Dear WG Members, Please see below the action items and discussion notes captured by staff from the meeting on 07 August. These high-level notes are designed to help PDP WG members navigate through the content of the call and are not meant as a substitute for the transcript or recording. The MP3, transcript, and chat room notes will be provided separately. See the referenced documents: Drafting Team Discussion – Applications Assessed in Rounds (https://docs.google.com/document/d/1u3UzvZIXzjnxtklgPmqArqff6dyckUbyuzWyLz7d...) Drafting Team Discussion – Predictability Framework (https://docs.google.com/document/d/1lzXxBLMtFr03BKnHsa-Ss7kR7EAJt7pCI1EP3H81...) Excerpts from the chat room are included for ease of reference. Best regards, Julie Julie Hedlund, Policy Director Actions/Discussion Notes 1. Work Track Updates: Work Track 1: Christa Taylor -- We have a call on 08 August at 0300 UTC.  Looking at the CC2 responses on systems and applications fees. Work Track 2: Phil Buckingham -- We began to review the CC2 input on reserved names and proposed changes to the AGB.  Looking at the matrix on reserved names at the top and second level.  Next call is 10 August at 2100 UTC. -- The two "matrix" documents related the reserved names are available here -- Top-Level Reserved Names: https://docs.google.com/spreadsheets/d/1x74w58a9UaTTVulCMmrI45iTiHao6Hf1s8eV... Reserved Names: https://docs.google.com/spreadsheets/d/1WgsYlUpKI_QGuIOlOxtu4uBBj8ZWgD0bTw8G... -- See:  Top-Level Reserved Names: https://docs.google.com/spreadsheets/d/1x74w58a9UaTTVulCMmrI45iTiHao6Hf1s8eV...; See: Second-Level Reserved Names: https://docs.google.com/spreadsheets/d/1WgsYlUpKI_QGuIOlOxtu4uBBj8ZWgD0bTw8G... -- Note that this discussion excludes geographic names at the top level, which will form the basis for Work Track 5. -- Question: On reserved names on registries having a certain number or a limited number of names.  That became an issue.  What was the public comment on that and where does the group stand?  Response: That is one of the topics in the matrix. Work Track 3: Robin Gross -- Met last week to go over some of the legal rights objection questions and independent objector issues.  We got as far as 3.1.7.  Next call is Tuesday, 15 August.  We will get into some of the community objection issues. -- Question: Are we on a correct track to have an independent objector?  What is our experience? -- That issue is ongoing in the Work Track.  There have been positives and negatives about the independent objection.  There was a discussion about the statistics too.  People are re-reading the independent objector's report.  Recordings are available from the last discussion.  The recording from the most recent WT3 call is available here: https://community.icann.org/x/twAhB  -- Question: Which track is addressing independent review?  Response: Work Track 3. Work Track 4: Cheryl Langdon-Orr -- Met last week with a continuing discussion on CC2 responses and proposed text we may be making.  We will continue the discussion with the coming meeting.  Last week we focused on Universal Acceptance.  Next call is 17 August at 0300 UTC.  Finalizing the CC2 responses.  Other pieces of work that are poised to be done. Work Track 5: Jeff Neuman -- Sent emails to the chairs of the SOs and ACs -- GAC, ccNSO, GNSO, and ALAC -- asking for them to put forward representatives to be co-chairs.   Goal is to have a call with the co-chairs in August.  Then we will put out a call for volunteers for the community.  Hoping to have a scope ready by mid-September. -- Question: What is the mandate for this group and how many representatives will participate from each SO/AC?  Response: At this point we've asked for one leader to be nominated from each SO/AC.  Then we will hold an initial call to discussion membership, voting, etc.  All working groups are open to everyone so membership will be open to everyone.  The leaders will discuss question on how voting will be taken.  Terms of reference have not been drafted yet.  The PDP WG is working on an outline.

