ABCDEFGHIJKLMNOPQRSTU
1
New gTLD Subsequent Procedures PDP Working Group Initial Report
2
Substantive comments submitted by Working Group members via WG mailing list (April 2018 - June 2018)
3
4
DateContributorTopic sectionTextLink
5
11 Apr 2018Christopher Wilkinson2.12.1 TLD Rollout, 2.12.3 Contractual ComplianceAllow me to amplify and confirm my few Chat comments during the conference call last Monday afternoon, 9 April 2018.
In general the working document (1.12 Deliberations etc .) is an excursion into a working method with which I am quite unfamiliar, so I ask those of you who have this for your bread-and-butter to bear with me for a little while. I also have noted that the document is in the form of questions for discussion and not recommendations at this stage. So I hope that some of my comments may still be taken into account.
Indeed, at some points, I find the drafting on some issues rather uncertain; there are ambiguities and options that - from the point of view of a rather more conventional approach – might be described as loopholes. ICANN and GNSO will no doubt have gathered that the next 'round' will be scrutinised by third parties far more thoroughly than was the case in 2004 or 2012. More specifically:
1. Rollout: It would be helpful to have information about how many new TLDs have still not been implemented, and why.
For instance, after the 24 months allowed, it is not clear why “… extensions…should continue to be available according to the same terms and conditions as they were allowed during the 2012 round.”
In a related question, “The Work Track was also careful to avoid drawing the conclusion that only having <nic.TLD> registered constituted 'squatting' or 'warehousing' “
Taken together these two statements leave the strong impression that the Work Track would in practice accept squatting and warehousing of new TLDs. Was that intended?
If there has been 'unwanted behaviour' – and the subsequent discussion suggests that there has been – then one might have expected a rather more proactive approach to discouraging such in the future.
2. Contractual compliance - pricing for premium domains.
The document discusses the issue of 'pricing for premium domains' but reports that 'The WT has not reached any conclusions on this issue'. Whereas it is quite likely that applicant representations and related authorisations would address prmium pricing.
For instance, in the case of a geographical name, there would normally be a presumption of non-discrimination between Registrants, such as towns or other communities, within that geo-TLD.
More generally, ICANN might consider moving towards a policy whereby the economic 'rent' for a 'premium name' should accrue to the Registrant and not to the Registry. Otherwise it would appear that the Registry would be taking advantage of its monopoly over their TLD in question in an unjustifiably discriminatory manner.
3. Contractual compliance – enforceability of representations
The document reports that the WT considered a proposal 'that all applicant representations should be included in the registry agreement' and that 'There was no agreement … in support of this proposal.'
This would appear to be a rather weak conclusion which might be queried at a later stage because it does not really respond to the four questions indicated under f. Deliberations.
The above is but a summary of the main concerns evoked by this section of the draft. I look forward to discussing these and other aspects in due course.
https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000988.html
6
12 Apr 2018Maxim AlzobaPricing for premium domainsI'd like to make a comment and a note.
1. a note about pricing for premium domains - 'picket fence' protects both registries and registrars from influence of policies on pricing
(also direct price regulation from ICANN will lead to investigations of anti-monopoly agencies around the world[in some countries it will constitute a crime])
2. a comment on "Registry would be taking advantage of its monopoly over their TLD in question in an unjustifiably discriminatory manner."
Fact that a particular company has a monopoly can be established only by the relevant market regulator.
Registrants are free to choose another TLDs, if they are not happy with the terms of the particular ones.
So assumptions that Registries are monopolists are not established facts and we can not use it.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000989.html
7
12 Apr 2018Michele NeylonPricing for premium domainsWasn’t pricing and why attempting to regulate it already discussed to death in the last round?
I have some recollection of this.
