Nominet announces new policy consultation for expiring .UK domains

The rush to introduce .uk seems to have been heavily influenced by the new gTLD mania that was permeating ICANN at that time. Unfortunately, that was driven by people who did not realise that it was the Domain Tasting of 2005-2008 that had caused massive problems for would-be gTLD registrants. ICANN just sat on their hands chewing the cud. It took external action by Dell and Google to stop that problem. However, new gTLDs were being pushed as a solution to the problem of "all the good names are taken" even though that ceased to be a major issue in 2009.

What is happening with .uk is a classic Junk Dump. This is where domain names registered during land rush periods that were not flipped, developed or needed are dropped. In some respects the .uk demand was artificially inflated by some large registrars. Without these players, the .uk would have had a much less impressive launch. The .co.uk dominates the UK TLD market. Much of the .COM activity is legacy (old registrations being renewed) and the UK market changed to being ccTLD dominated years ago. The decision to introduce .uk might have made sense in 2008 or so but the market had changed by 2012 and the 2020 market is vastly different to both. Most of the new gTLDs are struggling. The large new gTLDs (>1M regs) are bubble TLDs that have been inflated by heavily discounted registrations. Web usage is low compared to usage of .UK domain names. The interesting thing was that the web usage of .uk domain names was also low and many just redirected to the registrant's .co.uk or .com site. The problem with low demand TLDs or subdomains is that the renewal rates are much lower than expected and many of these domain names are never reregistered. The high value keyword domain names will be dropcaught and reregistered but it may take some time for the .uk to achieve the numbers of 2019. I'm not sure that Nominet's moves on expiring domain names will play out as it expects.

Regards...jmcc

 

After peaking at over 3,600,000 .uk domains after the mass-registrations by 123Reg, Namesco, Ionos1&1, and Fasthosts, it dropped by 1,200,000 by the time of Namesco's 'dump' in January this year, and will drop by a further 1,000,000 at least (and probably 1,200,000) when Fasthosts and Ionos dump their own mass-registrations next month, leaving .uk registrations at 1,200,000 or 1,300,000.

In other words, by facilitating these mass-registrations with 'ghost' registrants through their free registrations, Nominet created a fake peak but 2,300,000 of those were exactly that - fake.

The large registrars circumvented the rules requiring registrants to *request* names before registrations, but the close relationship of Namesco, GoDaddy (and their 123-reg) etc with Nominet seems to result in them doing what they like over things like this. Nominet's official attitude was we leave these companies to operate how they see fit.

'Laissez-faire'. From Nominet's position it was a misjudged, ineffective and desperate ploy to try to rescue .uk from its dismal and poor take-up. People just did not want .uk domains in millions of cases, as these vast name dumps demonstrate.

 

The dump seems to be in progress. This is the summary for the top of the .UK as of 01 August 2020. The Registrar/Registrars refers to ICANN registrars under control of the various operators.

Hoster / Group - .UK - New - Deleted
| Godaddy / Host Europe Group (Registrars) Portfolio Operator | 2,028,826 | 27,341 | 9,217 |
| United Internet (Registrars) Portfolio Operator | 1,602,046 | 18,751 | 653,533 |
| Godaddy US (Registrars) Portfolio Operator | 1,227,097 | 47,188 | 21,946 |
| Team Blue BE (Registrars) Portfolio Operator | 529,061 | 9,336 | 9,533 |
| Wix.com US (Registrar) | 285,337 | 7,837 | 4,914 |
| Cloudflare.com US (Registrar) (DDos mitigation/CDN) | 269,478 | 5,494 | 3,568 |
| dan.com (auction/sales) | 153,458 | 12,649 | 5,035 |
| Endurance International Group US (Registrars) Portfolio Operator | 144,569 | 2,377 | 6,143 |
| Web.com US (Registrars) Portfolio Operator | 143,048 | 2,449 | 5,342 |
| NameCheap, Inc. US (Registrar) | 132,092 | 5,944 | 4,707 |

This level of concentration of the large portfolio operators and their dominance is common across most of the gTLDs and ccTLDs. These operators have been on a buying spree for the last five years or so and many of the top tier hosters have been acquired. Even the "top registrars" thing with ICANN no longer means that the top registars are distinct registrars as it is common for some operators to own multiple registrars and brands. There are approximately 2,500 ICANN registrars but a significant number of them are being used simply for dropcatching. Dropcatch.com alone has a few hundred registrars.

