Expanding pie, shrinking piece
Hi all, As ICANN plunges itself into yet another new-gTLD feeding frenzy supported by ALAC, perhaps it should be reminded again that this ever-expanding universe of 'memorable' domains is chasing an ever-decreasing number of users. Perhaps it is worth contributing some perspective and research to the ALAC conversation when considering what end users really need from ICANN, and never-ending expansion of the domain space is not among those needs. Consider that the share of Internet traffic that is initiated by user-typed domains is in the range of *4-7*%. This is down from *9-14%* in 2017 and *13-20%* in 2010. This is clearly a downward trend that has not yet bottomed out. Statistics are hard to come by, because most surveys lump typed-URL traffic together with bookmarks (in which your browser remembers the domain name but you don't) and dark social into the category of "Direct Traffic" to differentiate from lookups done via search, QR codes, etc. Sources are linked below. This is what everyone here is fighting over: mid-single-digits of how people get to their Internet destinations. Yet ICANN operates on a much larger budget than orgs like IETF or W3C -- which critically affect how the *entire* Internet works -- because of all the money sloshing around from sources such as fees for new gTLDs that -- provably -- nobody needs except ICANN itself and domain speculators. That domains are really rented rather than sold obviously provides ongoing revenue; whether this model benefits end-users is certainly debatable. This isn't Internet Governance. It's heavily-conflicted regulatory cosplay, overseeing a tiny (and shrinking in importance) piece of the Internet infrastructure, designed to make money for some while diverting attention from real issues affecting end-users. Such issues include domain abuse that goes well beyond the industry-created ICANN definition, that even its non-political technical bodies refuse to acknowledge. Where is ALAC on this? I used to think that the ICANN community bubble of self-delusion insulated it from the broader Internet ecosystem. I now realize that it actually insulates the broader Internet ecosystem against being infected by ICANN; this is why absolutely NOBODY is considering ICANN processes for AI or other critical Internet governance. The term "multistakeholder" now appears toxic outside the bubble, though its good intentions are already being reinvented under different names elsewhere. I strongly suggest that At-Large re-evaluate its role within this bubble, to redouble its advocacy of genuine end-user needs, starting with an honest evaluation of what those needs are (as opposed to what we wish to be or what other communities wish us to chase). That is: Just how can ALAC make life better for that 4-7% of the world that still depends on typed URLs? And what else matters? Sources: ACM <https://dl.acm.org/doi/fullHtml/10.1145/3473343> Sparktoro <https://sparktoro.com/blog/new-research-dark-social-falsely-attributes-signi...> Mental Momentum <https://research.mental-momentum.ai/r/dark-social-marketing-attribution-2026...> Averi <https://www.averi.ai/blog/57-of-our-traffic-is-direct-unknown-what-it-means> GR0 <https://gr0.com/blog/what-is-direct-traffic> Sitechecker <https://sitechecker.pro/what-is-direct-traffic> -- Evan Leibovitch, Toronto Canada @evanleibovitch / @el56
Evan, Although I tend to agree with your general sentiment I disagree that the primary target are typed-URLs. One value of a domain is an identifier for a business or activity, like a phone number or trade name. What would you put on an ad if not your domain along with name, phone number, etc.? Whether few transcribe that to a URL bar is not really the concern of ICANN, what if you were selling million-dollar sports cars? You're being too prescriptive. Either people buy domains or they don't, either the economics works for those involved or they don't*. Otherwise it's like someone, an industry regulator like ICANN, saying sorry we have enough flavors of ice cream we don't need more. On that note my feeling is that ICANN charges far too little for TLDs and this is why they flood the net with hundreds of new TLDs most of which are bound to become moribund and fail. It's become a high-level version of "domain tasting" if any remember that controversy. It encourages speculation in a product which should be providing trust and reliability. My off-hand guess is that new TLDs should cost over US$1M to start, maybe much more, with various considerations for the various "worthy causes" before someone goes there. * Criminality and abuse of the net aside, another area where new TLDs contribute but that hasn't been raised here, yet. -- -Barry Shein Software Tool & Die | bzs@TheWorld.com | http://www.TheWorld.com Purveyors to the Trade | Voice: +1 617-STD-WRLD | 800-THE-WRLD The World: Since 1989 | A Public Information Utility | *oo*
On Sat, May 30, 2026 at 9:22 PM <bzs@theworld.com> wrote: One value of a domain is an identifier for a business or activity, like a
phone number or trade name. What would you put on an ad if not your domain along with name, phone number, etc.? Whether few transcribe that to a URL bar is not really the concern of ICANN, what if you were selling million-dollar sports cars?