I think the key thing with regard to pricing is transparency, which is already baked into the contracts.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000990.html
8
12 Apr 2018Marc TrachtenbergPricing for premium domainsExcept that such transparency was not, and still is not, being enforced by ICANN.https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000991.html
9
12 Apr 2018Vanda ScarteziniPricing for premium domainsAny regulation on price will lead to Cartel organizations and corruption associated to it. Agree that transparency is the only demand we shall make related to pricing and from my view majority of it is in the contract words.https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000993.html
10
12 Apr 2018Rob HallPricing for premium domainsCan someone explain to me what the concern is about transparency on price ?
Registries publish their pricing to their clients, the Registrars.
Registrars set whatever price they want for Registrants.
Are we concerned about transparency of a Registrar not knowing what price a Registry is charging them ?
Or are we concerned that a Registrant can’t tell what price their Registrar is charging them ?		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000994.html
11
12 Apr 2018Rubens KuhlPricing for premium domainsWhile not mentioned in this thread, the only valid pricing concern I saw over the years is that while registries are obliged to inform registrars in advance the raising of renewal prices, the same doesn't happen with registrars and registrants. So a registrant that might prefer to anticipate renewals might not be informed of the upcoming raise.
But since this out of scope of GNSO policies, that's possibly something to raise when RAA is amended.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000995.html
12
12 Apr 2018Kris SeeburnPricing for premium domainsIt needs to be in the RAA for sure. I see no other ways and with the laws around it needs be under the RAA where one can also add compliance etc., I think and pretty sure it makes logical sense to have these in the RAA.https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000996.html
13
12 Apr 2018Rob HallPricing for premium domainsI suspect Registrars would resist that obligation, as they already believe they have too many mandatory touches on a client with mandatory renewal notices and mandatory whois check reminders.
This would add another layer to what is mandatory for them.
I fear we are trying to solve a theoretical problem that doesn’t exist in practicality.
I was on the last 2 RAA negotiating teams and this was never mentioned as a concern.
But I am thankful that it is not now within our remit.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000998.html
14
12 Apr 2018Rob HallPricing for premium domainsI think you will find that every Registrar informs clients immediately, as they go after the opportunity for the Registrant to buy up to 10 years at the lower price.
But you are correct, that is driven by market forces, I don’t think there is any obligation in the RAA to inform the Registrant of the change in their renewal price at the time the Registry informs them, just at time of Renewal. And that might be too late and the price might already be in effect, as the Registry only has to give 6 months notice to the Registrar.
I concur that Registrar-Registrant issues are out of scope for us and governed by the RAA.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000997.html
15
12 Apr 2018Volker GreimannPricing for premium domainsI think the main issues is that with an endless variety for pricing models for especially premium names, it can be sometimes a bit difficult to discern the renewal pricing of a particular domain name.
In the previous world, one only had to look at the standard price list of a registrar, look for a TLD and find your renewal price. If you do that now, you may find a bad surprise down the road when you notice that this price does not apply to your premium name, no matter how big the letters on the website of the registrar pointing out the possibility of
differential pricing between names in the same TLD were.
It has become somewhat consumer-unfriendly.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/000999.html
16
12 Apr 2018Christopher WilkinsonPricing for premium domainsThankyou, all, for several comments and corrections regarding my initial post.
Pace Jeff Neumann's obiter dicta, I feel that I should respond briefly to some of them.
1. Pricing: I did not open a general discussion about Registry-Registrar pricing; I was referring specifically to reported discriminatory pricing for 'premium domains', an issue that was raised in the text.
I expect this also to be an issue in Geo-Names. For example should <cork.munster> turn out to be several times more expensive than <killarney.munster> it is likely that ICANN would hear about it. Verb sap.
1. Competition: The argument about 'choice' is of limited validity in this market, particularly with reference to Geo-names. To follow the previous example, if <cork.munster> is too expensive or already taken, then <cork.donegall> would presumably be of no use whatsoever as an alternative. Furthermore, once the initial registration completed, should the registrant's domain become successful, there is normally no more choice of Registry.
I suggest that further discussion of those matters that are not within the GNSO's or the PDP's mandate be taken off List.