It resulted in a Potemkin village of a subdomain. They were valid registrations but the domain names were either inactive or parked on holding pages. As had been seen with .XYZ and what followed in the new gTLDs, zone stuffing is a numbers game. The .XYZ registry got lucky. The normal renewal rate on a heavily discounted/free promotion is less around 5%. The registrars that .XYZ had chosen were registrars with a bluechip clientbase. That meant that some of these customers would not worry about automatically renewing domain names. The ccTLDs also tend to have a higher first year renewal rate (>=70% compared to .COM's 57%) so there is a chance that the renewal rate for these free .uk registrations might be higher than 5%.

I'm not sure of the politics of the situation with Nominet but the whole thing was badly thought out and contradictory signals were being given by Nominet over the years. Initially, it seemed that the .uk subdomain was to be a kind of expensive UK-only option. That was fine. The numbers for those kinds of TLDs or subdomains are always small but they have very strong renewal rates because of the expense of the registrations. The problem seemed to be a lack of demand. That was followed by changes about the nexus requirements over who could register, the necessity of a UK address, the pricing changes and then the RoR mess. What may not have been apparent was that some of the registration as a service operators were giving old information to their customers (registrars and hosters that use them to register domain names in ccTLDs or gTLDs rather than go though the process of becoming accredited registrars) about elegibility.

Each new TLD or subdomain has a landrush curve where new registrations spike for the first six months or so before falling to a normal monthly registration rate. That trend is seasonally affected (increased registrations on the run-up to Christmas and a decline for the Summer months) but stable unless the registry goes off the deep end with discounting. The RoR created what was, in effect, a second landrush which is being followed by the Junk Dump. The echo of the .EU landrush and Junk Dumps were visible in the graphs for over ten years after the launch of the ccTLD. Eurid eventually stopped publishing monthly statistics. (It has largely become a truckstop TLD where people go before going on elsewhere and natural in-TLD development is low compared to real ccTLDs.)

Perhaps the people in Nominet were unaware of all this as some of this only played out after the decision to launch .uk had been taken. In terms of .COM being a viable competitor to the .UK or, more specifically, .CO.UK, that's a gross simplification. The legacy gTLDs (.COM/NET/ORG/BIZ/INFO) have been sidelined in most ccTLD positive markets. The new registrations volume of the local ccTLDs is higher than the combined new registrations of the legacy gTLDs. The .BIZ is effectively dead in Europe and the decision by Afilias to use a Boom and Bust model for .INFO has made it largely irrelevant. The historical registrations (often brand protection) in some of these gTLDs keep renewing.

The Boom and Bust model is a cynical numbers game. Basically, the registry or the registrar runs a heavily discounted or free registrations promotion. Most of these registrations will not renew but a small percentage will. If the promotion is run over a long enough timeframe, the number of renewed registations will increase. Most of my work deals with gTLDs and web usage but I also track some ccTLDs and run periodic web usage surveys on them. With .uk, Nominet seemed to have been muddling along from one decision to another. In the background, there were some mad expectations that the new gTLDs would be a major threat to the established legacy gTLD/ccTLD order. That may have influenced Nominet's thinking with .uk as almost everyone expected the new gTLDs to do better than they did. The market had changed by the time the new gTLDs had launched and .uk seemed to be intended to compete with a threat that never materialised. The .LONDON, .SCOT, .WALES, .CYMRU new gTLDs never really gained momentum. Between the launch of .uk and the RoR, the domain name market had completely changed. Nominet and .uk, were victims of changed circumstances. But at least .uk has not ended up like .LOAN or some of the other new gTLDs.

Regards...jmcc

 

Tomorrow is the last day to make your views known on the Expiring Domains Consultation. Click here if you want to add your comments and have not already done so.

The consultation presents just the Auction and ECA models as 'fait accomplis' and when asked to choose, those are the only options offered. However, what I did was decline to choose any of their options, and made clear in the comment box provided that I believed the majority of correspondents would actually have chosen an Option C - other models or ways forward, had that been available.

It is interesting that at the roundtable (which I was unable to attend as I was nursing that day) apparently only 55 people by then had taken part in the consultation. In contrast, 101 people have signed Andrew Bennett's petition, wanting some kind of 3rd option, which wasn't offered in the consultation. Nominet may report on the consultation, and announce 51% voted for auctions, 49% voted for ECA. Whereas I suspect if people had not just lost the will to engage, the reality would have been that maybe 75% actually wanted an Option C.

Anyway, just over 24 hours left to add a contribution.

Feel free to copy and paste any parts of my reply, if anyone wants to, extracts follow:

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


Do you have any comments on the options for highly desired expiring domains?


