RE: [nc-whois] Summary: Comments on Accuracy.
Thanks to Thomas for preparing these useful summaries. To further assist the Task Force, I attach a summary (in reverse chronological order) of all substantive comments archived on the comments site. I emphasize that this document is NOT intended to form part of the final report but is simply circulated for the convenience of Task Force members. I second Thomas's recommendation that that the actual text of comments also be reviewed. A number of the comments have injected some new issues onto the agenda with regard to Whois data accuracy. Let me try to put some of these new issues in a context which I hope will prove useful for Task Force members. What follows focuses mainly on the com/net/org space although some of it is applicable to the other open gTLDs. Roughly half of the 3000 respondents to the Task Force survey last year reported that they had encountered some problem with respect to inaccurate Whois data. Presumably the other half had not and perhaps they would not consider Whois data accuracy a major concern. But I do not recall any respondents who thought that less should be done than is being done today to try to collect and maintain accurate Whois data. The current environment, under the RAA's that have been in effect for about three years now, includes some of the following features relevant to Whois data accuracy: · Registrants are contractually obligated to submit accurate and complete Whois data and to keep it current. · Registrars are empowered to treat any "willful provision of inaccurate or unreliable information" as a material breach that can lead to cancellation of the domain name registration. · Another, separate basis for cancellation because of material breach is the registrant's failure to respond for over 15 calendar days to Registrar inquiries about the accuracy of contact details. · For the past six months, a complaint mechanism has been in place on the internic.net site for reports of inaccurate Whois data. This mechanism is intended to trigger Registrar inquiries to registrants about the accuracy of contact details. The Task Force proceeded on the assumption that these features of the current environment, if more rigorously enforced, would lead to some improvement in the overall quality of Whois data. (Obviously there is room for dispute about how much improvement would result.) In the comments on the Interim Report, we are now seeing for the first time in this process some strong advocacy AGAINST these long-standing features of the current environment. For example, some are questioning the conclusion that if a registrar is presented with contact data that on its face is "inaccurate or unreliable," it has the authority (and in appropriate circumstances should exercise it) to cancel the domain name registration. Other comments attack the provision of the current RAA that allows cancellation of registrations if the registrant does not respond to a registrar inquiry within 15 days, asserting that the time period should be longer. Other commenters worry about abuse of the internic.net complaint system and call for restricting the ability of third parties to invoke it. What if these recommendations were followed? If the RAA were changed so that even if obviously false contact data were submitted, the registrar would be required to make an inquiry and give the registrant a chance to submit different data, clearly some registrations that could be cancelled under today's rules would remain viable, at least for some period of time. If the time period within which a registrant would need to respond to a registrar inquiry were extended from 15 days to 30 or 45 days, as some commenters advocate, then inaccurate data would remain in the Whois database that much longer. If some parties were deterred or discouraged from using the internic.net complaint system, then fewer instances of inaccurate Whois data would be brought to the attention of registrars, and thus opportunities to improve data quality would be lost. In short, all these recommendations for change to the current agreements would have the effect of increasing the amount of inaccurate or unreliable data in Whois. They would do nothing to address the problem identified by half the respondents to our survey, and which the Task Force's reports up to now have identified as a serious concern. One key issue is whether, despite this impact, these proposed changes would have countervailing benefits. I am not sure what the countervailing benefit would be of allowing registrants to submit blatantly false data in the first place and giving all such registrants another chance to correct it (or of eliminating the contractual obligation to submit and maintain accurate data, which amounts to the same thing), but maybe someone can suggest one. With regard to extending the 15-day period, it is certainly true that some legitimate registrants could have lost their domains for failure to respond to inquiries in 15 days under circumstances in which they would have responded in 30 or 45 days, but it would be useful if the registrars could supply some data indicating the number of registrations that have