PS: More generally, regarding competition policy, in my view ICANN IS the market regulator.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/001000.html
17
17 Apr 2018Michele Neylonpricing for premium domainsSo the official policies, which are contractually binding, are as follows:

“Section 4.1 of the Expired Registration Recover Policy (ERRP) (found at https://www.icann.org/resources/pages/errp-2013-02-28-en ) states, “Registrars must make their renewal fees, post-expiration renewal fees (if different) and redemption/restore fees reasonably available to registered name holders and prospective registered name holders at the time of registration of a gTLD name”. In addition, Section 4.1.1 of the ERRP states “At a minimum, these fees must be clearly displayed on the registrar’s website and a link to these fees must be included in the registrar’s registration agreement. Publishing of registration fees is not required by the Policy.”		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/001011.html
18
18 Apr 2018Christopher Wilkinson2.2.2. Predictability, 2.2.2.2 Clarity of the Application Process, 2.2.3 Application Submission Periods, Competition policy aspectsA few points regarding this draft (35pp.):

It is not clear at which stage of the process, someone (who?) decided that these were the overarching issues. Also, it should clarified whether references to 'existing policy' refers to the 2007 reports or to the 2012 AGB. It would be preferable if the WG/WTs could cease reference to the 2007 reports. Very few participants today, with the exeaption of a few GNSO members at the time, can really appreciate what that policy was, more than ten years ago.
1.2.2 Predictability: At present the discussion focusses almost entirely on predictability from the perspective of eventual applicants, whereas predictability is equally important from the perspective of the ultimate final users or third parties and that would be impacted by new TLDs.
The reader gathers that the demand for predictability arises from the (external) changes that were made to the policies post 2007 and post AGB. Whence the demand for a predictable process to address such changes as and when they arise in ethe future.
However, one could equally well argue that those problem arose from a lack of multi-stakeholder consultation during the previous policy development processes. For instance the requirement for (enforceable) Public Interest Commitments (PICs) – or their equivalent - was manifestly inevitable and desirable, long before the matter reached the application stages.
1.2.2.2 Clarity of Application Process: There is an underlying assumption in this part of the draft that all applications would be subject to the same guidelines and evaluation processes irrespective of the nature of the proposed TLD. That is very unlikely to be the case because different categories of applications will manifest quite different characteristics. This is elaborated later in the draft (p.26) . Thus the statement to the effect that “Normally no subsequent additional selection criteria should be used …” (p.3) is only credible if (a) there has been much more thorough prior consultation with all stakeholders and (b) there are distinct selection criteria for identified categories of applications.
Arguably, many of the problems with the previous 'round' arose from the attempt to fit too many different categories of applications into one standardized policy and process. We should learn from that experience and not try to do it again.
We encounter the analogous issue with the concepts of “criteria fully available to the applicants prior to the initiation of the process.” and the “pre-defined roadmap”. (p.12). That is all very well and very desirable provided that the necessary degree of differentiation has been achieved. To date there is little indication of that.
For other aspects of the application process, such as multiple applications and Registry Service Providers see below under Competition. For application prioritization, see below under multiple rounds and 'batches'.
1.2.3 Application Submission Periods. For the credibility and acceptability of another new TLD programme, in the global context, it is absolutely essential that the vast imbalances of the previous round be recognised and corrected. This is the principal over-arching issue that is not recognised - indeed hardly mentioned – in the current draft. For instance, there is but one reference to IDN and the reference to Applicant support and Community applications are referred to 'outreach' to ALAC and the GAC suggesting that At Large and GAC have not effectively participated to date. (p.11). It is time GNSO took applicant support on-board on its own behalf.
The principal constraint on managing the evaluation of new TLD applications is the capacity of the ICANN.org Staff. Questions about ICANN's “scale” (p.13) and hints about “ICANN subcontract[ing]…tasks:” ( p.6) amount to wishful thinking in the current budgetary context. No. We need a positive statement from ICANN staff about how many applications per month and per topic they will be able to process.