From the chat:

Anne Aikman-Scalese (IPC): COMMENT:  In Work Track 4, name collisions, two issues relate to who will make the judgment as to whether a string applied for represents a low, medium, or high risk string.  Second question relates to when ICANN.org may grant a discretionary waiver in relation to the 90 day controlled interruption period.  We may need technical advice on these two points. 2. Drafting Team Discussion – Applications Assessed in Rounds: See: https://docs.google.com/document/d/1u3UzvZIXzjnxtklgPmqArqff6dyckUbyuzWyLz7d... -- Change title to "Application Submission Periods".  The Google Document is open to anyone to edit. Requirements Considered -- look at future methodology: -- On the last call the feeling from the group was that we acknowledged that the first application period will be that of a round.  A time period in which we would collect applications, then stop, and then go into an evaluation period -- rather than first come, first served. -- Criteria that we should measure any methodology against -- clarify and predictabililty; no undefined gaps between processing and acceptance; choice of methodology must address the impact on other areas of the program; should not negatively impact operational effectiveness and the fiscal feasibility of the program. -- In 2012 we had a round that started in January and closed in early June or late May.  Between that time period and today we are over 5 years past that and no definitive application period has been announced.  So there still is a lack of clarity and predictability.  Assume that there will be more predictability. -- Question we started talking about last week was whether or not review periods should be built in at least from the next application window and whatever we do after that.  Some said no and others were advocating for review periods. Discussion: -- Don't see how we can presume that there will be no problems in the round so no potential adjustments.  Don't see how we can't allow for some level of review. -- Perference is just to move forward with the next round.  Rather find another mechanism to do some tweaking rather than going through the process as we are doing now.  Move forward as quickly as possible. -- Perhaps we should call them "adjustments" rather than a "review". -- What do we mean by "review"?  Is it an ongoing process or do we stop everything?  Could we calendar multiple rounds?  Need to define this. -- Sounds like -- to recap -- there is a feeling by some that we should go straight into a following round, but also a sense that there should be some kind of lightweight adjustment period. Questions: 1) If we have this adjustment period when would be start that -- what would be a triggering event?  When the application period closes? 2) If we get more applications than expected that would be one variable to think about extend the period between one ending and the next one starting.  What would increase or decrease that time-period between rounds? Discussion: -- On other variables: the processing rate of applications.  Currently only 1000 per year.  Ask SSAC or someone else to come up with another figure. -- We have a policy in place that says we are going to have a round 1 year after the last one, which we didn't do.  We can make a recommendation but without agonizing over our words there since we have no way of controlling what happens after this next round. -- Two choices: 1) recommend a round and a follow up; 2) recommend a round and then the open sort of application period. -- Other ways to think about how we would recommend an adjustment period -- such as a standing committee that looks at issues that arrive to some level and then they have to solve the issues within a certain amount of time.  It is a tough issue and there are intervening events.  -- Even if we said that the third round should be 1 year after the close of the acceptance window for applications, we know that the number of applications could be a variable in the delay in the next round, as well as the rate in processing applications -- 1) 1,000 per year for delegation, and 2) rate at which ICANN could process the evaluations. -- Perhaps we just fix one year and then say depending on the number of applications. -- The real question is where do we want a delay to be.  Are we really serious about worrying about 1,000 per year being the limiting factor?  Some information could be useful there. -- If we can map out the variables then we can provide for predictability. -- One of the reasons we had the long delay in getting TLDs to delegation was not necessarily the processing but the substantive discussions on plural, name collisions, changes to RAA, which took a long time to reserve.  Otherwise we would have seen a faster delegation rate. -- Question: Can we have some information from ICANN on the processing capability based on type of application?  Response: An outcome could be from this WG to present ICANN with variables -- how long would it take you to process if we present these scenarios.  Once we present these variables it isn't going to be a definitive number but will be a scale. -- Reasonable assumption that even when this round opens there will be road bumps.  There also is the issue of how long it takes to get SSAC advice.  Also, the name collision evaluation will take time.

From the chat:

