[registrars] WLS comments from Registration Technologies, Inc.

[Jim Archer <jarcher@registrationtek.com>]

This email contains our comments on the WLS proposal.  The comments are included herein and attached as a PDF file.  I apologize that these are a bit late to the table.  However, I feel we have raised significant issues not covered in the Constituencies response by the drafting team and ask that these be reviewed for possible inclusion.


Jim

*****************************
Jim Archer, CEO
Registration Technologies, Inc.
10 Crestview Drive
Greenville, RI 02828
voice: 401-949-4768
fax: 401-949-5814
jarcher@RegistrationTek.com
http://www.RegistrationTek.com



***************************************************************************
**************


WLS Proposal Comments from Registration Technologies, Inc.
January 16, 2002


Registration Technologies, Inc., appreciates the opportunity to review the Verisign WLS proposal.  After careful review we have prepared the following 
comments.  Please feel free to contact us with any questions or comments you may have.


Executive Summary
After a careful review of the available material, we have decided to oppose 
the implementation of this proposal.  We strongly feel that the proposal, as presented, contains provisions that are detrimental to registrars, resellers, registrants and consumers generally worldwide.

Of particular concern is Verisign's stated desire to enter the secondary domain marketplace: a legitimate and established market segment Verisign currently is not present in, to use their monopoly advantage to bring about 
the "Elimination of many desired domain name registrations from the speculator market."  In other words, Verisign intends to enter an existing market segment in which they currently have no presence and with a stated objective to eliminate existing participants in that market segment.

Further, we question why Verisign is permitted to control this comment process.  As a monopoly, Verisign has no place regulating itself.  It is not made clear in the proposal that anyone other than Verisign itself shall 
evaluate these comments and decide whether or not it will be launching this 
service.  We believe that as a regulated monopoly, Verisign has no place mediating this process.  Several significant players who have long cultivated the established market segment for secondary domains, such as BuyDomains, are not registrars and have been offered no opportunity to be heard in this process which affects their core business.

We point out that the material we received to evaluate was incomplete.  The 
proposal states that registrars would be required to complete a new "service agreement" but this agreement has not been presented for review. Therefore, this proposal is incomplete and missing a key element.

Additionally, the time frame for preparing comments was very short.  This is especially a problem since Verisign did not provide the service agreement for review.

Detailed Comments
We have organized our comments into four major topics; Ethical Issues, Legal Issues, Technical Issues, and Practical and Administrative Issues.

Ethical Issues

We feel that the primary marketing mechanisms for selling waits will be fear and false hope.  In the case of a registrant, a wait should have no value, as they can simply renew their domain.  If a registrant initially registers their domain or renews it for more than one year, then there is no value to the registrant at all.  Most likely, these waits will be sold by scaring registrants, as SnapNames currently does.

Further, registrants do not get any right of first refusal or right to deny 
third parties buying a wait for the registrant's domain name.  We feel that 
if this service does go forward that existing registrants should be allowed 
to purchase a wait on their domain name at any time and that this wait should take precedence over any other wait, whether it was placed previously or not.  Of course, as explained, it is our view that no registrant who is sufficiently well informed would ever want to buy a wait on their own registered domain, but that the proposed system creates an ideal opportunity for a con game to be played to victimize the uninformed or naïve registrant, and we feel that some protection against this should 
be built in.

As for those who are not registrants buying waits on a name, the subscriber 
may get nothing for their money.   Most domain names worth placing a wait on will be renewed by their owners.  If a domain name is renewed, then the subscriber gets nothing for their money, except false hope or maybe the ability to extort the registrant.  In the case of false hope, we have done a disservice to the subscriber.  In the case of extortion, we are doing a grave disservice to the registrant.

Because the UDRP does not apply to a wait, we are creating a secondary market in blackmail.  If a person buys a wait on some famous trademarked domain, the trademark holder has no recourse until their domain is transferred, and then must use the time consuming UDRP action to get their domain back.  Therefore, it is in the interest of the registrant to purchase the wait from the subscriber.  Although there is no transfer provision for a wait, nothing stops parties from effecting transfers by simply executing contracts.

This system is also unfair to registrars and their resellers.  The startup procedure, to the extent it has been articulated, is unfair.  Snap has had the ability to market waits for a long time, whereas other registrars have not.  Many of us chose not to sell this product because it cannot guarantee 
success.  SnapNames, on the other hand, is being rewarded because they have 
no such scruples.  Therefore, the Verisign wait is a new product, but Snap is poised to market this immediately.  We barely have time to implement and 
test our software.  If this is to happen, there should be a startup procedure similar to that used by Afilias for the .INFO land rush.


Potential Legal Issues

	Monopoly and Antitrust

It is our opinion that by implementing this proposal, Verisign would be illegally leveraging their monopoly status to enter and dominate an existing market segment of this industry.  Currently, there exists a market 
in "secondary domains," or domains that are registered with the intent of being sold.  Several different business models exist for this activity with 
a variety of pricing models.  Despite Verisign's view, this is completely legal and legitimate activity.