I think you missed my point entirely. I look from the focused perspective of the Internet user, who is neither domain buyer nor seller, where none of the above matters. So long as I can reach my intended target easily I'm happy. How I got there is simply not a big deal, and as I mentioned the number of people who get where they're going by typing a URL is an extremely small proportion. Were I to care about the marketing of a brand I would think its primary question would not be "which domain should I buy?" but rather "should I spend my money on a premium domain or SEO or referral advertising?". But I *don't* care. This is At-Large and its role is not to sell domains or even the idea of domains; it's to advance that when domains *are* used they're in the best interests of Internet consumers who aren't anywhere on ICANN's revenue chain.
You're being too prescriptive. Either people buy domains or they don't, either the economics works for those involved or they don't*.
For the end-user the level of choice available to registrants is irrelevant. Indeed for the end user, most of ALAC's interventions -- notably on processes related to the domain buyer-seller relationship -- are irrelevant. That's why I keep asking for greater focus on the actual single global constituency ALAC is mandated to serve. Otherwise it's like someone, an industry regulator like ICANN, saying sorry
we have enough flavors of ice cream we don't need more.
Except ice cream actually requires physical ingredients. Allocate all your milk and sugar to some flavours and you're not able to make others. These are natural restraints. Once a gTLD is allocated out of thin air, it can then create a nearly infinite number of subdomains (also out of thin air) so long as it can convince enough suckers to buy. No restraints. I suspect that ICANN is not interested in the world knowing that no more than 7% of users (and likely even less) make use of hand-typed domains, let alone memorable ones, as it has a conflicted vested interest in maximizing revenue from applications.
On that note my feeling is that ICANN charges far too little for TLDs and this is why they flood the net with hundreds of new TLDs most of which are bound to become moribund and fail.
A relevance to end-users does exist here because stability matters. Bookmarking a destination URL, only for that bookmark to be invalid after some years because the TLD went away or changed policies decreases one's effective use of the Internet and it diminishes trust in the DNS. This stability/trust issue is probably the only element of gTLD expansion that should matter to At-Large. In that regard I agree with your assessment, and that ALAC should advocate for ICANN to charge much higher fees for gTLDs. I would go further and advocate for increasing the ICANN rental price of subdomains too.
It's become a high-level version of "domain tasting" if any remember that controversy. It encourages speculation in a product which should be providing trust and reliability.
I would posit that one of ALAC's only true advancements of end-user interests -- ever -- was its successful effort to eliminate domain tasting. Alan Greenberg deserves much praise for his role leading that. * Criminality and abuse of the net aside, another area where new TLDs contribute
but that hasn't been raised here, yet.
That depends on what you mean by "here". I raised it specifically in my original post in this thread, paragraph 5. It's a wholly valid claim that ICANN and ALAC have been asleep at the wheel on this very important end-user issue, which even SSAC refuses to sufficiently address. Only ALAC can make this issue matter to ICANN. - Evan
I have never been one of the "Let a thousand flowers bloom" proponents. It was a foundational concept within ICANN and it obviously still is. It may not help (or particularly hurt) end users - with a few possible exceptions. Regardless, we are not going to stop this process, so I think that At-Large has correctly focused on a few areas of the program that do have importance to us: - IDNs - Community TLDs - Underserved areas - Contention resolution/auctions (which impacts ICANNs credibility and generates funds for "good things"). Alan On Sun, May 31, 2026 at 8:45 AM Evan Leibovitch via At-Large < at-large@icann.org> wrote:
On Sat, May 30, 2026 at 9:22 PM <bzs@theworld.com> wrote:
One value of a domain is an identifier for a business or activity, like a
phone number or trade name. What would you put on an ad if not your domain along with name, phone number, etc.? Whether few transcribe that to a URL bar is not really the concern of ICANN, what if you were selling million-dollar sports cars?
I think you missed my point entirely.
I look from the focused perspective of the Internet user, who is neither domain buyer nor seller, where none of the above matters. So long as I can reach my intended target easily I'm happy. How I got there is simply not a big deal, and as I mentioned the number of people who get where they're going by typing a URL is an extremely small proportion.