been cancelled on this ground in which the reply from the registrant would have been satisfactory but simply arrived too late. (We should also consider to what extent any adverse effects would be ameliorated if the consequence of failure to respond were not cancellation of the domain name but placement on indefinite hold, as several commenters have suggested and as I believe many Task Force members would support.) Finally, with regard to abuse of the internic.net complaint mechanism, there is at least some experience over the past six months with this mechanism. Perhaps we should ask ICANN staff to report to the Task Force on it. Dan Halloran made a presentation on this topic to the registrars in Shanghai. I was present and I append my notes below but it would probably be more appropriate (and no doubt more accurate than my notes!) to ask Dan if he can provide us with any report on how this process has been working and whether there is evidence of abuse. (I was surprised to learn that many of the "complaints" are actually coming from frustrated domain name registrants who had been trying unsuccessfully to get registrars to change or update their Whois data and had decided in seeming desperation to try this route of reporting their own existing data as false!). I am continuing to review the specific comments but I wanted to highlight the fact that we are now getting suggestions that will not strengthen enforcement of the existing obligations but will in fact weaken what those obligations are. We need to think long and hard before we accept any proposal that will have the effect of increasing, not decreasing, the amount of inaccurate Whois data in the system. Steve Metalitz NOTES re internic.net complaint mechanism: I attended part of the Registrar Constituency meeting in Shanghai on Monday, October 28, during which Dan Halloran, the registrar liaison on ICANN's staff, gave a brief presentation on the operation of the Whois data complain mechanism established under the Registrar Advisory of last May (i.e., the complaint form on the Internic.net site). Dan reported that just under 2500 complaints had been received, about half of them in the first two weeks after the complaint form was made available to the public. All have been individually reviewed or given at least a "quick look" by ICANN staff. Most seem to have been motivated by receipt of spam from an e-mail address for which the Whois data corresponding to the domain name was clearly false. A surprising number of complaints were from domain name registrants who had been trying unsuccessfully to get registrars to change or update their Whois data and had decided to try this route. Relatively few appeared to be from intellectual property owners. Dan said that ICANN had not released any analysis of the complaints but that they were distributed across registrars in proportions "close to market share." It came out in the discussion that about 4-500 of the complaints involved registrations with Verisign Registrar, which has the largest market share of registrations. Verisign is taking the position (and did so in the meeting) that it is under no contractual or legal obligation to "close out the ticket" on a complaint, i.e., to advise ICANN of the action taken in response to a complaint of false Whois data. Verisign asserted that this would need to be done "by hand" and they could not justify the expense of doing so. Most of the other registrars present seem to be making such reports, but others may be following Verisign's lead, so it is apparently impossible to tell how many cases are being resolved by the submission of new contact data, and how many are resulting in cancellation of the domain name. There was some discussion of whether this aspect of the process should be automated. Two other concerns were mentioned by registrars in the meeting. A French registrar said that under French law, they would be unable to process any complaint unless the identity of the complainant were disclosed. Dan Halloran mentioned that as part of its screening process, ICANN pings the e-mail address given by the complainant, and complaints which contain an invalid e-mail address are not forwarded to registrars. This apparently satisfied this registrar. Second, one registrar worried about abuse of the complaint process, stating that it had received complaints that were "on their face frivolous" and that had been filed by "people with legal axes to grind" against the registrar or registrant. This registrar said that "an acceptable definition of 'inaccurate'" was needed in order for the complaint system to work as intended. -----Original Message----- From: Thomas Roessler [mailto:roessler-mobile@does-not-exist.net] Sent: Sunday, November 17, 2002 1:45 PM To: Kristy McKee Cc: nc-whois@dnso.org Subject: Re: [nc-whois] Summary: Comments on Accuracy. On 2002-11-17 08:57:44 -0600, Kristy McKee wrote: > Thomas, are all of the submission included in this summary or are > some of the submissions included? I hope that all submissions relevant to accuracy are included. -- Thomas Roessler (mobile) <roessler-mobile@does-not-exist.net> DOMAIN NAMES Summary of Whois TF Comments RML 111402.doc
|