On 17/09/2023 20:18, Antony Van Couvering wrote:
Domain Tasting stopped domain names that would have naturally gone through the deletion process becoming available for reregistration. That created an artificial demand for new gTLDs and drove a lot of the enthusiasm for the new gTLD round of 2012.
John,
While I agree with the rest of your excellent post, the idea that domain tasting was in any way responsible for the demand for new gTLDs is speculative at best and doesn’t at all accord with my own (extremely close) experience of the introduction of new gTLDs.
If you are saying that consumers were frustrated with the (lack of) choice in the .com space, and that (therefore?) consumers were pushing for new gTLDs, I couldn’t agree less. Consumers at the time had no idea about new gTLDs; my bet, and those of other business people at the time, was that upon the introduction of new gTLDs, consumers would wake up to the idea of choice once it was presented to them. Consumers didn’t have much idea of choice outside of local ccTLDs before the new gTLDs were introduced, and so they couldn’t have driven demand for new gTLDs.
The problem had a number of parts, Antony, Much of each days deletions were being hoovered up by tasters and many of the "valuable" domain names were shifted to sales and auction sites before they ever had a chance to drop. This created the infamous "all the good names are gone" feeling with prospective registrants. Some of the ccTLDs capitalised on this and gained registrations as a direct result. Tasting was creating an artificial shortage of good domain names that consumers could reregister. People wanted to be able to register the domain names that they wanted. The new gTLDs were not around then but the end of tasting changed the demand (along with ccTLDs modernising and meeting some of that demand). To some extent, the new gTLDs would have met that demand if large-scale Domain Tasting had still been a problem. When ICANN introduced the restocking fee and the key registrars where taken out of the market, many of the domain names that had been effectively hijacked were making their way back into the deletion process and on to the market. ICANN's action had unwittingly fixed part of the problem and in doing so turned a lot of prospective registrants of new gTLDs into registrants of legacy gTLDs.
Quite the contrary — we bet on consumers being interested in new gTLDs *despite* there being no direct evidence that they were. I often used the quote from Henry Ford — “If I asked my customers what they wanted, they’d all say they wanted a faster horse” — as an illustration of how we approached this possible new market.
Interest is one thing. Converting that interest into paying customers is another. Without the artificial shortage created by tasting, registries had to rely on the attractiveness of their gTLD and marketing. The latter caused major issues for some registries because they completely underestimated the money required to get their new gTLDs established in the market. It was as if some of the had adopted the "Field of Dreams" business plan and when they launched their gTLDs, the registrants didn't come. Some, like .WED, had business plans that were, to be diplomatic, novel.
Domain tasting could not have created an artificial (or any) demand for new gTLDs, because no-one outside of ICANN even knew about new gTLDs prior to their introduction. The long fight to get them introduced was ignored by the press, including the tech press, and if you recall correctly, ICANN ran exactly one advertisement to promote them. As I said at the time: “One World, One Internet, One Ad.”
That gets back to ICANN being a reactive organisation rather than proactive one. It is as if the real world part of the Internet doesn't even know that ICANN exists.
Of those that did know about new gTLDs, many, including ALAC, worked to quash them. The only ones promoting them were the new registries, and we were having to spend all of our time and money trying to win over the supposed champions of choice and competition.
If large-scale Domain Tasting had still been underway, that process would have been much easier because it was directly affecting the core legacy gTLDs (.COM/NET/ORG). I'm not sure that ALAC, at the time, appreciated the extent of what was happening.
So no, there was no consumer demand for new gTLDs prior to their introduction, whether artificial or otherwise, and certainly none created by domain tasting.
Some of the ccTLDs met that demand for domain names (not necessarily new gTLDs or even choice) and it kickstarted growth in some of the ccTLDs. What's happened as a result of this is that the new registration volume in the countries of these ccTLDs is often 2:1 in favour of the ccTLDs with the legacy gTLDs falling to replacement level. The Internet, while ICANN wasn't looking, became local. Domain Tasting created an artificial shortage of deleting "good" domain names. Once that was stopped, consumers were often able to register the deleted or non-tasted domain names they wanted in C/N/O. There was a rise in ccTLD registrations in parallel with the Domain Tasting period and that has continued. Many of the new gTLDs could have provided an alternative for that consumers during tasting. ICANN's wonderful economic report on new gTLDs did mention switching. I think that it underestimated the stickiness of the legacy gTLDs and the ccTLDs. People, registrants and users, recognise the legacy gTLDs and a few ccTLDs. That hasn't really changed except for geo gTLDs or niche gTLDs.
Maybe you have evidence that shows I’m wrong; I’d love to see it.
