On the GA list, Karl Auerbach writes that he has been convinced by the discussion there that he should vote against WLS today. I'm not sure he should be.

At least my most important concerns about WLS are actually resolved in the general counsel's report:

• The perverse incentives for registrars created by WLS - namely, not extending existing registrations, and letting a WLS subscription ripen - are kind of mitigated by the redemption grace period.
• The business model to hoard domain names in order to later sell them off through WLS subscriptions (and subsequent "deletion") is addressed by the agreement to prevent sponsoring registrars from placing WLS subscriptions on a domain name in the 60 days before the name expires. (Note that this restriction is messy and entirely unnecessary when the current registrant wants to place the subscription!)
  
  At the same time, this removes part of registrars' incentives to hoard domain names in order to unfairly compete about WLS subscriptions - delete the name when a subscription is placed through the sponsoring registrar, don't delete it when the subscription is placed through a competitor.
  
  (Actually, the version of this argument given by Louis in his report is not accurate.)
  
  Once again, this is not the perfect solution - the correct approach would be to mandate the point of time when registrars send the DELETE command to the registry.

Note that these are genuine policy matters, and clearly within ICANN's scope.

The single most important concern which remains - and it's the one George and friends are emphasizing - is that WLS may put existing registrar-based approaches out of business. The existing used-domain dealers demand that their business models be protected. I'm sorry to say this, but they are pretty close to Hollywood's crusade against the net in that - Bruce Schneier paraphrased the latest legislation around intellectual property like this: "They are trying to invent a new crime: interference with a business model."

ICANN should not get into the business of prosecuting this "crime" in the registy/registrar market - and that's precisely what not accepting WLS would mean. Don't let George's argument fool you, when he says that "changing the contract" amounts to "action" and "not changing the contract" amounts to a hands-off approach. I'm sorry, George - this is (good, and, it seems, effective ;) rhetorics, and not an argument: Hands-off means that you let others do what they want. Normally, that's equivalent to "do nothing". However, when to "do nothing" means to forbid an activity, a hands-off approach implies rubber-stamp approval.

Of course, this was only half of the argument, so far: I didn't address the fact that existing competition would be replaced by a monopoly. It's the context in which I believe that protecting business models is a good idea and a requirement: When existing, lawful business models would be threatened by a new business model which can only be implemented as a monopoly. There are two tests in this:

• Does the new business model necessarily imply a monopoly service? The answer is obviously yes.
• Are the existing business models lawful? Can current expired-domain registration services compete without registrars violating their accreditation agreements by letting others control "their" transactions with the registry? I'm not sure that this question will ever be answered based on information easily accessible to ICANN and the public. But there are some indications that the answer is "no". Ups.

Given the lack of information, and given that this is a classical question for governments and courts to address, I believe that ICANN should do precisely that: Leave the question to governments and courts, i.e., do not address it in your decision.

Thus, ultimately, I would strongly recommend that ICANN approves this service - fully aware of the perspective that it will quite likely be sued for this decision.

Finally, there is one relatively deep point to this, well worth the board's attention - for the future, because it's too late to address this for WLS: It's Steven Heath's question whether new monopoly services based on the registry's data should automatically be the registry's privilege, or whether these services should be subject to a public bidding process.

I'm not entirely sure whether Louis' argument that the .name services set a precedent for this actually works: After all, with .name the new service was introduced from the very beginning. There may be some merit to treating services to be added later differently than services included in the original TLD concept.