Re: [gnso-pro-wg] Proposed Principles for discussion

[Liz Williams <liz.williams@xxxxxxxxx>]

Hi Kristina

Many thanks -- my role in the process is to pose questions to ensure  
full discussion.  I also have to be able to "explain" WG discussions  
to non-expert readers and translate principles into implementation  
where necessary.

More discussion is needed and that can happen tonight.

Liz
.....................................................

Liz Williams
Senior Policy Counselor
ICANN - Brussels
+32 2 234 7874 tel
+32 2 234 7848 fax
+32 497 07 4243 mob




On 14 May 2007, at 18:05, Rosette, Kristina wrote:

> Liz and all,
>
> See my comments below in CAPS.  As a general note, many of these
> proposed principles were drawn from PoC reports, the TLD summaries
> and/or third party reports (such as the Summit Strategies report.)
>
> (Liz, if my answers are not clear or you have further questions, let's
> discuss off-list.)
>
> K
>
> -----Original Message-----
> From: owner-gnso-pro-wg@xxxxxxxxx [mailto:owner-gnso-pro-wg@xxxxxxxxx]
> On Behalf Of Liz Williams
> Sent: Monday, May 14, 2007 5:44 AM
> To: Rosette, Kristina
> Cc: gnso-pro-wg@xxxxxxxxx
> Subject: Re: [gnso-pro-wg] Proposed Principles for discussion
>
> Kristina
>
> Thanks so much for sending these.  A couple of questions for the group
> to consider.
>
> 1.  It was not my sense of the direction of the group that principle
> 1 -- all new TLDs must provide RPMs.  Could that be confirmed please?
> THESE ARE PROPOSED PRINCIPLES FOR DISCUSSION AND I BELIEVE IT'S
> IMPORTANT THAT THE WG DISCUSS THIS ONE.
>
> 2.  With respect to 1, the existing recommendation, why isn't the
> existing RAA sufficient to address that principle?  ANSWER VARIES
> DEPENDING ON WHICH BIT OF THE RAA YOU'RE REFERENCING.  (A) UDRP:  IT
> DOES NOT (AND WAS NOT INTENDED TO) DETER OR PREVENT ABUSIVE
> REGISTRATIONS.  (B) 3.7.7.9 ("RNH SHALL REPRESENT THAT, TO THE BEST OF
> THE RNH'S KNOWLEDGE AND BELIEF, NEITHER THE REGISTRATION OF THE
> REGISTERED NAME NOR THE MANNER IN WHICH IT IS DIRECTLY OR INDIRECTLY
> USED INFRINGES THE LEGAL RIGHTS OF ANY THIRD PARTY"):  REPRESENTATIONS
> (AND WARRANTIES, FOR THAT MATTER) ARE NOT SELF-ENFORCING. IN OTHER
> WORDS, THIS  REPRESENTATION IS SUFFICIENT ONLY IF (i) VIOLATION OR
> BREACH OF IT CONSTITUTES A MATERIAL BREACH OF THE AGREEMENT ON WHICH
> TERMINATION OR CANCELLATION CAN BE BASED; AND (ii) REGISTRARS DO, IN
> FACT, TERMINATE OR CANCEL DOMAIN NAME REGISTRATIONS ON THIS GROUND.  I
> AM NOT AWARE OF ANY REGISTRAR THAT DOES, IN FACT, TERMINATE OR CANCEL
> DOMAIN NAME REGISTRATIONS FOR VIOLATION OF THIS PROVISION. (IT  
> WOULD BE
> WORTH KNOWING, THOUGH.  PERHAPS AN AREA TO BE EXPLORED IN REGISTRAR