I am not selecting any option because I believe there should have been an Option 3 - 'Other proposals and ways forward'. I believe a majority of correspondents to this consultation would have selected an Option 3. The consultation needs to acknowledge that in its statistics, and you have not made that possible by limiting selection to 1(a) 1(b) and 2. Your consultation loses credibility with valued and experienced Nominet members as a result - for example the not insignificant signatories of the recent petition.

Surely, Nominet should be approaching this consultation without preconceptions or minds made up - otherwise what is the point of involving members and asking for their opinions?

Two factors this year have made the whole issue more fraught and challenging.

1. The domination of certain parties, catching a disproportionate % of most valuable domains, in the mass dump of Namesco's mass-registrations in January and ROR the previous July. This seemed to demonstrate that Nominet was unable to enforce respect for its own principles and processes.

2. A month later Covid-19 had changed the whole way we would have to live our lives for the foreseeable future. In the context of pandemic and economic crisis, the proposal to radically change the way domains could be obtained when they dropped - which could deeply impact on the economic models and businesses many members relied on - could not have come at a worse time.

I've looked at many routes forward. But I've also listened on forums to people's fears for their families. Now is just not the right time. If you analyse Andrew Bennett's petition (which I signed for the reasons I'm outlining here) these were not a crowd of malcontents. There were many well-respected Nominet members (over 100 have signed) and some very significant businesses represented.

To present the consultation question as "my way", "my other way", or the highway... is disrespectful. Or perhaps it is just sloppy communication. To date, it seems to me that Andrew's petition has been marginalised. It should be welcomed by Nominet, and discussed at AGM, as a significant statement by significant and valued Nominet members, who are not comfortable with this seeming short-listed binary of options. It should also be formally factored in to your report on the whole consultation process.

I believe that - given the pandemic and the harsh economic crisis - it would be respectful to pause for a year or 24 months, and afford time for exploring and testing out other possibilities, and also, to afford people the basic dignity and respect of working out how to re-model their family and business finances if they have to. Right now is just not the right time.

I recognise that this process has already been extended once, but that was before we realised the massive hit our economies were going to take with the pandemic. I believe it should be formally extended, with the interim period openly exploring the ways of making possible option 3's work. And also to give small businesses breathing space.

Events have changed everything. The problem has been what I suspect have been in some cases 'rogue players' and Nominet's inability to enforce its own rules. And yet it is the honest registrars who have kept to the rules who will take the economic 'hit'.

Nominet allowed 5 years for rights holders to claim their .uk domains (actually more than 6 years for those mass-registrations dropping next month). And yet there seems to be a bit of a headlong rush now over pushing the present issue to a conclusion, at exactly the wrong time for many members. As the petition demonstrates, serious and valued members are not happy with this.

If you want to talk ethics and good practice, I commend you to read Simon Blackler's 'creed' at Krystal. It is maybe not chance that Simon is one of the signatories on the petition. So is Stephan Ramoin. These are quality people and I think the concerns raised on the petition need further acknowledgment, and a little more respect.


Where should Nominet direct the profits from holding auctions for expired domain names, or charging for drop catching connections?


In the event that Nominet goes ahead with one of its chosen options, I believe that ideally Nominet should distance itself from the distribution of these funds. I think a separate body should oversee funds distribution. Personally I should like to see those funds directed towards collaborative work with existing charities, aimed at using technology to solve problems (for example in the social care sector) and helping to increase digital access for marginalised communities, groups and individuals. Nominet should be applauded for some of its recent initiatives, such as its collaboration with the Samaritans. However, I do think that the financial windfall from any new model should be channelled to a Trust that is distanced from Nominet. That way you avoid being accused of doing it just for PR, or of simply switching financial costs of you existing public benefit programme to the windfall fund, which would mean that Nominet was not actually channelling any more funds to good causes, but simply associating it with the new model to win people over. I believe that Nominet should continue its own public benefit work, but avoid all charges of profiteering by making the new revenue source accountable to a separate and independent Trust.

 

I really think it is either a or b, i doubt Nominet are interested in anything else at this point. I chose b all day long over a auction model.

 

Agreed, although I was despondent at first I did eventually realise if we all took that view then Nominet will just assume we don't care and press ahead with their plans so I submitted my response a while back. I suggest anyone who was thinking like me previously should get a response in so at least we can say we tried. Saying there's no point will just 100% ensure they press ahead with auctions.

The 2 options does feel like a bit of a trap though as most of us don't want either of them, but i went with the economically controlled access option over the auction model because it is obvious they will be implementing one of these options so it is a case of making the best out of a bad situation.

 

it's still open at
https://surveys.nominet.org.uk/s/ZHP616/

If you're a Nominet member and have not done so, please go complete it now

 

Yes this is so important.