[ga] Paper on the Release of Deleted Domains

["Richard Henderson" <richardhenderson@xxxxxxxxxxxx>]

Copy of Paper sent to Bhavin Turakhia and Tim Cole for presentation/discussion at Capr Town meeting:

Dear Bhavin and Tim,

I submit the following paper (also posted at www.atlarge.org) on the release of deleted domains. I would appreciate it if you could distribute this paper and allow for its presentation at ICANN's Capetown meeting.

I have been an active member of the At Large for several years and have contributed in detail on the development and evaluation of New gTLDs. It would be advantageous if the issue of releasing deleted domains could be clarified prior to the Agreements being created for the next round of gTLDs.

I also have a good first-hand understanding of the Deleting Domain market and processes, having monitored for a number of years.

In my presentation, I review the status quo and offer a number of possible solutions.

With kind regards,

Richard Henderson

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

PAPER ON THE RELEASE OF DELETED DOMAINS (for ICANN's Meeting in Cape Town):

"PROCESSES FOR FAIR ACCESS TO DELETING DOMAINS AND A FAIRER MARKET FOR ESTABLISHED REGISTRARS"



For some time, the market for selling deleted domains has been substantially captured by secondary agents who use selected or created registrars to increase their access to the Registries at the time names are released. This has arguably handicapped the rights and opportunities of people to a fair and open process, using the registrar of their own choice. In the past year this process has resulted in the creation of approximately 100 "shell" registrars set up solely for the purpose of increasing access to Registries for their parent company. This, in turn, may have contributed to the decision of certain bona fide registrars to take matters into their own hands and develop their own policies for the re-distribution of domain names which were previously registered with them.

The leading "catcher" of deleting domains used to be Snapnames. They succeeded by using a fairly large number of registrars to run scripts and snap desirable deleting names. Snapnames got their cut and the Registrar got their cut. But they succeeded because statistically if you have the largest number of registrars running scripts for a deleting domain name, then you get the reputation as the best company to pay into to get the name you wanted.

Then POOL.com emerged, with what was arguably a better business model than Snapnames because you didn't have to pay anything up front - only if you got the name (provided you could outbid any other people who also wanted the name). To succeed in this business, POOL needed to have more registrars working for them than Snapnames had working for them.

Initially Namescout was used as a proactive registrar and "holding warehouse" for these incoming names. POOL realised they needed more registrars to run their scripts for deleting names, so what you got was a proliferation of new registrars who were set up with the intention of running scripts for deleting domains. The more registrars, the more chances of getting names, so just create scores of new registrars for your company!

In short, I imagine most of these new "Registrars" are just the same registrar, operating under a different name, but getting extra privileges and extra access to the Registries to obtain deleting domains.

ICANN seems to have unquestioningly accredited all these "registrars". But then, other registrars have operated very similarly before, like Domain Bank which seemed to set up Domain Pro as an extra "registrar" so they could have a second "short list" to submit to the .info Landrush 2 and the .biz launch. Both DomainBank and DomainPro operated from the same offices with the same phone numbers and the same accounting outlet, but they "counted" as 2 registrars, so they got twice as many lists into these newly launched TLD round robins when they launched. Just think what fun and games the 98 "Namescout/POOL" registrars will have if there are more round-robins for .eu or .travel in the future!

In its MoU with the Department of Commerce, ICANN is obliged to ensure the fair distribution of domain names to the internet public. I wish to offer a number of Proposals which may make the market in Deleting Domain fairer both for the public and for mainstream registrars who help maintain the smooth running of the DNS, and do not merely exist to exploit registrar privilege and access to Registries and capitalise on the deleting domains market.

PROPOSAL 1: FAIRER AND CLEARER INFORMATION FOR THE INTERNET PUBLIC.

At present, the deleting domains market is dominated by a fairly small number of buyers who have understood the processes and cornered the market for premium deleting domains. The ordinary member of the public, or average company, may well be unaware of the names which are "On Hold", then "Expired", then in "Redemption", then "Pending Delete". I propose that Registries should be obliged to publish lists of ALL domains which have expired, reached redemption, and finally are pending delete (with clear date provided for when they will be deleted). These lists should be published and accessible on their Registry websites so that ordinary internet users do not have to scrabble around, scouring DeletedDomains.com or other sites to glean their needed information. There seems to me to be no good reason why we cannot "open up" this part of the information - at Registry level - so that the whole world can easily see which domains are coming up for deletion. What possible reason can there be for not creating this open public access?