The next 'rounds' should be focussed in phases over time, and designed to address specified priorities. Their scale at any point in time should be related to the evaluation capacity of the ICANN staff and related (independent) community support. From this point of view, the characterization on page 26 is a useful start, but incomplete and lacking prioritization. For instance, there is no mention of IDN TLDs, and we already know that Geographical Indications will have to become a specific category, not unlike Brands.
Thus, most of the discussion (pp. 17-20) could conveniently be dropped from the draft, since none of those 'Models' are likely to be sustained.
Competition policy aspects: The authors of the draft seem to be unaware that whilst they seem to believe that the 2012 round contributed to competition, most of the – limited – data that is referred to rather indicates an increase in concentration.
Thus we have three companies who applied for more than 100 new TLDs! Furthermore one of them is a Registrar, presumably taking advantage of the flawed concept of vertical integration in the 2012 round.
The next round should ensure (a) that diverse entities had access to the application process and (b) that new applicants have the option of engaging an RSP that is independent of pre-existing Registries or Registrars. From this point of view, the WT should have considered a cap on new applications from individual entities. That would be a particularly sensitive issue in the case of geographical names.
There would also be great merit in having an independent market for so called 'Registry back-end' services, also known as Registry Service Providers (RSP). But we learn from the draft that “The top five RSPs accounted for over 70% of the 2012 new gTLD applications.” We are not told who they were. Some of the RSPs are also historical Registries or Registrars. It is difficult to conceive of a competitive DNS market unless there is at least a clear structural separation between the RSP activities and the Registry and Registrar activities within the same entity.
Finally, it is rather odd that the WT determined that accreditation of RSPs was not required, whereas- to the best of my knowledge – ICANN does accredit escrow service providers, albeit they are a rather less critical function than the RSP.
New bodies and entities required to implement the proposals
Predictability: Standard Implementation Review Team
Consensus Policy Implementation Framework (CPIF)
Implementation Review Team (IRT) for Subsequent Procedures
The draft refers to the “Duty of the ICANN organisation… to follow recommendations of the Standing IRT” (p. 11). This is potentially problematic: The ICANN staff are accountable to the CEO and Board. The Board is responsible for the public interest, including taking account of GAC advice.
In what sense has the ICANN Board a “duty” to respect the IRT?
Also, in the light of the complexities of the proposed IRT (page 10), who is responsible for its creation, with what budgetary resources or authority?
What is not in the draft report?
The draft refers on more than one occasion to the fact that “… the WG has not agreed upon a set of arguments …” (p.2), or “ …it has not yet reached any conclusions on specific success metrics.” (p.3).
It is not clear whether these are the result of disagreements within the WT or a decision that they were not necessary - ?
Furthermore, regarding establishing additional categories beyond the ones coming from the 2012 round, we have “It is time consuming to develop policy using an approach with many categories” (p. 25).
This reverts to the discussion above about categories of TLDs and the merits of dealing with applications in batches scheduled according to priority and distinct categories. Even within the context of the existing draft we have 14 categories which may be needed (p. 26), most of which would in any event require distinct evaluation criteria.
Multiple rounds and 'batches'
In my view, The evaluation and implementation of new TLDs will become quite specific to the objectives and policies of each application. The time for 'vanilla' generic applications is probably past. In any event, the hundreds of generic applications in 2012, many of which – I understand - are still not operational, suggests that even more Generic gTLDs are hardly a global priority for the next rounds.
This is tacitly accepted by the draft's recognition of thirteen or more categories of TLDs in addition to the standard open registries – 2012 category. (p.26)
Apart from the technical and security related considerations, which should be maintained as a common trunk to all TLDs, most of the other evaluation criteria will differ among categories. It is quite possible to prepare these criteria, and it is regrettable that a start has not yet been made.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-April/001034.html
19
21 May 2018Vanda ScarteziniSection 1.51. ( Section 1.5 item c.2 ) - I am still not think that to implement this recommendation ( refund applicants) will be an easy and/or fair task- I believe keep the excess in a fund for destinations as described in item c.4, will cost less to ICANN’s administration, otherwise the administration cost may generate more fee cost to all.