> AUDITS?).  IT DOES NOT SEEM LIKELY TO ME THAT A SUFFICIENT NUMBER OF
> REGISTRARS TERMINATE A SUFFICIENT NUMBER OF REGISTRATIONS ON THIS  
> GROUND
> TO CHARACTERIZE THIS PROVISION AS A SUFFICIENT RPM.  I DO BELIEVE,
> THOUGH, THAT IT'S WORTH KEEPING IN THE RAA.
>
> 3.  I have attached an overview of the principles and proposed
> recommendations and implementation guidelines -- take a look at
> Recommendation #2 & #3 on p3.  I would have thought that was  
> sufficient
> for new TLDs and for registrations within those TLDs  MAY BE  
> SUFFICIENT
> FOR NEW TLDS, BUT RECOMMENDATIONS RESTRICTED TO TOP LEVEL, NOT SECOND
> LEVEL, AND OUR WORK IS RESTRICTED SOLELY TO SECOND LEVEL.
>
> 4.  With respect to principle 2, these are ideas that would fall into
> implementation guidelines?  POTENTIALLY, YES.  What happens if a TLD
> applicant propose no RPM in its application?  UP TO COUNCIL AND  
> STAFF TO
> DECIDE.  ONE OPTION IS TO REJECT THE APPLICATION.  ANOTHER OPTION  
> IS TO
> IDENTIFY THE APPLICATION AS INCOMPLETE AND RETURN TO REQUEST  
> COMPLETION.
> WHAT HAPPENED IN THE 2000 ROUND IF AN APPLICANT DID NOT ANSWER THE
> "PROTECTION OF THE RIGHTS OF OTHERS" AND/OR "PREVENT ABUSIVE
> REGISTRATIONS" PORTIONS OF THE RFP?
>
> 5.  Another question with principle 4.  Referring to the survey  
> results,
> you will see that registrants have listed that they have  defensive
> registration results in both sponsored and generic TLDs.   It seems to
> be that eligibility and membership verification mechanisms do not
> prevent defensive registration nor minimise the perceived necessity  
> for
> having RPMs?  We need far more robust statistics to justify the
> reasoning that just having a sponsored/ registrant verified system was
> sufficient to prevent defensive registrations?  AS AN OVERALL POINT, I
> INTENTIONALLY USED "MAY NOT" INSTEAD OF "SHOULD NOT" FOR FLEXIBILITY
> PURPOSES.  I ALSO INTENTIONALLY DID NOT USE "SPONSORED" BECAUSE THE
> "SPONSORSHIP" IS NOT DISPOSITIVE; IT'S THE ELIGIBILITY AND NAME
> SELECTION VERIFICATION.  IT'S NOT A SURVEY; IT'S A QUESTIONNAIRE.  WE
> CAN'T EXTRAPOLATE HARD AND FAST CONCLUSIONS FROM IT FOR THAT VERY
> REASON.  BASED ON THE TLD SUMMARIES, IT APPEARS THAT THE NUMBER OF
> CHALLENGES (LACK THEREOF TO BE MORE ACCURATE) IN THE NOTED TLDS  
> SUPPORTS
> THE "MAY NOT" CONCLUSION.  THE FACT THAT SOME REGISTRANTS SECURED
> DEFENSIVE REGISTRATIONS DOES NOT MEAN THAT THE ELIGIBILITY/NAME
> SELECTION WAS NOT A VALUABLE MEANS FOR DETERRING/PREVENTING ABUSIVE
> REGISTRATIONS.
>
> 6.  Principle 6 is way too deep into the detail of registry operations
> -- surely how a registry accepts payment for an RPM (presuming that  
> they
> have one) is up to them.  DON'T DISAGREE, BUT THIS POINT WAS MADE IN
