differentiated by the string and/or by the marketing and functionality associated with the string. 4. Diversity: New TLDs are important to meet the needs of an expanding Internet community. They should serve both commercial and non-commercial goals. 5. Honesty: A TLD should not unnecessarily increase opportunities for malicious or criminal elements who wish to defraud net users. 6. Competition: The authorization process for new TLDs should not be used as a means of protecting existing service providers from competition. Next steps In addition WG C recommends that the Names Council sets up a new working group to consider the application of these principles as practical guidelines. The WG C also recommends that the Names Council should charter a working group to develop policy regarding internationalized domain names using non-ASCII characters. - ------------------------------------------------------------------------ So, Jonathan et al, is this something upon which the majority of the WG can agree? Philip ------------------------------ Date: Mon, 10 Apr 2000 08:41:04 -0700 From: Kent Crispin Subject: Re: [wg-c] S/K principles On Mon, Apr 10, 2000 at 05:13:22PM +0200, Philip Sheppard wrote: > Charter and open > This is not a black and white choice. Even dot com has some defining > characteristics. Actually, it is much closer to black and white than you imply. "Charter" in context usually means "*enforced* criteria for restriction of registrations to particular registrants." - -- Kent Crispin "Do good, and you'll be kent@songbird.com lonesome." -- Mark Twain ------------------------------ Date: Mon, 10 Apr 2000 08:56:57 -0700 From: Dave Crocker Subject: RE: [wg-c] S/K principles [Was: Working Group C agenda] At 12:39 AM 4/10/00 -0400, Jonathan Weinberg wrote: > 6. As a general matter, new TLDs should be designed to serve >non-English-speaking as well as English-speaking communities. As Jonathan, This is an excellent example of a very well-intentioned rule that will likely, instead, have ill-effect. Essentially it mandates that new names be non-english and then asks for an effort to go off an study the issue further. That guarantees more, open-ended delay before we get ANY new gTLDs. No, that is not the intention of the rule. Yes, the rule is trying to do a Good Thing. But remember the political environment this activity is suffering from. The rule will be used as one more excuse for delay. d/ =-=-=-=-= Dave Crocker Brandenburg Consulting Tel: +1.408.246.8253, Fax: +1.408.273.6464 675 Spruce Drive, Sunnyvale, CA 94086 USA ------------------------------ Date: Mon, 10 Apr 2000 12:09:04 -0400 From: Jonathan Weinberg Subject: Re: [wg-c] S/K principles Now that Philip's back, I'm delighted to turn to the draft he just circulated as the basis for our work, and to abandon the version I sent round yesterday. I do have a few suggested changes to his draft, which I'll append below. Again, we're operating under a painfully tight timeframe: If we're to have a week for a formal consensus call before giving the result to the NC on 4/17, in time for their discussion and vote on 4/18, the text needs to become final *tonight* (well, tonight US time, in any event). My specific thoughts: To folks within this WG, it's obvious that domain names need not have their primary semantic meaning in English. I do think it's useful, though, to say that explicitly, especially for the benefit of non-English speakers for whom this issue is often very important. In the draft below, I've tried to do so gracefully (and in a way responsive to Dave's concerns). One of the things that came through to me quite strongly from last week's discussion (as well as last month's straw poll) was that WG members do support, and desire, an explicit statement that the initial rollout should include both open and restricted TLDs. Maybe the best thing to do is to break this point out into a *separate* rough consensus point, and vote on it separately, rather than including it in the S/K principles. I've added the words "and desired" to the enforcement principle to convey the sentiment people expressed last week that a registry may choose not to adopt an affirmative enforcement mechanism. I don't think we should recommend that the NC set up a new working group to operationalize these principles, b/c I don't see how the schedule supports it. The ICANN Board has asked the NC to report its views by April 20 so that ICANN staff can develop implementation documents for new TLDs, and subject them to public comment, in time for the Board to take action in Yokohama. That suggests to me that the work of operationalizing these principles, at least for the initial rollout, is going to be done by ICANN staff, not a DNSO WG. Jon - - - - - - - - - - - Proposed rough consensus item #1: The initial rollout should include a range of top level domains, from open TLDs to restricted TLDs with more limited scope. Proposed rough consensus item #2: Criteria for assessing a gTLD application, subject to current technical constraints and evolving technical opportunities, should be based on all of the following principles : 1. Meaning: An application for a TLD should explain what meaning will be imputed to the proposed TLD string, and how the applicant contemplates that the new TLD will be perceived by the relevant population of net users. The application may contemplate that the proposed TLD strong will have its primary semantic meaning in a language other than English. 2. Enforcement: An application for a TLD should explain the mechanism for charter enforcement where relevant and desired. 