Verisign was not publishing full deletion data as the 2014 reporting rules had not happened. It only started publishing Add Grace Period deletions in April 2007. Month - .COM count - total new - deleted - AGP deletions 200704 | 62,364,687 | 1,961,272 | 111,324 | 42,476,307 This shows the damage that tasting was doing to the market: .COM - Count - Deletions - AGP Deletions June 2008 - 77,382,243 - 15,909,449 - 15,738,292 July 2008 - 77,191,881 - 2,617,167 - 2,483,953 The restocking fee policy came into effect on 01 April 2009. Month - .COM count - Deletions - AGP Deletions March 2009 - 82,229,830 - 2,871,076 - 2,721,859 April 2009 - 82,689,221 - 174,193 - 37,519 I'm not sure if prospective registries were compiling their own statistics on .COM/NET in the absence of deletion figures in the ICANN registar reports. (There was a lot of snakeoil floating about at the time.) Relying on these ICANN registrar reports without accurate deletion figures would have seen that the .COM was continuing, largely, to increase without undertanding what was happening. Without tasting, the new registration volume in .COM stabilised and new registration volume started to increase. April 2014 is when Verisign started using the new reporting format. (I extracted the data from the ICANN registrar reports going back to July 2001 including all those spreadsheets printed as PDFs.) Month - .COM count - New - Deleted - AGP Deletions | 200705 | 63,334,380 | 2,002,028 | 101,573 | 45,732,043 | | 200706 | 64,693,754 | 2,131,267 | 96,790 | 45,574,584 | | 200707 | 65,881,394 | 2,135,242 | 107,661 | 14,126,500 | | 200708 | 67,065,864 | 2,192,416 | 101,401 | 55,919,742 | | 200709 | 68,276,056 | 2,122,711 | 104,912 | 50,370,006 | | 200710 | 69,621,530 | 2,407,358 | 91,570 | 14,593,119 | | 200711 | 70,698,420 | 2,132,379 | 18,146 | 30,454,737 | | 200712 | 71,361,736 | 1,907,321 | 214,802 | 18,382,882 | | 200801 | 72,441,682 | 2,277,038 | 143,749 | 18,886,175 | | 200802 | 73,649,885 | 2,182,419 | 117,305 | 15,804,124 | | 200803 | 74,823,774 | 2,255,869 | 143,777 | 22,181,045 | | 200804 | 75,785,462 | 2,207,236 | 147,333 | 17,590,550 | | 200805 | 76,625,770 | 2,159,536 | 223,617 | 16,708,036 | | 200806 | 77,382,243 | 2,122,794 | 171,157 | 15,738,292 | | 200807 | 78,191,881 | 2,078,606 | 133,214 | 2,483,953 | | 200808 | 78,762,924 | 1,912,456 | 162,582 | 1,911,396 | | 200809 | 79,377,991 | 1,945,023 | 139,793 | 1,903,177 | | 200810 | 79,898,475 | 1,904,691 | 140,057 | 1,533,837 | | 200811 | 80,333,074 | 1,803,623 | 128,944 | 1,579,387 | | 200812 | 80,450,204 | 1,711,711 | 135,603 | 1,951,765 | | 200901 | 80,759,835 | 1,962,766 | 144,857 | 1,330,227 | | 200902 | 81,434,408 | 2,058,343 | 190,034 | 1,380,330 | | 200903 | 82,229,830 | 2,368,840 | 149,217 | 2,721,859 | | 200904 | 82,689,221 | 2,084,868 | 137,154 | 37,519 | | 200905 | 82,887,947 | 2,021,929 | 134,817 | 39,376 | | 200906 | 83,263,339 | 1,999,639 | 184,357 | 43,575 | | 200907 | 83,588,661 | 2,037,927 | 147,765 | 35,971 | | 200908 | 83,890,030 | 2,010,759 | 122,335 | 44,133 | | 200909 | 84,357,889 | 2,043,854 | 117,983 | 60,584 | | 200910 | 84,946,678 | 2,164,497 | 132,848 | 43,131 | | 200911 | 85,715,975 | 2,287,230 | 130,249 | 56,619 | | 200912 | 86,046,517 | 1,918,437 | 131,930 | 33,925 | | 201001 | 86,677,330 | 2,256,890 | 146,991 | 41,943 | | 201002 | 87,415,546 | 2,146,326 | 138,867 | 43,190 | | 201003 | 88,509,746 | 2,607,402 | 146,523 | 45,053 | | 201004 | 89,187,401 | 2,362,072 | 137,186 | 45,543 | | 201005 | 89,712,873 | 2,309,272 | 131,078 | 74,182 | | 201006 | 90,267,011 | 2,230,614 | 127,064 | 41,798 | | 201007 | 90,798,616 | 2,132,195 | 141,129 | 42,630 | | 201008 | 91,366,289 | 2,174,797 | 139,699 | 39,604 | | 201009 | 92,032,442 | 2,198,343 | 128,941 | 44,071 | | 201010 | 92,739,962 | 2,286,669 | 137,508 | 49,817 | | 201011 | 93,383,155 | 2,244,633 | 157,806 | 50,800 | | 201012 | 93,739,357 | 2,075,948 | 159,403 | 43,771 | | 201101 | 94,352,914 | 2,354,541 | 153,817 | 49,478 | | 201102 | 95,185,529 | 2,287,129 | 147,577 | 47,212 | | 201103 | 96,284,285 | 2,608,413 | 157,528 | 50,060 | | 201104 | 96,926,465 | 2,343,558 | 133,801 | 43,812 | | 201105 | 97,395,538 | 2,358,225 | 129,604 | 47,350 | | 201106 | 97,968,486 | 2,410,783 | 128,739 | 90,091 | | 201107 | 98,569,830 | 2,274,808 | 121,996 | 71,566 | | 201108 | 99,118,230 | 2,287,194 | 135,545 | 52,960 | | 201109 | 99,850,728 | 2,380,479 | 110,261 | 57,761 | | 201110 | 100,540,971 | 2,376,910 | 82,631 | 53,769 | | 201111 | 101,161,004 | 2,318,165 | 79,464 | 58,196 | | 201112 | 101,621,129 | 2,223,835 | 97,805 | 61,887 | | 201201 | 102,220,788 | 2,467,165 | 132,359 | 70,383 | | 201202 | 103,224,211 | 2,617,244 | 109,609 | 68,933 | | 201203 | 104,350,457 | 2,765,631 | 108,098 | 75,141 | | 201204 | 105,009,536 | 2,537,906 | 78,707 | 64,112 | | 201205 | 105,601,144 | 2,582,657 | 83,195 | 66,837 | | 201206 | 106,003,864 | 2,331,031 | 81,308 | 59,499 | | 201207 | 106,400,013 | 2,330,066 | 75,804 | 66,652 | | 201208 | 106,804,483 | 2,270,299 | 77,283 | 59,515 | | 201209 | 107,212,100 | 2,299,212 | 85,547 | 59,392 | | 201210 | 107,717,350 | 2,385,091 | 92,124 | 57,059 | | 201211 | 108,205,473 | 2,368,644 | 91,656 | 74,490 | | 201212 | 108,505,915 | 2,284,710 | 113,173 | 108,494 | | 201301 | 109,149,651 | 2,640,699 | 131,790 | 92,938 | | 201302 | 109,738,139 | 2,357,263 | 205,577 | 82,311 | | 201303 | 110,585,571 | 2,785,307 | 110,729 | 106,864 | | 201304 | 111,163,489 | 2,671,414 | 90,248 | 126,674 | | 201305 | 111,476,933 | 2,639,003 | 91,374 | 108,202 | | 201306 | 111,719,747 | 2,381,716 | 83,833 | 88,655 | | 201307 | 112,071,447 | 2,518,917 | 74,955 | 107,483 | | 201308 | 112,533,569 | 2,486,559 | 76,994 | 108,041 | | 201309 | 113,008,994 | 2,386,879 | 73,217 | 90,505 | | 201310 | 113,593,013 | 2,532,606 | 73,589 | 101,676 | | 201311 | 114,076,050 | 2,441,576 | 92,269 | 116,373 | | 201312 | 114,359,327 | 2,313,129 | 77,226 | 114,845 | | 201401 | 114,765,220 | 2,456,962 | 77,822 | 79,213 | | 201402 | 115,710,951 | 2,424,094 | 76,014 | 56,157 | | 201403 | 116,480,005 | 2,805,746 | 108,941 | 68,959 | | 201404 | 116,856,602 | 2,584,718 | 2,439,119 | 62,626 | Some of this .COM growth is due to broadband and infrastructure improving in countries. This amplified the switching cost because .COM was known and the new gTLDs were not. It would probably have been beyond the financial abilities of most new gTLD registries to target all countries equally. I still can't understand how ICANN expected there to be 33 million new gTLD registrations in the first year of their operation. That .COM new registrations volume is what the new gTLD registries were facing when trying to get their gTLDs established. With Domain Tasting creating an artificial shortage of "good" domain names, there was a potential market for some new gTLDs. Without that artificial shortage and the pent-up demand that it created, new gTLD registries had to rely on their own marketing and publicity to drive registrations. The problem was that the market, since Domain Tasting, had changed and a large part of it outside of the US had become ccTLD focused. The new gTLDs that successfully gained registrations were actually ones that were targeting communities. The pricing of some of the new gTLDs also had an effect. Some of the legacy gTLDs are pricing themselves out of the developing markets as an alternative to the local ccTLDs and .COM/NET. But that's a completely different issue. 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