My Comments to ICANN Opposing the 2024 .COM Renewal

The public comment period regarding the .COM renewal ends today (November 5, 2024). ICANN routinely ignores public input, and I expect that will continue with this comment period.

Regardless, I’ve submitted a comment opposing to the .COM renewal, in order to be on the record. You can also read it here (PDF).

NB: After reviewing comments by others, I decided to supplement my submission with an additional statement addressing those “off topic” comments. It’s included below:

We’d like to also address some of the off-topic comments relating to transaction prices for the transfers of ownership of domain names, that attempt to distract from Verisign’s abusive monopoly for .com registry services.

Verisign charges fees for registry services (i.e. managing a central database, the zone file, the nameservers that spit back the nameservers for the various domain names in that registry). They have managed those services without having to compete for the contract by those who wish to replace them. The market value of those registry services is below $1/yr/domain (e.g. 70 cents per domain per year for the .IN ccTLD, as per their tender won by Neustar).

http://domainincite.com/23976-neustar-completes-in-migration

which is more than 90% less than what Verisign is charging as its fees to registrars (fees which are ultimately paid by consumers, namely domain name owners).

The so-called “secondary market” or “aftermarket” is a marketplace for the asset value of the domain names themselves, which is an entirely different market than that for registry services.

To understand this distinction, consider a TM registry, like the USPTO, which has a fee structure for the services they provide for the registration and maintenance of TMs. Those fees are entirely unrelated to the value of the TM itself. A TM owned by Google for “GOOGLE” or by Nike for “JUST DO IT” has set fees, and Google or Nike are free to sell, assign, license, etc. those trademarks to others at whatever the market will bear. Companies like Hilco Streambank routinely auction off the intellectual property of companies, including their trademarks, see:

https://www.hilcostreambank.com/closed-deals

and of course, owners of TMs do these kinds of transactions all the time. For example, Hooters sold their TM for $60 million.

https://www.bizjournals.com/tampabay/stories/2001/04/09/story1.html

The same comparison exists for a land registry and houses. Or a copyright registry, and the underlying copyrighted works. The story of the Beatles catalog makes interesting reading:

https://www.billboard.com/articles/columns/rock/7662519/beatles-catalog-paul-mccartney-brief-history-ownership

and that catalog only has value due to the copyrights (once those copyrights expire, the works fall into the public domain, and the works are free for everyone to use). The cost for JK Rowling to register the copyrights for her valuable Harry Potter novels is the same as the cost for a random author to register their random novel. Ms. Rowling is free to assign those copyrights to others at the full value of the works, which are unrelated to the fees paid to register the works.

Verisign and its allies (some of whom do not disclose their association with Verisign) attempt to distract from their abusive monopoly by pointing to the healthy market for ownership transfers of certain domain names, to suggest that “something is wrong” that needs to be addressed. This could not be further from the truth. Ownership transfers of already-registered domain names operate in a highly competitive market that is textbook capitalism. No one is concerned if the Lennon estate transfers ownership of John Lennon’s music at market prices (i.e. at the value of the music itself, rather than at the fees that were paid to register their copyrights in a central database). No one is concerned when land changes hands at market prices between a willing buyer and a willing seller — that market price is unrelated to the costs of the land registry that records the ownership records.

This is completely different than Verisign’s abusive monopoly, which is crony capitalism. They are simply operators of a central database, and a set of nameservers that resolve domain names. The cost of managing the database entries for “voice.com” or “Google.com” or “Apple.com” is the exact same amount as the cost of managing the database entries for “jkhjkhjkkjhkhgkjh.com” or “19097198ghgkjhg.com” (2 random domain names with little value for ownership transfers).

Verisign would have you believe that it is “special”, and that it can do things that nobody else can do. This is entirely false. The two main activities it conducts are the management of a central database, and the resolution of domain names via nameservers. Let’s take these one at a time:

First, its database is extremely small. The zone file (the list of all registered domain names) for .com is can be downloaded by the public, and is approximately 3.9 Gigabytes compressed (22.6 Gigabytes uncompressed). In other words, it can fit into an iPhone with ease, or even a $5 USB stick. It is much smaller than databases operated by CERN for particle physics data, by VISA for e-commerce transactions, by NASA for astronomy data, or numerous other entities that manage commercial databases. The Blockchain data for Bitcoin is over 600 GB:

https://ycharts.com/indicators/bitcoin_blockchain_size

Secondly, many other companies operate nameservers at scale, such as Google, Akamai, GoDaddy, Amazon and Cloudflare. They face all the same security issues that Verisign does, and operate without hiccups. Indeed, the root zone is handled by a group of different organizations,

https://en.wikipedia.org/wiki/Root_name_server

which suggests that ICANN could separate name resolution from the central database management, for even greater resilience of .com. There is no need for Verisign (or a replacement contractor) to operate all the nameservers for .com as well as the central database itself. Those 2 distinct functions can be assigned to different entities, or even groups of entities, as per the root zone itself.

It’s worth reminding ourselves that Verisign doesn’t own the .com TLD. They wish they did, and they often act like they do (especially around uninformed politicians or regulators), but they are simply contractors with a specific task. The owners of a 500 unit condo building might hire a property manager (a contractor) to offer concierge services, handle repairs, and manage the keys to the individual units. But, that doesn’t make the property manager equivalent or equal to the condo unit owners. The condo owners alone benefit from the ups and downs of real estate values. The property manager can be swapped at will, if a different property manager can offer concierge services, handle repairs, or manage keys at a more competitive fee to the condo owners. Concierge services, handling repairs, and managing keys are akin to the “registry services” for the dot-com TLD, whereas domain name registrants are akin to the condo owners. While property managers might get jealous that condo prices skyrocket for desirable units in good locations, ultimately they are not the owners of those units and those changes in market values are irrelevant to the cost of providing their concierge services, handling repairs, or managing keys (which are services that many other prospective property management companies can provide).

A property manager that held condo units “hostage”, attempting to prevent replacement property managers from competing for the management contract, would find itself in court by authorities and/or property owners, due to its anti-competitive and abusive behaviour. One wonders why ICANN and/or national antitrust agencies to this day have failed to scrutinize Verisign’s abusive and anti-competitive behaviour, in relation to dot-com. Perhaps one day they will.

In conclusion, don’t be distracted by those attempting to compare apples with oranges. The market for registry services (i.e. management of a central database and resolution of names via nameservers) is entirely different than ownership transfers of the domain names themselves. Just like the markets for land registries, trademark databases and song copyright databases (all of which are specific databases) are entirely different than the markets for the land itself, the underlying trademarks, or the songs.