PROPOSAL 2: REGISTRY AUCTION OR ROUND ROBIN MODELS FOR DELETING DOMAINS

The round-robin method used at the time of the .info Landrush and the .biz2B demonstrate that it is possible to operate a process where names which are becoming available can be distributed *through registrar applications* in a manner which might more fairly reflect the market reality of existing registrars and their relative size and contribution to the DNS industry.

(i) Release of Deleting Domains through a Registry-level "Round Robin" model: coupled with the open lists from Registries proposed in PROPOSAL 1, deleting domains could be released in monthly "batches" using a Round-Robin model. This would enable registrants to apply through their regular registrars, rather than getting a released domain through an unknown entity with the problems that may occur. It would also entirely by-pass companies like POOL.com and Snapnames, whose services (in my view) are not really needed and tend to distort the market in these names. Certain problems would need to be resolved if the Round-Robin model was to work in a manner which was helpful to registrants and fair to Registrars... In the Landrush 'Round-Robins' it became clear that the process could be "gamed" by some registrars (like Signature Domains and DomainPro) submitting unusually short lists and effectively queue-jumping those registrars who opened their lists to all the public. To overcome this, I propose the insertion of "blanks" in each Registrar's list to top up their actual applications, so that all registrars have a statistically identical chance of procuring a particular deleting domain. The market would then reward registrars in proportionate relation to the number of customers they can attract.

(ii)Release of Deleting Domains through a Registry-level "Auction" model: a possibly simpler model would be an Auction of all deleting domains at Registry level, coupled to the advance posting of names by the Registry (proposed in PROPOSAL 1). To protect the interests of Registrars, each bidder would be required to nominate a Registrar of choice, to whom payment would be made and through whom the name would be registered, in the event of winning the auction for that name. The impact of this process would be to exclude the secondary market profit made by companies like Pool.com and Snapnames, and reward Registrars and Registry directly. It would also redress the balance in favour of regular and well-established Registrars, because people would tend to nominate a Registrar they knew, trusted, and possibly already used. It would render the 100 or more "shell" registrars superfluous, because why would people want to nominate these completely unknown entities? The auction model might be set up on a name by name basis, as names came up for deletion; or the deleting domains could be batched on a weekly or monthly basis. Once again, the whole process should be based on open advance access to information, and open access to the process of obtaining a domain name.

PROPOSAL 3: ADOPTING THE 'TUCOWS' MODEL FOR INCLUDING PREVIOUS REGISTRANTS IN THE PROFITS FROM DELETED DOMAIN SALES.

I am by no means sure about this third proposal, but I offer it for discussion and consideration. Under the plan recently floated by Tucows, domain names due to expire/delete would be offered for auction prior to this process, and a share of any profit would be distributed to the previous registrant. This model could also work if the Auction was operated at the Registry level that I propose in PROPOSAL 2. At present the DNS is hindered by the large number of domains that get "warehoused" by mass-registrants. I believe that a model that offered a proportionate share of the profits from the deleting domains market to current registrants might have the effect of freeing up the supply and creating more fluidty in the domain market (so that a higher proportion of domains might actually get used). I am not wholly convinced by this proposal because it may add complexity to a process which should ideally be simple.

I thank you for reading, and perhaps discussing, this Paper. I particularly commend PROPOSAL 1 and PROPOSAL 2(b) to you. In all my comments, my uppermost thought is what will make the supply of the DNS most open and fair for the people who actually want to use it. In addition, I recognise the role and livelihoods of established registrars (who provide a wide range of services beyond just grabbing deleting domains) and who deserve a process which reflects their commitment to the wider DNS market and its services. In achieving this, I advocate the removal of secondary businesses like Snapnames and Pool from the process, and the disempowering of the scores of "shell" registrars which have recently distorted this part of the market (and may also create havoc when .eu and .travel finally come onto the market with round-robins etc for the launch of their names). By using a Registry auction, where bidders nominate their regular Registrars, there will be little further need for these recently created registrars which have merely come onto the scene to 'game' the system.

With kind regards,
Richard Henderson
www.AtLarge.org
richardhenderson (at) ntlworld.com