2. ( Section 1.5 item e) : What are the considerations/implications if we move to continuous rounds, especially as it relates to ensuring the program is run in a revenue neutral manner?
Refund solution will not work in this alternative. the Application administration for a continuous rounds will probably use the same “more accurate applicant fee” and will need to have a more reduced but permanent structure to face continuous application. Adjustment of the applicant fee will be a continuous task and shall be monitoring annually reporting to the community. Additionally the whole external support, as independent panels, shall be registered as a list of persons open to join in a different panels, any time, for an specific task they are experts. For each application, when needed a reduced panel of 3 will be selected for a working day to analyse the application.
1. Under the circumstance where the application fee is set at the floor amount, do you have additional suggestions or strategy on the disbursement of excess funds?
I believe the best alterative will be in Fund to be disbursed as defined in isection1.5 c.4.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-May/001069.html
20
21 May 2018Christopher WilkinsonTax havens, 2.10.2 Registrar Non-Discrimination / Registry/Registrar Under Notes 2. The co-Chairs are aware of my reservations about the restriction on substantive discussions.
When I join a PDP WG Call it is precisely to discuss the proposed outcome of the WTs of which I am NOT a member.
Under Section 1.5 Caveat. I note that the issue of tax havens has been recognised. The jurisdiction of the incorporation of a Registry IS an issue, that may well give rise to restrictive conditions.
- at the present time when several international entities and governments are concerned with tax fraud and tax evasion, ICANN's image would not be enhanced should we be seen to be facilitating tax evasion.
- regarding Geographical Names, I have already pointed out to WT5 that one might expect that public authorities issuing letters of non-objection (or similar) would normally require incorporation within their jurisdiction.
Under Section 1.10.2 Registrar Non-Discrimination: I would enter a general reservation about the neutrality of Registrars among Registries under current conditions of 'vertical integration'. I shall return to this issue when we have had an opportunity to see the CCT-RT final report.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-May/001073.html
21
11 June 2018Christopher Wilkinson2.3.1 Competition, Consumer Choice and Consumer Trust, 2.3.2 Global Public Interest, 2.3.3 Applicant Freedom of ExpressionSection 1.3 – A few quick comments:
1. Competition …As I have pointed out before, it is most difficult to have a serious discussion without the CCT-RT Report. Meanwhile, there are strong indications that the 2012 Round contributed to concentration in the DNS markets:
- Registry Service Providers: only a few major providers. Some of which are themselves Registries or Registrars, leading to risks of conflict of interest.
- Certain Registrars accumulating very large portfolios of Registries; an anomaly arising from the flawed implementation of vertical integration.
2. Global Public Interest:(a) We need more clarity as to the UDHR limitations to freedom of speech as against claims elsewhewre that there are freedom of speech rights for 'applicants'.
The most important freedoms of speech - at least in the context of Geographical Names – is the freedom of speech of Registrants. I do not understand the scope and objectives of the eventual freedom of speech of Registries.
(b) First bullet: The original purpose of vertical integration was to permit new Registries to register names directly, at lease before reaching a certain threshold. Today, new Registries could reasonably anticipate anti-competitive bnehaviour by accredited Registrares who are their natural competitors.
(c) Mandatory PICs: Those imposed by ICANN on the basis of community and GAC advice to the Board.
Voluntary PICS: Those proposed by the applicant who shall then be obliged contractually to respect them permanently (subject perhaps to a contractual revision procedure.)
ICANN supervision of contractual compliance, including transparency, becomes important in this context.
Sensitive Strings associated with GAC Category 1 Safeguard Advice: Just to note that the third bullet is internally inconsistent.
If the PICs have effectively prevented abusive behavior, it follows that there will not be data to demonstrate that effect. So the mandatory PIC will have done its job!
3. Applicant Freedom of Expression:(b) There needs to be a balance between the freedom of Expression of the Registrants and that of the Registry. (a) the jurisdiction of the incorporation of the Registry should be transparent for purposes of tax and other requirements (b) At least for Geo-Names, the jurisdiction of incorporation of the Registry must be the same as the territory or community relating to that Geo-Name.