Were I to care about the marketing of a brand I would think its primary question would not be "which domain should I buy?" but rather "should I spend my money on a premium domain or SEO or referral advertising?". But I *don't* care. This is At-Large and its role is not to sell domains or even the idea of domains; it's to advance that when domains *are* used they're in the best interests of Internet consumers who aren't anywhere on ICANN's revenue chain.
You're being too prescriptive. Either people buy domains or they don't, either the economics works for those involved or they don't*.
For the end-user the level of choice available to registrants is irrelevant. Indeed for the end user, most of ALAC's interventions -- notably on processes related to the domain buyer-seller relationship -- are irrelevant. That's why I keep asking for greater focus on the actual single global constituency ALAC is mandated to serve.
Otherwise it's like someone, an industry regulator like ICANN, saying sorry
we have enough flavors of ice cream we don't need more.
Except ice cream actually requires physical ingredients. Allocate all your milk and sugar to some flavours and you're not able to make others. These are natural restraints. Once a gTLD is allocated out of thin air, it can then create a nearly infinite number of subdomains (also out of thin air) so long as it can convince enough suckers to buy. No restraints. I suspect that ICANN is not interested in the world knowing that no more than 7% of users (and likely even less) make use of hand-typed domains, let alone memorable ones, as it has a conflicted vested interest in maximizing revenue from applications.
On that note my feeling is that ICANN charges far too little for TLDs and this is why they flood the net with hundreds of new TLDs most of which are bound to become moribund and fail.
A relevance to end-users does exist here because stability matters. Bookmarking a destination URL, only for that bookmark to be invalid after some years because the TLD went away or changed policies decreases one's effective use of the Internet and it diminishes trust in the DNS. This stability/trust issue is probably the only element of gTLD expansion that should matter to At-Large. In that regard I agree with your assessment, and that ALAC should advocate for ICANN to charge much higher fees for gTLDs. I would go further and advocate for increasing the ICANN rental price of subdomains too.
It's become a high-level version of "domain tasting" if any remember that controversy. It encourages speculation in a product which should be providing trust and reliability.
I would posit that one of ALAC's only true advancements of end-user interests -- ever -- was its successful effort to eliminate domain tasting. Alan Greenberg deserves much praise for his role leading that.
* Criminality and abuse of the net aside, another area where new TLDs contribute
but that hasn't been raised here, yet.
That depends on what you mean by "here". I raised it specifically in my original post in this thread, paragraph 5. It's a wholly valid claim that ICANN and ALAC have been asleep at the wheel on this very important end-user issue, which even SSAC refuses to sufficiently address. Only ALAC can make this issue matter to ICANN.
- Evan
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
Thank you all, Evan raises an uncomfortable but important question that At-Large should not dismiss simply because the answer may challenge long-held assumptions. The issue is not really whether typed URLs account for 4%, 7%, or 10% of user navigation. The deeper question is whether At-Large is aligning its priorities with how Internet users actually experience the Internet in 2026. There is little doubt that the visibility of domain names in the user experience has diminished. Search engines, social media platforms, mobile applications, QR codes, AI assistants, recommendation engines, and embedded links increasingly mediate access to online resources. Most users no longer consciously interact with the DNS in the way they did twenty years ago. However, I would caution against drawing the conclusion that domain names have therefore become irrelevant. The DNS remains one of the foundational trust and identity layers of the Internet. Users may no longer type domains frequently, but they continue to depend on them indirectly every day. Domain names remain critical for authentication, email, business identity, digital sovereignty, routing, resilience, and trust. The fact that infrastructure becomes invisible does not make it unimportant. Most users never think about TCP/IP either, yet we would hardly conclude that Internet protocols no longer matter. The more important challenge for At-Large is therefore not whether domains matter, but whether our policy priorities accurately reflect the risks and opportunities facing end users. In that respect, Evan's criticism deserves serious consideration. Too often, At-Large discussions become absorbed in registrar-registry relationships, procedural debates, and implementation details that may be highly relevant to contracted parties but only indirectly relevant to Internet users. The question we should continually ask is: *What tangible benefit does this policy discussion provide to Internet users?