> SEVERAL POC REPORTS AND THE SUMMIT STRATEGIES REPORT SO IT MADE  
> SENSE -
> TO ME - TO INCLUDE IT HERE.
>
> 7.  Principle 7 -- how do you determine those costs?  Would any  
> registry
> operator have any idea, going into the process, who would  
> participate in
> an RPM to determine what any fee would be, prior to the process  
> actually
> starting?  YES, I THINK IT WOULD.  I WOULD HOPE THAT SUNRISE
> REGISTRATION FEES, IP CLAIM FEES, AND VALIDATION FEES ARE BASED ON
> ACTUAL CALCULATIONS.  THE NUMBER OF POTENTIAL PARTICIPANTS IN THE  
> RPM IS
> TIED TO THE ELIGIBLE RPM AND SCOPE OF THE TLD.  THE COST WILL ALSO BE
> BASED ON WHO DOES VALIDATION (IF ANY) AND AT WHAT POINT.  FOR EXAMPLE,
> REGISTRY COST IS PRESUMABLY LOWEST IF NO VALIDATION DONE AT ALL,
> AUTOMATED OR SPOT CHECKING WOULD SEEM TO FALL IN THE MIDDLE, AND FULL
> VALIDATION WOULD PRESUMABLY BE HIGHEST.  IF A THIRD PARTY WILL BE
> VALIDATING AND APPLICANTS ARE INTERFACING DIRECTLY WITH THE THIRD  
> PARTY,
> THE REGISTRY'S COSTS SHOULD BE LOWER THAN IF REGISTRY WAS DOING
> EVERYTHING ITSELF.  TO USE AN EXAMPLE FROM THE SUMMIT STRATEGIES  
> REPORT
> AND TO KEEP IT TO THE TRADEMARK EXAMPLE USED THERE, COSTS WILL BE  
> LOWEST
> IF VALIDATION CAN BE DONE BASED SOLELY ON THIRD-PARTY PREVIOUSLY
> EXISTING DATABASE (E.G., NATIONAL TRADEMARK OFFICE DATABASE, FOR
> EXAMPLE); COSTS WILL BE HIGHER IF RIGHTS CANNOT BE VALIDATED SOLELY
> BASED ON ONLINE DATABASES (E.G., NATIONAL TRADEMARK OFFICES THAT DON'T
> HAVE ONLINE DATABASES OR THAT DO, BUT ARE NOT CURRENT OR COMPLETE);  
> AND
> COSTS WILL BE HIGHER STILL IF THERE IS NO DATABASE AT ALL (E.G.,  
> COMMON
> LAW RIGHTS FOR WHICH EVIDENCE OF USE WOULD BE REQUIRED.) JEFF AND MIKE
> P. - COMMENTS?
>
> 8.  Principle 8 -- isn't this already done under the existing UDRP
> rules.  NO, NOT REALLY.  UDRP AND RPM CHALLENGES REQUIRE THE  
> CHALLENGER
> TO PROVE A NEGATIVE - THE ABSENCE OF RIGHTS.  THIS PROPOSAL WOULD
> REQUIRE AFFIRMATIVE ACTION BY PARTY SEEKING BENEFIT OF RPM.  TO ME IT
> SEEMS ONLY FAIR THAT THE PARTY SEEKING TO AVAIL ITSELF OF THE RPM
> BENEFIT SHOULD PROVIDE ITS ENTITLEMENT TO IT.  Why wouldn't that  
> system
> remain sufficient?  ABUSE.  It wasn't clear to me what rights the  
> group
> would include in any RPM in any new TLD
> -- that needs further discussion.  YOU'RE CORRECT - WE NEED TO DISCUSS
> THIS.  And then validated by whom at what cost?  COMPLETELY OPEN FOR
> DISCUSSION By the objector?  NO, THAT'S THE DOWNSIDE TO THE CURRENT
> SYSTEM.  (SEE MY COMMENTS ON 10 BELOW).
>
> 9.  Principle 9 -- how is this different from Principle 8? VALIDATION
> VS. TIMING OF VALIDATION.
>
> 10.  Principle 11 -- how could this process be automated?  In any new
