ICANN punts on Oman meeting decision

There can be few in the ICANN community feeling confident that the Org’s annual general meeting is going to take place in Oman, as currently planned, but its board has kicked the can on a decision to go ahead or cancel.

The AGM is due to happen in Muscat from October 17, but the US-Israel war on Iran has raised serious question marks about the venue’s suitability to host a large international meeting.

Oman has been one of several allied countries targeted by Iranian retaliatory drone and missile strikes. Three people were killed there by such attacks in March, at the start of the war.

With news about the likely end date of the conflict changing on a daily basis, and seemingly subject to the whims and domestic pressures of Donald Trump, it remains unclear whether Muscat will be a safe or practical destination come October.

ICANN’s board of directors had “October 2026 ICANN Meeting Venue Contracting” on its consent agenda for its meeting last Sunday, but there’s currently no published resolution, if one was passed.

It’s possible that the motion was shunted from the consent agenda to the main agenda, where two “confidential” topics were discussed. But the absence of a public announcement suggests no decision to relocate the meeting has yet been taken.

The likely issue with Oman as a destination likely isn’t so much one of safety — though that is certainly a concern — but one of practicality. Air travel in the region has been seriously complicated by the war.

I recently had a 12-hour layover in Abu Dhabi, another Iranian target, and had to take shelter from drone or missile attacks at the airport on five separate occasions.

While that’s no joke, it was quite chucklesome that the UAE government’s alerts, pushed automatically to every phone in the danger zone, were signed off with “MOI” for Ministry of Interior, which my phone decided to read aloud in an almost flirtatious female voice as “mwah”.

Over 100 new gTLD bids have already been announced

With the 2026 ICANN new gTLD application window now officially open, one striking difference compared to the 2012 round is the number of organizations that have broken cover to announce that they will apply.

Back in 2012, consultants excitedly pointed to Canon, the Japanese electronics firm, as the convincer. It had openly revealed it planned to apply for the dot-brand .canon two years earlier.

A handful of gTLDs related to communities, cities or causes — such as .gay, .nyc and .eco — had also been announced. Several had multiple announced applicants and were ultimately contested.

But today, with the application window now five days open, DI’s free risk analysis tool, Stringtel, currently has more than 100 announced applications in its database.

With dot-brands expected to make up a sizeable chunk of 2026 applications, very few major brands have actually put their heads above the parapet. Salesforce seems to be the most well-known, though its revelation came via an ICANN director’s conflict disclosure rather than a full-throated formal announcement.

The biggest cohort of announced applications seems to come from applicants representing an if not entirely then certainly uncommon type of gTLD — which you could call the non-dot-brand-dot-brand, or perhaps more simply “open dot-brand”.

By this I mean organizations, typically in the blockchain or cryptocurrency space, that have said they will apply for gTLDs matching their brand but make second-level domains available to their users, rather then keeping the whole namespace in-house.

There are dozens of such announced bids. Some are already selling domains that resolve on blockchains rather than DNS. Due to cost, complexity, or risk, I don’t expect every announcement in this space to translate into an eventual application.

We’ve also seen announcements of applications for generic strings, often in hot industries such as AI, crypto or podcasting. Stringtel already has records for the likes of .crypto, .agi, .podcast and .blockchain, for example. A couple, such as .chain and .anime, are already contested.

While we’ll likely discover how many applications ICANN has received not long after the application window closes in August, we won’t find out what strings have been applied for until October at the earliest.

2026 new gTLD round has actually opened

As of today, you can now apply to own a piece of the internet’s root zone for the first time since 2012.

Almost unbelievably, given some of its relatively recent history, ICANN has hit its deadline and opened up its systems for companies and organizations to file applications for new gTLDs.

The application window opened late April 30 and will close August 12. Many applicants will have been working on their bids for some time, but may hold out until later in the window to actually commit.

ICANN CEO Kurt Lindqvist said in a press release: “Whether building a brand for a company, spotlighting a geographic region or city, strengthening a community, or launching a business to offer domain names under a new registry, a new gTLD can be an innovative tool for commerce, security, and communication.”

ICANN noted that it’s expecting to receive applications for gTLDs in non-Latin scripts — 27 are available — to broaden the linguistic diversity of the DNS. Whether the Org has done enough awareness-raising outreach in non-Anglophone regions is an open question, however.

The cost of applying begins at $227,000. That’s the base application fee, but it will likely often be bumped up by thousands for applicants that need special extra evaluation services.

There’s also going to be an auction of last resort for competing applications for the same strings, where ICANN pockets the proceeds. Unlike the 2012 round, there’s no ICANN-endorsed pathway to privately resolving contention sets for cash.

As many as 75 organizations around the world may have qualified for the Applicant Support Program, which will subsidize application fees by as much as 80%.

We won’t know who has applied for what until probably around mid-October. Applicants then get two weeks to change their strings to their back-ups if they find themselves in unwelcome contention sets. Final strings are confirmed in November.

That’s all assuming the 2026 room is more or less the same size as the 2012 round, in which there were 1,930 applications. A larger batch of applications may delay things a little.