* Applied to the next round of new gTLDs, the answer cannot simply be "more choice." Choice is only valuable if it advances identifiable public-interest goals. From an end-user perspective, the more compelling questions are: - Will new gTLDs improve trust and reduce abuse? - Will they increase linguistic and cultural inclusion through IDNs? - Will they support underserved communities and regions? - Will they strengthen competition without creating instability? - Will they improve accessibility and meaningful participation in the digital economy? These are questions where At-Large can and should add value. I therefore agree with Alan's observation that At-Large's focus on IDNs, community applications, underserved regions, and fair contention resolution reflects areas where clear user interests are at stake. At the same time, I also agree with Evan that At-Large should devote greater attention to issues that increasingly shape users' actual online experience: - DNS abuse and fraud; - trust and security; - privacy and data governance; - digital inclusion; - the impact of AI-driven discovery systems on Internet openness; - concentration of power in platforms and intermediaries; - preserving user choice and agency in a world where navigation is increasingly mediated by algorithms rather than browsers. Indeed, one could argue that the future challenge for At-Large is not the decline of typed URLs but the rise of algorithmic gatekeepers. If users increasingly reach information through search engines, app stores, social media feeds, and AI assistants, then questions of openness, interoperability, transparency, and user choice become central public-interest concerns. *These concerns are directly connected to the Core Internet Values that At-Large has historically championed.* The real issue, therefore, is not whether the DNS is shrinking in importance. Rather, it is *whether At-Large is sufficiently evolving its conception of end-user interests to reflect how people actually interact with the Internet today.* In my view, the answer is that we must do both: continue protecting the stability, security, and inclusiveness of the DNS while expanding our attention to the broader ecosystem of digital intermediaries that increasingly shape users' online experiences. That evolution would not diminish At-Large's mission. It would strengthen its relevance. Cheers, Pari Esfandiari -- Pari Esfandiari President *Global TechnoPolitics Forum <http://www.technopolitics.org/> * *Pario <http://www.parioconsultants.com/>- Architects of Ideas* info@TechnoPolitics.org <info@technopolitics.org> *Linkedin Profile <https://www.linkedin.com/in/pariesfandiari/>* *Tel: * US :+1-202*-735-1415* (Office) US : +1-310-435-0888 <http://voice.google.com/calls?a=nc,%2B13104350888> <http://voice.google.com/calls?a=nc,%2B13104350888> <http://voice.google.com/calls?a=nc,%2B13104350888> <http://voice.google.com/calls?a=nc,%2B13104350888>(Cell) Europe: +30-694-1607131 <http://voice.google.com/calls?a=nc,%2B306941607131> <http://voice.google.com/calls?a=nc,%2B306941607131> <http://voice.google.com/calls?a=nc,%2B13104350888> <http://voice.google.com/calls?a=nc,%2B13104350888>(Cell) UK : +44-731-210-4049 <http://voice.google.com/calls?a=nc,%2B447312104049> <http://voice.google.com/calls?a=nc,%2B447312104049> <http://voice.google.com/calls?a=nc,%2B447312104049> <http://voice.google.com/calls?a=nc,%2B447312104049> (Cell) -- On Sun, May 31, 2026 at 6:35 PM Alan Greenberg via At-Large < at-large@icann.org> wrote:
I have never been one of the "Let a thousand flowers bloom" proponents. It was a foundational concept within ICANN and it obviously still is.
It may not help (or particularly hurt) end users - with a few possible exceptions.
Regardless, we are not going to stop this process, so I think that At-Large has correctly focused on a few areas of the program that do have importance to us: - IDNs - Community TLDs - Underserved areas - Contention resolution/auctions (which impacts ICANNs credibility and generates funds for "good things").
Alan
On Sun, May 31, 2026 at 8:45 AM Evan Leibovitch via At-Large < at-large@icann.org> wrote:
On Sat, May 30, 2026 at 9:22 PM <bzs@theworld.com> wrote:
One value of a domain is an identifier for a business or activity, like
a phone number or trade name. What would you put on an ad if not your domain along with name, phone number, etc.? Whether few transcribe that to a URL bar is not really the concern of ICANN, what if you were selling million-dollar sports cars?
I think you missed my point entirely.
I look from the focused perspective of the Internet user, who is neither domain buyer nor seller, where none of the above matters. So long as I can reach my intended target easily I'm happy. How I got there is simply not a big deal, and as I mentioned the number of people who get where they're going by typing a URL is an extremely small proportion.
Were I to care about the marketing of a brand I would think its primary question would not be "which domain should I buy?" but rather "should I spend my money on a premium domain or SEO or referral advertising?". But I *don't* care. This is At-Large and its role is not to sell domains or even the idea of domains; it's to advance that when domains *are* used they're in the best interests of Internet consumers who aren't anywhere on ICANN's revenue chain.
You're being too prescriptive. Either people buy domains or they don't, either the economics works for those involved or they don't*.