Alexander Schubert: ICANN will provide us with an annual rate of application processing. E.g.: 2,500 applications per year. Annebeth Lange: Do we expect that when the next round opens, the AGB (or whatever it will be called) then will be the same for later rounds? Vanda Scartezini: we did a survey in this region about the intention to apply Maxim Alzoba (FAITID): @Alexander, as I understand it is more like 1000 applications per year now Alexander Schubert: If we get 5,000 applications then for 2 years ICANN won't be able to process any new application submissions anyway. So we have those years to make adjustments. Kavouss Arasteh: "Remedial Period " or "Adjustment Period " instead of " review Period " Alexander Schubert: So number of applications divided by ICANN's processing speed determines the earliest start of the next round! Phil Buckingham:  @ alexander - it is a question  of what  delegation rate  the root zone can take .  10000  per year  possible ? Kavouss Arasteh: Jeff , That seems an option to establish "Stading Panel is Maxim Alzoba (FAITID): Alexander@ I think some ICANN letters might give a hint on a next round time, like https://www.icann.org/en/system/files/correspondence/crocker-to-diaz-26jul17... Kavouss Arasteh: Kavouss Arasteh: Jeff , That seems an option to establish "Stading Panel Vanda Scartezini:  i beleive 1.5 years after the end of the first round looks great Alexander Schubert: Why allowing application submissions when knowing that ICANN can't process any for X years? Let's pile up demand - why avoiding competition? Maxim Alzoba (FAITID): @Alexander, I think it might be an interest rate or something, for example 26k applications give 25 years for the last thousand Maxim Alzoba (FAITID): so finances are secured and it gives some stability Kurt Pritz: I think if you paid someone 185K per application, they could be processed "infinitely" quickly Maxim Alzoba (FAITID): I hope we do not see auctions for queue Maxim Alzoba (FAITID): numbers Susan Payne: well I think there are some things we are discussing which could assist on the evaluation rate - pre-certifying RSPs for example would speed up subsequent evaluation Paul McGrady: +1 Susan.  Why evaluate and revaluate and revaluate and revaluate and revaluate the same people endlessly. Alexander Schubert: Precisely. And some here want to reduce the effective application cost to just  $50k - then we easily get a 5 figure application roster. Cheryl Langdon-Orr (CLO): I suppose many would agree with not having money held Alan Cheryl Langdon-Orr (CLO): WT4 has asked for clarification on that Alan Cheryl Langdon-Orr (CLO): the 1k "limit" I mean Maxim Alzoba (FAITID): Also companies bleeding money during the time in the queue, and it is more than 185 for the single application (in case a single TLD applicand) Vanda Scartezini: for me limiting 1,000 for each round and considering time we had last round ( around 1.5 year)  keep the same lenght looks viable Alexander Schubert: Maybe we simply double the application fee next round - and depending on demand only lower it if ICANN can process fast enough. Trang Nguyen: 2.1.4.2 of the PIRR has some information regarding evaluation timeframe. Per the timeline provided for in the AGB, the estimated timeline for processing 1930 applications is 25 months. ICANN completed IE in 18 months, that's with prioritization taking place late in the process (December of 2012) and with additional time (outreach process) for applicants to respond to CQs. Maxim Alzoba (FAITID): @Vanda, the last time it was 2000 and not 1000 Jeff Neuman: Ultimately, I would love to be able to give ICANN a list of the types of variables for them to model out. Maxim Alzoba (FAITID): digital archery attempt of the queueing Vanda Scartezini: yes - i am considering limiting for 1,000 just to have a number and time enough to not expend too much paying large groups to evaluation Steve Chan: To Trang's comment, PIRR stands for Program Implementation Review Report and is available here: https://www.icann.org/news/announcement-2016-01-29-en Gg Levine (NABP): GAC Advice may be an additional variable Robin Gross: There will always be "glitches" ;-) Jeff Neuman: hopefully less minor glitches in the future Maxim Alzoba (FAITID): @Vanda,  given the 7-8 years between rounds (what we see now), 1000 might not be  enough Maxim Alzoba (FAITID): @Jeff, also we need to hope not to have major ones too Trang Nguyen: ICANN's criteria for vendor selection in the 2012 round was that the firms had to be global with the ability to scale because we did not know what the volume would be. Trang Nguyen: All 3 firms engaged for financial and technical evaluation had the ability to scale. anda Scartezini: maxim, i am proposing 1.5 years between rounds - and for this new round I am not betting on more than 1,000 Aslam G Mohamed: To scale up ICANN can also consider outsourcing? Donna Austin, Neustar: Trang: will ICANN be starting from scratch in terms of building a replacement to TAS or is something else in train? Robin Gross: That makes sense, Jeff. Kavouss Arasteh: Dear Secretariat, pls capture my thoughts in summary Maxim Alzoba (FAITID): as I understand the current model - applicants pay and then , over time , the plan is revealed (or created) Maxim Alzoba (FAITID): in mathimatics  attempts to optimize model with too many independent variables usually have no real solutions Trang Nguyen: @Donna, TAS was retired after Initial Evaluation of the 2012 round and it's not expected that we will resurrect it. 3. Drafting Team Discussion – Predictability Framework: See: https://docs.google.com/document/d/1lzXxBLMtFr03BKnHsa-Ss7kR7EAJt7pCI1EP3H81...) -- If there are no objections as we walk through the document we will start accepting changes. Anticipated Outcome -- Second paragraph talks about policy implementation and governed by the Consensus Policy Implementation Framework -- compliment that and not to change it. Community Engagement -- Last time the process was to put together a general policy framework with a few implementation guidelines. -- Method of developing policy has gotten more complicated.  Strive for "clear recommendations that can be implemented in order to result in a program where there is minimal ambiguity or change needed." -- Hopefully we are producing something that won't be a surprise for people. -- Question: Trying to understand difference between the Consensus Policy Implementation Framework and in the manuals that were developed for implemation in the GNSO procedures?  Responses:  The Expedited PDP has been implemented.  