Several registrar and non-registrar companies compete in this market segment, providing competition that benefits consumers by offering a variety of business models and pricing.  For example, Dotster, Enom and SnapNames all concentrate on registering names pursuant to a customer request.  Dotster uses an auction model in which customers pay only if they 
win the name, although the price may be high.  SnapNames uses a subscription model in which customers pay regardless of whether or not they 
win the name, but the price is always fixed.  BuyDomains has an inventory of names and prices vary.  Clearly, there is an established market that is offering consumers a wide variety of options.

Verisign, with this proposal, is seeking to enter this market segment.  As stated in their proposal, they would have a 100% success rate in acquiring domain names for their customers.  This would ruin the business of all the existing participants in this market.  Registrars would be left with no choice but to cooperate with Verisign on terms dictated by Verisign, contracting the existing wide range of consumer choice in pricing and pricing models.  Non-registrars would be cut out of the secondary domain market segment entirely.  Further, Verisign has stated in their proposal that they have set a high price point deliberately, the expressed purpose:

"Elimination of many desired domain name registrations from the speculator
market so that the current excessive demand on operational resources is reduced
and system access is maintained at a much more reasonable level." [Emphasis 
Added.]

The effect of this will be that Verisign is able to set pricing, causing a price increase for consumers.  Verisign's open admission that the high price is proposed by them for the purpose of erecting a barrier to entry into the market is, in our opinion, an impermissible abuse of monopoly power to narrow customer choice and fix prices in this market segment.

Further, Verisign, as a monopoly, should not be allowed to arbitrarily set the price of this or any other service derived from their monopoly position.  The price should be based on a "cost plus" fee arrangement, where the costs are examined by a regulatory agency.  As a monopoly service, wait listing should not be priced at what the market will bear. Verisign, in not submitting management of this service to bid, has further demonstrated that they are not interested in keeping costs low, but rather finding reasons to keep them high.  Even if we were to agree that a service 
such as this has value for consumers, we find the $40 per year per wait price to be outrageous.  If Verisign wishes to launch and outsource this service, they should be required to submit its management to competitive bidding.

We believe that the entire proposed Wait List System is not a natural monopoly and therefore need not be administered as if it was.  The registry 
function, which we all recognize as a natural monopoly and which is therefore highly regulated and priced on a "cost plus" basis, involves considerably more infrastructure than the proposed Wait List System.  The existence of a robust and thriving market segment for expired domains is ample evidence that a monopoly is neither necessary nor desirable.  At present, for example, customers seeking to acquire a domain might reasonably maximize their chances by purchasing the services of a number of 
competitors, buying from SnapNames, Dotster, and others at the same time. Under the proposed scheme, the customer would instead be expected to make an arbitrary choice among competitors, buying from one only, whose products 
are coerced to be undifferentiated other than by price.  The coercion of the market to a single undifferentiated product will force competition almost entirely on the basis of price, meaning that Verisign's wholesale price will fix the market.

At present, SnapNames is a major player in the secondary domain market segment.  If they acquire monopoly power through association with the Verisign registry, we believe that they must recuse themselves from competing in the direct retail sale of waits.  This presents a significant problem in that SnapNames has already sold waits under their business model 
and will have a strong incentive to dump them all into the new system as it 
is started up.  The commingling of monopoly and competitive roles clearly calls for independent oversight to prevent abuses of this kind from occurring, including what is commonly referred to as a "wall of isolation" or "clean room techniques" between the different parts of the company.

	Unlicensed Commodity Trading

We have considerable discomfort in the possibility, as we see it, that a wait on a domain, as distinct from a domain registration, may be held to be 
a security or commodity subject to careful rules and regulations, especially if a secondary market emerges, as seems inevitable, for the trading of waits themselves.  If we elect to participate in the sale of waits, we are concerned that we may thereby subject ourselves to unknown and potentially unlimited liability as unlicensed commodity traders.  For example, one can buy and sell corn and wheat with little concern for regulation, but one becomes exposed to extremely complicated and arcane regulations if one enters the completely different business of trading in futures for corn and wheat.  United States federal law defines a "commodity" subject to regulation as:

The term "commodity" means? all other goods and articles? and all services, rights, and interests in which contracts for future delivery are presently or in the future dealt in.
[Commodity Exchange Act, 7 USC 1(a)(4)]

This observation regarding the legal definition of a "commodity" leads, in turn, to the interesting observation that Verisign, through this proposal, may actually be backing the business model which is most likely to be held as unlawful trading in commodities.  That is, the SnapNames pricing and business model, which charges whether the wait eventually results in delivery of the domain or not, seems to us to be most problematic when analyzed in terms of unlicensed commodity trading.  By contrast, the Dotster model, which charges only if the domain sought can be registered, escapes this problem, and the BuyDomains model, which sells from an inventory of registered domains, completely avoids the issue.