For the end-user the level of choice available to registrants is irrelevant. Indeed for the end user, most of ALAC's interventions -- notably on processes related to the domain buyer-seller relationship -- are irrelevant. That's why I keep asking for greater focus on the actual single global constituency ALAC is mandated to serve.
Otherwise it's like someone, an industry regulator like ICANN, saying sorry
we have enough flavors of ice cream we don't need more.
Except ice cream actually requires physical ingredients. Allocate all your milk and sugar to some flavours and you're not able to make others. These are natural restraints. Once a gTLD is allocated out of thin air, it can then create a nearly infinite number of subdomains (also out of thin air) so long as it can convince enough suckers to buy. No restraints. I suspect that ICANN is not interested in the world knowing that no more than 7% of users (and likely even less) make use of hand-typed domains, let alone memorable ones, as it has a conflicted vested interest in maximizing revenue from applications.
On that note my feeling is that ICANN charges far too little for TLDs and this is why they flood the net with hundreds of new TLDs most of which are bound to become moribund and fail.
A relevance to end-users does exist here because stability matters. Bookmarking a destination URL, only for that bookmark to be invalid after some years because the TLD went away or changed policies decreases one's effective use of the Internet and it diminishes trust in the DNS. This stability/trust issue is probably the only element of gTLD expansion that should matter to At-Large. In that regard I agree with your assessment, and that ALAC should advocate for ICANN to charge much higher fees for gTLDs. I would go further and advocate for increasing the ICANN rental price of subdomains too.
It's become a high-level version of "domain tasting" if any remember that controversy. It encourages speculation in a product which should be providing trust and reliability.
I would posit that one of ALAC's only true advancements of end-user interests -- ever -- was its successful effort to eliminate domain tasting. Alan Greenberg deserves much praise for his role leading that.
* Criminality and abuse of the net aside, another area where new TLDs contribute
but that hasn't been raised here, yet.
That depends on what you mean by "here". I raised it specifically in my original post in this thread, paragraph 5. It's a wholly valid claim that ICANN and ALAC have been asleep at the wheel on this very important end-user issue, which even SSAC refuses to sufficiently address. Only ALAC can make this issue matter to ICANN.
- Evan
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
On Sun, May 31, 2026 at 1:52 PM Pari Esfandiari <pariesfandiari@gmail.com> wrote:
Thank you all,
And thank you for your response, with which I largely agree. The DNS is far from irrelevant. Regardless of the demand for "memorable" domains, or the share of manually-typed domains in personal web access, having a trusted, secure and useful DNS will remain paramount to a smoothly functioning Internet, at least in our lifetimes. I only have one general comment to your response.
From an end-user perspective, the more compelling questions are:
- Will new gTLDs improve trust and reduce abuse? - Will they increase linguistic and cultural inclusion through IDNs? - Will they support underserved communities and regions? - Will they strengthen competition without creating instability? - Will they improve accessibility and meaningful participation in the digital economy?
As I mentioned in my response to Alan, domain names are not the solution to all of the Internet's inclusion and access challenges. And it might be a mistake for ALAC to tackle every such problem using only ICANN's fairly limited toolkit. I agree that it is imperative to maximize linguistic and cultural inclusion. However I assert that trying to do this through IDNs is futile and becoming moreso every year. A genuine commitment to the end-user perspective means exactly that: understanding what communities, and the underserved, and linguistic/cultural minorities need from us ... not starting with our own assumptions and tools and trying to twist the problem to fit our "solution". (As one example: what about the many linguistic groups whose scripts are supported by Unicode but not punycode -- do we demand for them to wait for ICANN to catch up? Or do we support a Unicode-based solution that might, just, not involve domain names?) We must put ourselves into the perspective of the communities, the underserved, those challenged with access. Right now ALAC looks at them from the perspective of domain landlords, and every challenge assumes some kind of domain-based answer. Sometimes it's appropriate, but just as often not. While such a hit-and-miss approach can be sustained indefinitely with the support of ICANN, it's not serving the end users which means ALAC isn't fulfilling its mandated role. It's also expending significant volunteer resources on efforts that have no chance of success. At least, ALAC needs to choose its battles and realize that it can't solve all problems alone and only with domains. For a start, it needs a direct and formal relationship with the Internet Society whose main mission is inclusion and access. (As a head start to this, many ALSs are ISOC chapters). ALAC could also apply for ICANN project funding to research and survey appropriate groups to know what the-world-at-large needs from ICANN-At-Large. ALAC's being hyper-focused on the end-user first and foremost will demand a substantial change in priorities and outputs. It will demand ALAC members who filter every resolution, every statement for relevance to end-users only. That won't be an easy transition. But it's long overdue. - Evan
