On 12/01/2026 12:02, gopal wrote:
John,
Thank you again.
During the CPWG Call on 7 January 2026, my team gave me an opportunity to speak on the ICANN SWORD Algorithm.
The justifiability of the ICANN SWORD algorithm is a subject of debate, with arguments focusing on its purpose of providing an objective process versus significant concerns regarding its proprietary nature, transparency, and perceived lack of flexibility in specific cases.
I suggest the usage of this algorithm due to the following reasons:
#1. number of applications anticipated #2. Most countries in the region host more of their gTLDs outside their country's Internet infrastructure [as asserted by you in the trace] #3. Automated support for safety and stability of the Internet.
ICANN's SWORD algorithm is a good approach, Gopal, The adoption of IDNs is closely linked to the ccTLDs rather than the gTLDs. The reason might be more to do with the users and registrants rather than the TLDs themselves. An IDN is, at their most fundamental level, a linguistic expression. That tends to be concentrated geographically or on the basis of languages of the script users. The gTLDs might not be the best TLDs for that unless they are specifically targeted at the market. In a global TLD, those IDNs fade into the background noise. The state of readiness of registrars and service providers (DNS, e-mail and web) is an important factor. (That gets back to the UA problem.) With a ccTLD, they are more visible because the people using the ccTLD are already thinking in terms of the IDN scripts. If people consider a TLD to be *their* TLD, then they will use it more and IDNs would be much more useful in such a TLD because the web developers and resellers would be familiar with the scripts. (The ICANN Domain Name Marketplace group covered some of this but it focused on the number of languages available on registrar sites.) The Local/External hosting issue is a complex one. There are three main reasons for hosting outside a country's IP infrastructire. The first is the development level of the Internet infrastructure. The more data centres, the more websites can easily be hosted in the country. The second is the "age" or development phase of the market in the country. This is slightly different to the first point as it is more focused on the digital economy of the country whereas the first point was focused on the hardware and connectiviy. Data centres rarely appear overnight and there has to be a certain level of demand to make them economically viable. The early phase of a market sees the local telco and ISP companies providing web hosting as an add-on service for their customers. The next stage sees web developers and dedicated web hosting providers appearing. Then, when there is sufficient demand, the data centres start appearing. If there is a large market, Cloud hosting providers will also appear in the market. The third is pricing. It is often cheaper to host externally than host locally in the early phase of a market. The hardware requires investment and bandwidth and traffic costs can be more expensive. The costs for larger web hosting providers and Cloud hosters in other countries might be lower (economy of scale) and thus the hosting costs might be lower than those for hosting in the country. With a developing market, costs are even more important than they would be in a developed market with a lot of competition. (The economists on the list can probably explain that far better than I could.) There is a kind of segmentation of the market that occurs as it develops. A lot of the e-commerce might be externally focused and people want to sell products and services to countries where they can make a greater profit. As the local market grows, the e-commerce shifts to people in a country selling more to each other than externally. That's the typical evolution. The rise of DIY website services such as WIX makes it a lot easier for people to build their websites on these services and that means that those DIY website services will take a share of each country's market. There are also transnational providers who either develop businesses in various countries or acquire players in various countries. Godaddy, Newfold Digital and Team Blue would be good examples of these operators. It is not unusual to see some of the top ten providers in various countries owned by the same operator but with separate brands. At the upper level of the gTLD market, approximately 95% of the gTLD registrations are concentrated on approximately 7,500 hosting brands (including registrars and off-registrar hosting). The off-registrar domain name hosting is approximatley 26%. It is more complicated in terms of website hosting. Some of the larger providers use Cloud hosting for their services but have a distinct brand identity. An end user or regisrant may be completely unaware that their website or domain name is CLoud hosted because the branding remains constant. Some smaller providers only exist within Cloud hosting and do not have their own distinct DNS. Some of the figures that I've seen quoted for the market share of various Cloud operations are either wrong (based on extremely limited data) or simplistic (by people who don't understand the complexity of usage). That Verisign short selling incident late last year where short sellers decided that the end of the Google Adsense for Domains program would result in a big hit on the .COM/NET registration volume is a good example. Verisgn's own figures (which were more reliable) stated that approximately 2% of its domain names were registered purely to make money from PPC parking (to make money from advertising). The others with PPC adverts were largely domain names that had been parked by the registrars and had no developed websites yet.
Comments from the members of this mailing list as most welcome.
On 7 January 2026, I opted to use excerpts from just a couple of disputes on String Similarity from a collection of 69 disputed cases with another one very specific to Tamil script. I stopped collecting more case studies.
Also, I found a draft report "Policy, Business, Technical and Operational Considerations for the Management of a country code Top Level Domain (ccTLD)" from the ITU during April 2008. But the present focus is on gTLD.
It might be interesting to see how the ITU and its members (mainly telcos) thought that the market would develop. The gTLD domain name market at the time was being hammered by Domain Tasting (some registrars exploited the 5 day Add Grace Period to test newly deleted domain names with PPC ads and kept those thrat made money and deleted the rest.) and that created an artificial shortage of "good" domain names that should have been deleted and been available for re-registration. That, in turn, kickstarted the 2012 round of new gTLDs. The problem was that beween 2009 and 2012, the large-scale Domain Tasting problem was fixed and the demand for some new gTLDs had collapsed (the artificial shortage created by Domain Tasting had disappeared) before the new gTLDs even launched. Regards...jmcc -- ********************************************************** John McCormac * e-mail: jmcc@hosterstats.com MC2 * web: http://www.hosterstats.com/ 22 Viewmount * Domain Registrations Statistics Waterford * Domnomics - the business of domain names Ireland * https://amzn.to/2OPtEIO IE * Skype: hosterstats.com ********************************************************** -- This email has been checked for viruses by Avast antivirus software. www.avast.com