3. Differentiation: The selection of a TLD string should not confuse net users, and so TLDs should be clearly differentiated by the string and/or by the marketing and functionality associated with the string. 4. Diversity: New TLDs are important to meet the needs of an expanding Internet community. They should serve both commercial and non-commercial goals. 5. Honesty: A TLD should not unnecessarily increase opportunities for malicious or criminal elements who wish to defraud net users. 6. Competition: The authorization process for new TLDs should not be used as a means of protecting existing service providers from competition. Proposed rough consensus item #3: WG-C recommends that the Names Council charter a working group to develop policy regarding internationalized domain names using non-ASCII characters. At 05:13 PM 4/10/00 +0200, Philip Sheppard wrote: >Hello WG C, sorry that I was away last week and could not contribute to the >good discussions on the S/Kprinciples. Much commentary was good though often >went into greater detail than is necessary. The principles as written can >not be directly applied as part of an agreement. They express principle >(hence their title) - and are offered for another group to turn into a form >in which they might apply. > >Their objective is the establishment of consumer/net user confidence. >They are intended as "Criteria for assessing a gTLD application from a >registry operator". Thus Brett Fauset's clarification last week is correct. > >Languages >On multilingualism, this point is to my mind unnecessary. We went to pains >to ensure that the principles did not imply Latin1 or English exclusivity. >They were written to encompass wide variation. Making a new principle that >supports multi lingual names diminishes the application of the rest of them >to multi-lingual names. > >Justification >I am also concerned about adding too much detail or justification. Better to >state a clean list of principles and add justification later if it is deemed >necessary. > >Guidelines or principles >This is perhaps just semantics but it seems to me that a set of detailed >guidelines may be drawn up from such a set of broad brush principles. Our >task should be the broad-brush principles. > >Initial roll-out >Mixing in the consensus statement on 6-10 is confusing. This is not a >"Criteria for assessing a gTLD application from a registry operator". It is >already in the report anyway and these principles should be for the next >6-1000! > >Charter and open >This is not a black and white choice. Even dot com has some defining >characteristics. It is not dot edu, not dot mil and not dot tv. The essence >is competition. Strong competition will be provided by names that add value >to the name space. A dot com2 will provide poor competition to dot com. The >point of the principles is that even a new open gTLD should, as some >commented last week, have a defining characteristic. We tried to capture >this in the original principles by >"3. Differentiation – the selection of a gTLD string should not confuse net >users and so gTLDs should be clearly differentiated by the string and/or by >the marketing and functionality associated with the string." > >This is not, as some have suggested, a call for only charter gTLDs. It >intentionally leaves it to a registry to be as chartered or as open as they >please, so long as they are different to all that has gone before them. >Given that the relevant "gone before" is dot com, net and org defining a >new open gTLD is pretty simple. > >So, let me offer this revision of the S/K principles based on the 6 April >Weinberg iteration which usefully tightened the phrasing and reduced the >number of principles by consolidating some of the original ideas. I have >added two points regarding next steps (which are not really principles but >seems to be what the WGC thinks is a good idea, as I do). > >---------------------------------------------------------------------------- >- >Criteria for assessing a gTLD application from a registry operator, subject >to >current technical constraints and evolving technical opportunities, should >be based on all the following principles : > >1. Meaning: A TLD should explain what meaning will be >imputed to the proposed TLD string, and how the applicant contemplates that >the new TLD will be perceived by the relevant population of net users. > >2. Enforcement: An application for a TLD should explain the mechanism for >charter enforcement where relevant. > >3. Differentiation: The selection of a TLD string should not confuse net >users, and so TLDs should be clearly differentiated by the string and/or by >the marketing and >functionality associated with the string. > >4. Diversity: New TLDs are important to meet the needs of an expanding >Internet >community. They should serve both commercial and non-commercial goals. > >5. Honesty: A TLD should not unnecessarily increase opportunities for >malicious or >criminal elements who wish to defraud net users. > >6. Competition: The authorization process for new TLDs should not be used as >a means of protecting existing service providers from competition. > >Next steps >In addition WG C recommends that the Names Council sets up a new working >group to consider the application of these principles as practical >guidelines. The WG C also recommends that the Names Council should charter a >working group to develop policy regarding internationalized domain names >using non-ASCII characters. > >------------------------------------------------------------------------ > >So, Jonathan et al, is this something upon which the majority of the WG can >agree? > >Philip > > > > > > > > ------------------------------ End of WG-C-DIGEST V1 #76 *************************