If it is eventually determined as a matter of law that domain waits are commodities subject to regulation, we may all be forced to refund charges to customers whose waits did not result in acquisition of the domain sought.  This could prove to be a ticking financial time bomb.

Contract Terms

It is clear that the Registry-Registrar Agreement (RRA) is intended to be a 
complete agreement governing the entire relationship of signatory parties with regard to the COM, NET, and ORG generic top-level domains.  Any modification, addition, or amendment to the RRA requires, as we understand it, the approval of ICANN and possibly other oversight bodies.  We view the 
proposed Wait List System as a substantial modification of the RRA, and we therefore believe very strongly that the Verisign registry must seek the review and consent of such oversight bodies before implementing it.  If the 
Wait List System goes into effect, some registrars may lose the ability to acquire dropped domains in the manner in which they have been doing it so far, and may lose substantial revenue as a result.

Technical Issues

No atomic method exists to create a domain name and a wait for it.  It is critical that registrants be able to be assured a wait on their name should 
they choose to get it and this can only be facilitated by an atomic method of creating both.  (A process is said to be "atomic" in software if it cannot be split up. In this case, since there is a delay between when a domain name is registered and when a wait for the same domain can be purchased, it affords some opportunity for a party other than the registrant to seize a wait for the domain shortly after it is registered.)

Because Verisign is creating an incentive for registrars to never delete names, root server resource consumption is increased.  This represents at least to some extent a transfer of the resource problem from Verisign, which is a profit-making entity, to the volunteer corps of root server operators.

We note that an auto-renew flag on a wait can never be cleared.  If the subscriber changes their mind and does not want to renew, it makes sense that the flag should be able to be cleared.  If the sponsoring registrar for the wait cannot clear the flag, setting the auto-renew flag would seem to commit them to paying $40 per year in perpetuity.

The inability of a customer to transfer a purchased wait from one registrar 
to another will tend to facilitate abuses by unscrupulous registrars.  For example, a registrar might charge $35 to switch a wait already purchased from one domain to another, something that we understand would cost the registrar nothing, and the customer would be left with no choice but to buy 
a new wait from a different registrar for at least $40.  This is merely an example of abusive conduct, and we see no reason that an unscrupulous registrar would stop at this.  Once a wait is sold, a registrar has little incentive to provide customer service related to it.  Because we see waits as of little real value, and most likely to be sold to unsophisticated or naïve customers, this customer pool will be unusually vulnerable to such abuses.

Because a wait can be changed (with respect to what name it is waiting for) 
and because waits expire, we are creating an additional "pool" that registrars will connect to and issue repeated attempts to ADD.  This is exactly the same situation as the existing "batch" pool.  Instead of having 
one batch pool, we now have two.  However, since it appears not to be public information when a wait will expire, it will be necessary for those attempting to acquire expired waits to hit the wait pool 24 hours a day, 7 days a week, as hard as possible.  Of course, unwaited names will still need to be dropped in the batch pool.  As long as names are being dropped, registrars will aggressively compete for these names.  Therefore, no problem related to load or access is solved here.  In fact, an additional and probably worse problem is created in the form of the WLS pool.

Practical and Administrative Issues

The existence of a Wait List System managed by the registry will introduce a new incentive for registrars to avoid allowing valuable domains to expire, preferring to offer these domains in the secondary market themselves if registrants choose not to renew them.  This will tend to warp 
the market, which presently has efficient mechanisms for determining customer demand for a particular expired domain.  More seriously, the registrars who for historical reasons -- and in particular the Verisign registrar -- have the largest inventory of domains which are not being renewed by their registrants will acquire an unfair advantage because they will be able to hoard domains for resale.  In fact, the existence of a wait 
could be used as an indicator of customer demand for a particular domain, and therefore influence the decision not to release it into the pool of available domains.

Also, registrars will undoubtedly encounter difficulty from credit card processing companies.  We all depend upon processing credit cards to accept 
payment.  We can see a situation where a customer places a wait on a name 30 or 60 days before it expires. If the registrant renews the domain, the subscriber will call his or her credit card issuer and charge back the cost 
of the wait.  This can be done up to 120 days after the charge is applied. This costs the registrar not only whatever the wait fee charged was, but an 
additional charge back fee, usually about $35.  If a registar gets enough of these, they could lose their merchant account.  Defending a "charge back" for a wait would be very difficult, especially if the customer never got anything tangible for their money.

Conclusion

For the reasons stated above, Registration Technologies, Inc. strongly opposes this proposal.  Additionally, we feel that the secondary market is a legitimate and honest one that is being made a scapegoat, and that any future proposals recognize this fact.  The existence of a robust and competitive secondary market satisfies consumer demand, and it is desirable 
that consumers have the widest possible choices of business and pricing models in a fair marketplace.  It is important to remember that the most important people in this industry remain the consumers.  Our actions should 
protect them and therefore protect our own interests.

WLS Comments.pdf