There cannot be a distinction between the jurisdiction of the territory and the jurisdiction of the Registry.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-June/001114.html
22
11 June 2018Vanda Scartezini2.8.1 Objections, 2.8.2 Accountability MechanismsSection1.8
Recommendations 3 – though I am totally in favor to respect intellectual property, this is quite difficult to follow and enforcement since Trade Mark is territorial.
International recognized brands have no problem to enter clearing houses, but it is not yet clear to me how local brands, trying to enter into the clearing house and found another brand similar to them from another country, as legitimate as its own, can really be protected…
We had no case till now, since we had very few organization in the whole South Hemisphere entering into 2012 bid. But now, with more knowledge spread, this will change…. Just a comment. We cannot change WIPO’s rules.
Recommendation 12 – I believe we need to define better Rules of Proceedings and not only the Process itself. Remembering that for 2012 was defined that “community interested strings” should collected worldwide letter of support from those communities the string will be set for. In the dispute time this had no real value (though had cost a lot of money) since the bid was set without ask the participants to prove they had make similar collection of support. If the Rules of Proceeding were not very clear and be followed, the dispute will not be fair.
Section 1.3 –
Besides the inclusion of topics from Final Report of Competition, Consumer Trust & Choice I have no other comment to add.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-June/001105.html
23
11 June 2018Christopher Wilkinson2.7.3 Closed Generics"this is one of the most contentious issues ..."
In what respect? THE Board decision on the last round seems to me to be quite enough.
1. If the applicant has an uncontested right to a name, then Okay, but I fail to see what the business model or other advantages might be.
2. If the application is for a Closed TLD on the basis of a name (dictionary, Geographical or other) to which the applicant has NO prior right, then NO.
https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-June/001117.html
24
11 June 2018Mike Rodebaugh2.7.3 Closed GenericsIt's contentious because the Board decision was made in complete 'top-down'
fashion. Not only without real community input, but also actually CONTRARY
to the GNSO policy and AGB which *allowed* closed generics. No public
interest has ever been identified for treating TLDs different than generic
.com domains, for example.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-June/001119.html
25
12 June 2018Christopher Wilkinson2.7.3 Closed GenericsOh! was that addressed to me?
Let us take time out during tomorrow's WT5 call to discuss it.
In short, from my point of view, the closed TLD is OK if the applicant already holds a global right to the name.
Otherwise the total privatisation of generic words is not acceptable. Generic TLDs have to be open.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-June/001133.html
26
13 June 2018Vanda Scartezini2.7.3 Closed GenericsI tend to agree with you. The Right to a name that belongs to you rightfully, is quite a principle.
The issue that will be no business model for them, not our problem. Is their money, their problem.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-June/001136.html
27
14 June 2018Christopher Wilkinson2.8.1 Objections1.8.1. Objections
In the light of the extensive discussion of objections, their 2012 antecedents, their procedures and costs, allow me to mention that the authorization/non-objection requirements for applications are also being discussed in Work Track 5 (Geographical Names).
Several WT5 Members (including myself among others) have argued that the 2012 objection procedures are quite inappropriate for Geographical Names. In this context, I consider that all applications for geographical names should benefit from appropriate prior approval or non-objection from the competent community or public authority.
The discussion in Section 1.8.1 (pp. 4-18) goes a long way towards reconfirming this point of view. ICANN cannot expect that all the government and local authorities, world-wide, together with related communities, would have the resources to monitor the whole new gTLD process in order to 'catch' each and every application that might concern their interests and eventual rights, including those for 'non-geographical' use.
The procedures, restrictions and high costs of objections, as described, are inappropriate when applied to applications for geographical names. All such applications should benefit from prior endorsement by the stakeholders most directly concerned.
This pre-condition should also normally obviate the need for subsequent objection procedures.		https://mm.icann.org/pipermail/gnso-newgtld-wg/2018-June/001146.html
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100