They are part of the framework for developing the policy now approved. -- Question: Where did the problem statement come from and where did this policy come from?  Response: This is the policy process established by the GNSO.  This is one of the overarching issues.  The original charter for the PDP WG had issues.  We grouped the issues logically so they would be in discrete work tracks.  This one is on a predictable model. -- The community collaborated in the development of the Consensus Policy Implementation Framework.  This document is not meant to substitute for the Framework, but provide a predictable way to resolve issues. -- The CPIF was established by GDD at the same time as we were revising the rules about the intersection of policy and implementation.  We are talking about how to deal with issues that come up during implementation.  What this does is accompany that document and say if this kind of process change is happening this is what needs to happen.  How can we do this within the CPIF and within GNSO consensus policy approved by the Board. -- On the GDD Framework that was one of the most collaborative effort between ICANN and the community. -- Don't know how we can get predictability unless we can hold the GNSO and its WGs to schedules. -- For this topic we think about things that came up where we wondered how to resolve the situation (such as digital archery).  There may be issues like that that arise for which we need to have a framework in place to resolve the issues without too much delay. -- From ICANN's perspective we are looking at two phases: 1) policy and implementation phase following the PDP with the  IRT and the CPIF; 2) issues that came up when the program is in operation -- an operational phase.  It would be great to have guidance and a framework for the issues that arise in the operational phase.  This is predominantly for the operational phase.  Does not replace the existing framework. -- The point is to have processes that can resolve issues in the operational phase.  Where does this takes us to the possiblity of a faster process, but is it a result of multi-stakeholder participation? -- There is nothing inconsistent in this proposal with the existing framework. -- Challenge from the last round: If the policy was silent is it an implementation issue or do we have to go back to the GNSO? -- There is a difference between implementation of the policy and implementation of the program.  Issues of intellectual property rights, for example -- very different than digital archery didn't work so we need a new solution to queue applications, or name collisions.  Things that had nothing to do with policy but for which we needed a predicable framework. -- Implementation is different from policy.  -- Suggest that people read this and also the discussion on the role of the implementation review team.  Anyone who has thoughts should comment on the document and to suggest changes and edits.

From the chat:

Paul McGrady: I guess I don't understand what the issue is?  This document, whether or not drafted by Staff, seems to flow directly from GNSO Principle A on new gTLDs.  "New generic top-level domains (gTLDs) must be introduced in an orderly, timely and predictable way." Plus, we are being given a chance to review it (on this call). Steve Chan: There is a footnote in the document to the existing framework that ICANN uses now to implement Consensus Policies: https://www.icann.org/policy/implementation Kurt Pritz: Following Paul's comment, the current policy is: "All applicants for a new gTLD registry should therefore be evaluated against transparent and predictable criteria, fully available to the applicants prior to the initiation of the process. Normally, therefore, no subsequent additional selection criteria should be used in the selection process." What edits are we proposing? Jeff Neuman: Take an example.......How to process applications for purposes of queueing.......In 2012, it was Digital Archery......well, we found out that implementation of that was flawed.  It took MONTHS to figure out a new queuing mechanism.  Hopefully having this framework will result in having a MUCH smaller delay if something like that comes up again Jeff Neuman: (Ignore the example), but focus on the principle Anne Aikman-Scalese (IPC): @Paul - issues arising during implementation was the whole point of Policy and Implementation Working Group work where participants from all stakeholders was open.    Who exactly is working on the revisions to the Framework? Kavouss Arasteh: Predictability is a complex process which depends on many factors on which we may not have all inputs Steve Chan: Thanks Trang, I had raised by head to try and draw that distinction. Kavouss Arasteh: We need to separate the policy development and its implementation Jeff Neuman: "Implementation of Policy" is different than implementation of the program Kurt Pritz: @ Jeff - Can you explain? Kavouss Arasteh: Policy and implementation are two different things Kavouss Arasteh: How we could implement the policy? Anne Aikman-Scalese (IPC): COMMENT:  Jeff, I see your point but you may recall there was huge disagreement in 2012 over what was policy and what was implementation.  Similarly we could easily have folks disagreeing in the next round as to whether something is implementation of policy or operational implementation. You go back down the same hole. Anne Aikman-Scalese (IPC): COMMENT:  Reminder that the whole issue in 2012 was that ICANN CEO and staff considered items implementation whereas GNSO considered them policy.  You actually authored the letter to the Board telling them they had to come back to the GNSO.  So why would your new category of "operational implementation" be any different? 4.  Next Meeting: -- Next call is in three weeks:  29 August at 0300 UTC.  Work Tracks are still meeting.  Google Doc calendar is available at: https://docs.google.com/spreadsheets/d/1O_NpWTXFMNHkJARveqCapEssRt1CaEqwWe8g...   -- As soon as Work Track 5 meetings are established we will send a note to the list.