Applications can be filed via ICANN’s new TLD Application Management System (TAMS). The Org has also made a great number of documents and archived webinars available that talk prospective applicants through the process.

Applicants, or simply the curious, can also use DI’s free Stringtel tool to investigate the risks and opportunities associated with their chosen gTLD strings.

Is a .tree gTLD very cool or very silly?

.tree is one of the first-ever publicly announced gTLD applications, predating even the 2012 application round, and it seems to have been proposed for a second time.

A web site using dottree.org, registered a week ago, is saying there’s a plan to apply for .tree in the forthcoming ICANN application round, with a charitable goal.

The project says it will donate a dollar from every .tree domain registration and annual renewal to a charity that plants trees in Dubai, with each $1 donation putting one tree in the ground.

At first glance, it seems like a nice idea, but I’m not sure it holds up to closer inspection.

There may be virtue-signalling advantages to running a site on a .tree domain if you’re in the environmental game or lumber business.

But if a registrant’s primarily interest is getting trees planted, why not just donate a buck to a reforestation charity directly? Why not donate the full amount you would otherwise have spent on the domain? Or more?

And how much demand would there be for .tree domains on their own merits?

Fortunately, DI’s new Stringtel tool has some data there. According to Stringtel, about 47,000 current .com/.net/.org domain names end in the substring “tree”, which may give an indication of potential registrations. About 2,000 of those end in “familytree”.

That’s the same frequency as we see domains end in “wine”, “dog”, or “paris”, and the three gTLDs matching those strings each have 17,000 to 18,000 domains under management, which is not terrible.

Should .tree perform just as well, envelope-based calculation suggest it would create the equivalent of a smallish forest that could be traversed on foot in minutes and not really suck much CO2 out of our increasingly fragile atmosphere.

But the .tree proposal would have a second forest of the same size planted in the second year (assuming 100% domain renewals) and so on. The effect would stack up over time, so maybe the idea does have merit.

The internet just got its first new TLD since 2022

There’s a new gTLD in the root, the first time ICANN has added a string in over four years.

Don’t get too excited though: it’s a dot-brand, kinda.

As of the weekend, .merck is live and resolving, though only the mandatory nic.merck registry domain exists right now.

The gTLD is interesting because it seems to be a rare example of two companies with the same name sharing a dot-brand.

The only other such arrangement I’m aware of is .sas, which is shared by the SAS Institute and SAS Airline, neither of which actually use it.

.merck seems to be jointly controlled by two pharmaceutical companies, one American and one German, both called Merck, which were under common ownership until World War I split them apart.

After they both applied for .merck in ICANN’s 2012 application round, over a decade of lawyering followed before they finally came to an arrangement.

There’s no Specification 13 in the .merck Registry Agreement, so it’s not technically an exclusive-use dot-brand at this time.

The back-end registry services provider, perhaps surprisingly, is South Africa-based DNS Africa, in what seems to be the company’s first deal outside its home continent.

.radio’s new owners might have a fight on their hands

Should gTLD registries that avoid costly auctions by promising to serve special limited communities be allowed to later change their business models to improve sales?

That’s the question being asked of .radio, which recently was taken over by a new owner and might be on track to liberalize its registration policies in the near future.

.radio is what is called a “Community” or “Spec 12” gTLD, meaning it has Specification 12 — which imposes eligibility and content rules on registered domains — in its ICANN registry contract.

Four organizations applied for .radio in 2012, but the European Broadcasting Union prevailed, avoiding a potentially expensive auction, by filing a Community application and winning a Community Priority Evaluation.

Afilias, Donuts and BRS Media — all commercial, non-Community applicants — lost out on their chances to run .radio, or make money losing a private auction, because of this successful CPE.

While .radio launched in 2017, it’s never had more than about 3,000 domains under management and earlier this year the EBU sold its contract to Digity, an emerging registry operator with ties to Sav.com.

The suspicion is that Digity is planning to abandon .radio’s community roots and make it more of a mass-market proposition.

That concern was recently expressed by Craig Schwartz, president of fTLD Registry Services, which runs the Spec 12 gTLDs .bank and .insurance, in a letter (pdf) to ICANN’s chair.

Schwartz noted that Digity, unlike the EBU, “appears to have no existing relationship with the originally stated radio community” and said that registries should not be allowed “to game ICANN by preferentially acquiring these assets under one premise and then blindsiding Registrants at a later date.”

ICANN should audit Spec 12 registries for community compliance and open to public comment any attempt to transfer ownership of the contracts, he wrote.

Schwartz reiterated these concerns during a call of the ICANN At-Large community yesterday, adding that ICANN should come up with a way to “wind down” failing Spec 12 gTLDs, rather than transferring them to a liberalizing successor.

He — saying he is supported by other community registries — appears concerned that if ICANN allows registries to play fast and loose with their original commitments, it could not only enable gaming but also tarnish the reputation of other Spec 12s such as .bank.