> round of TLDs how would one know what applications for registration
> would be false as we have  no way of knowing what would appeal to a
> potential registrant.  THE PROPOSAL DOESN'T REFER TO THE SECOND LEVEL
> NAMES THEMSELVES; INSTEAD, IT REFERS TO THE INFORMATION REQUIRED  
> FROM AN
> APPLICANT TO SUPPORT AN RPM.  I DON'T HAVE A TECHNICAL BACKGROUND,  
> BUT I
> WOULD THINK IT WOULD BE POSSIBLE TO WRITE SOME CODE THAT WOULD GO  
> TO THE
> "RED FLAGS."  AN EXAMPLE MIGHT BE HELPFUL:  I FILED (AND WON) A  
> SUNRISE
> CHALLENGE REGARDING THE AAA.MOBI DOMAIN NAME.  THE ORIGINAL REGISTRANT
> IDENTIFIED ITS TRADEMARK REGISTRATION NUMBER AS 123456.  WHILE IT'S
> THEORETICALLY POSSIBLE THAT A SUNRISE APPLICANT'S TRADEMARK  
> REGISTRATION
> NUMBER IS 123456, THE ODDS ARE SO SLIM AS TO MAKE IT QUITE LIKELY THAT
> THE INFORMATION IS FALSE.  UNDER THE PROPOSED PRINCIPLE, THAT
> REGISTRATION NUMBER WOULD BE FLAGGED AND THE APPLICANT COULD BE
> CONTACTED TO CONFIRM ACCURACY OF IT.  AS THERE WAS NO SUCH  
> MECHANISM FOR
> .MOBI, IT WAS NECESSARY TO PAY CHINESE COUNSEL TO SEARCH THE CHINESE
> TRADEMARK OFFICE REGISTRATION RECORDS TO CONFIRM THAT (A) THERE IS NO
> SUCH REGISTRATION WITH THAT NUMBER; (B) THE SUNRISE APPLICANT DID NOT
> OWN ANY TRADEMARK REGISTRATIONS; AND (C) NO THIRD PARTY  
> REGISTRATION FOR
> A SIMILAR MARK HAD A REGISTRATION NUMBER COMPOSED OF THOSE NUMBERS.
> THAT EXPENSE SHOULD NOT HAVE BEEN NECESSARY GIVEN THE STRONG  
> LIKELIHOOD
> THAT 123456 WAS NOT A "GOOD" REGISTRATION NUMBER. THIS DOES NOT MEAN
> THAT THE REGISTRY SHOULD NOT KEEP THE SUNRISE REGISTRATION FEE FROM  
> THE
> APPLICANT THAT PROVIDES FALSE INFORMATION; IT SHOULD.  (TWO OTHER
> COMMENTS:  IDEA WAS ADAPTED FROM THE SUMMIT STRATEGIES REPORT.  MAY  
> NOT
> BE NECESSARY IF ALL RIGHTS ARE VALIDATED PRIOR TO PARTICIPATION IN  
> RPM.)
>
>
> 11.  Principle 12 -- isn't that the same as Principle 8  NOT  
> REALLY, BUT
> I SEE YOUR POINT.  INTENDED TO GO MORE TO SCOPE OF RIGHTS THAN
> VALIDATION.
>
> 12.  Name Watch Service -- isn't it hard to comment on this idea  
> without
> more details about what it's about?  IT'S DISCUSSED IN DETAIL IN THE
> CHARTS POSTED LAST WEEK.  SHOULD HAVE NOTED THAT.  It sounds like a
> service a registry would offer to registrar customers (and would be
> subject to the Registry Services Funnel process) OR is it a registrar
> service OR is it a portfolio management service offered by trade mark
> attorneys to their customers?  NONE OF THE ABOVE.  .NAME OFFERED
> SOMETHING SIMILAR.
>
> Talk to you tonight.
>
> Liz
> .....................................................
>
> Liz Williams
> Senior Policy Counselor
> ICANN - Brussels
> +32 2 234 7874 tel
> +32 2 234 7848 fax
> +32 497 07 4243 mob