Consultant Michael Palage, on the same At-Large call, said the the “cheat code” for registries is to get their Spec 12 removed during bilateral negotiations with ICANN Org at the time of contract renewal. He pointed to the liberalization of GoDaddy’s .xxx as an example.

Probably not coincidentally, the current 10-year .radio Registry Agreement is due to expire this July and is currently up for renewal.

ICANN has yet to publicly respond to Schwartz’s letter.

ICANN cleaning house, cans four more registrars

ICANN has withdrawn the accreditations of four more long-defunct registrars, bringing this month’s terminations so far to 11.

They’re all Chinese, though they do not appear to be under common ownership. They are: Qinghai Yunnet Electronics Technology Co, Shandong Huaimi Network Technology Co, Xiamen Booksir Qiyoutong Technology Co and Xiamen Yuwang Technology Co․

What’s notable is how long it’s taken for ICANN to yank their accreditations. It’s been three or four years since Compliance opened tickets on each of them for non-payment of fees.

None of the four have any gTLD domains under management, and some don’t even seem to own their own original domains any more. One had its former web site turned into a blog in 2022. Another has its domain parked and listed for sale.

ICANN to throw millions more at cheapo gTLDs

ICANN wants to increase the number of new gTLD applications it will subsidize from its own coffers, as well as the size of the discounts it will provide.

The Applicant Support Program was originally budgeted at $10 million, with half coming from application fees and half coming from the proceeds of auctions from the 2012 application round.

The plan was to offer up to 45 qualifying applicants a 75% discount on their application fees, along with a collection of other perks such as hookups with pro-bono consultants.

But it turns out the ASP was oversubscribed — there are currently 75 organizations vying for the discount, and ICANN seems to suspect most or all will be found eligible for support.

About a month ago, after public comment, the ICANN board of directors requisitioned an extra $4.9 million from the auction proceeds war chest to cover the extra subsidies.

But the Governmental Advisory Committee and At-Large Advisory Committee have both recently recommended that ICANN increases the discount to 85% — about $192,500 of the $227,000 base evaluation fee.

So now the board wants to know if the community is cool with it dipping into the auction proceeds to cover the difference to the tune of an extra $3.2 million.

It’s asking for feedback via correspondence, outside of the usual more formal public comment process, by April 12.

Introducing Stringtel, my new free new gTLD tool

I’ve launched Stringtel, a free, industry-first string discovery and risk mitigation tool for new gTLD applicants.

Stringtel is designed to help applicants reduce the risk of their chosen gTLD strings being banned or incurring extra costs during the application process, as well as helping them discover potentially valuable undelegated strings.

The TL;DR

Stringtel gives you data that will help you pick a good string to apply for, and tells you some of the risks that string might present during the application process.

The goal is to help applicants avoid wasting tens or hundreds of thousands of dollars on crappy applications.

Risk mitigation

Stringtel implements the string-related rules in the latest version of ICANN’s New gTLD Applicant Guidebook, along with other risk factors, leveraging a database of almost a million possibly problematic strings, to offer applicants a quick look into issues that could kill or complicate their applications.

Enter a string and Stringtel will tell you if it would be blocked outright under AGB rules, or could trigger additional analysis, objections, fees, or contention with other applicants.

Stringtel does dozens of risk checks, along with countless associated string similarity checks, to give you a shortlist of the most likely reasons your application might fail.

Opportunity identification

New gTLDs might have a better chance of succeeding when they reflect how domains are already being used.

Stringtel analyzes over 180 million domains across .com, .net and .org, counting how often specific strings appear immediately to the left of the dot.

If a string is already widely used as a domain ending, that’s a signal of existing demand. For example, if tens of thousands of domains already end in “bakery”, that suggests a .bakery gTLD could have a ready-made market.

Essentially, Stringtel shows you where registrants are already behaving as if the dot were somewhere else — and where a new gTLD could turn that behavior into shorter domains.

It also helps filter out strings that look appealing but have little real-world usage, reducing the risk of applying for a gTLD nobody actually wants.

And because everyone keeps asking…

No, Stringtel does not record your searches. The data would be useless. I have no visibility whatsoever into what you’re searching for. Neither does anybody else.

Thanks to the sponsors

Many thanks go to Stringtel’s two launch sponsors: Hello Registry and John Matson Consulting.

Hello Registry, a venture of leading ccTLD registries CIRA (.ca) and SIDN (.nl) hosted a webinar on March 31 explaining how to apply for and operate a new gTLD.

Matson has launched TLD.fit, a financial modelling tool that helps new gTLD applicants build their business cases before they pull the trigger on an application.

Seven registrars get terminated

ICANN has terminated the accreditations of seven registrars for not paying their fees.

Haveaname, InstantNames, MisterNIC, NetEstate, Neudomain, OpenName, and TopSystem — all under common ownership in the US — have all been given their marching orders, effective April 17.

While Compliance said it will transfer the registrars’ domains to another registrar, in practice it seems that none of them actually have any remaining domains under management.

As I previously blogged, the seven all appear to have sacrificed their DUM when they lost their .com accreditations in late 2024. That’s about the same time as ICANN stopped receiving its fees.