From Pari: "*Choice is only valuable if it advances identifiable public-interest goals* ." I profoundly disagree with that proposition, if only because it assumes that the value of choice must first be justified by reference to some external public-interest test. The primary value proposition of choice resides with the individual user. Free people everywhere should retain the unfettered right to make choices that express preferences, priorities and aspirations that others may regard as unwise. Plainly put, they should be free to be foolish. I subscribe to the view that allowing ‘*a thousand flowers to bloom’* creates the greatest opportunity for the public interest, howsoever defined, to emerge organically. Social benefit – and you can add innovation and competition to this! - rarely arise because a central authority successfully predicts them in advance. They arise because individuals and communities have the freedom to pursue their own ideas and determine their own priorities. For underserved communities, the greater burden is often not a lack of public interest objectives but the requirement to satisfy someone else's definition of ‘*societal benefit’* before permission to participate is granted. We see the gatekeeping functions constraining rather than enabling local initiative, innovation and self-determination. From our corner of the world, the more pressing challenge has never been choice itself. It has been the structural realities of the DNS business model and market access that prevent underserved communities from participating on equal footing. Those barriers and other gatekeeping actions, not the availability of choice, are and remain the principal constraints on meaningful stakeholding by underserved communities in the DNS ecosystem. I engaged with the Applicant Support Program (ASP) a decade and half ago precisely because it offered an opportunity to address that imbalance. The current round shows encouraging signs of movement in that direction. Whether those measures ultimately produce broader and more equitable participation remains to be seen. We will know soon enough. Carlton ============================== *Carlton A Samuels* *Mobile: 876-818-1799Strategy, Process, Governance, Assessment & Turnaround* ============================= On Sun, 31 May 2026 at 14:02, Pari Esfandiari via At-Large < at-large@icann.org> wrote:
Thank you all,
Evan raises an uncomfortable but important question that At-Large should not dismiss simply because the answer may challenge long-held assumptions.
The issue is not really whether typed URLs account for 4%, 7%, or 10% of user navigation. The deeper question is whether At-Large is aligning its priorities with how Internet users actually experience the Internet in 2026.
There is little doubt that the visibility of domain names in the user experience has diminished. Search engines, social media platforms, mobile applications, QR codes, AI assistants, recommendation engines, and embedded links increasingly mediate access to online resources. Most users no longer consciously interact with the DNS in the way they did twenty years ago.
However, I would caution against drawing the conclusion that domain names have therefore become irrelevant.
The DNS remains one of the foundational trust and identity layers of the Internet. Users may no longer type domains frequently, but they continue to depend on them indirectly every day. Domain names remain critical for authentication, email, business identity, digital sovereignty, routing, resilience, and trust. The fact that infrastructure becomes invisible does not make it unimportant. Most users never think about TCP/IP either, yet we would hardly conclude that Internet protocols no longer matter.
The more important challenge for At-Large is therefore not whether domains matter, but whether our policy priorities accurately reflect the risks and opportunities facing end users.
In that respect, Evan's criticism deserves serious consideration.
Too often, At-Large discussions become absorbed in registrar-registry relationships, procedural debates, and implementation details that may be highly relevant to contracted parties but only indirectly relevant to Internet users. The question we should continually ask is:
*What tangible benefit does this policy discussion provide to Internet users?*
Applied to the next round of new gTLDs, the answer cannot simply be "more choice."
Choice is only valuable if it advances identifiable public-interest goals.
From an end-user perspective, the more compelling questions are:
- Will new gTLDs improve trust and reduce abuse? - Will they increase linguistic and cultural inclusion through IDNs? - Will they support underserved communities and regions? - Will they strengthen competition without creating instability? - Will they improve accessibility and meaningful participation in the digital economy?
These are questions where At-Large can and should add value.
I therefore agree with Alan's observation that At-Large's focus on IDNs, community applications, underserved regions, and fair contention resolution reflects areas where clear user interests are at stake.
At the same time, I also agree with Evan that At-Large should devote greater attention to issues that increasingly shape users' actual online experience:
- DNS abuse and fraud; - trust and security; - privacy and data governance; - digital inclusion; - the impact of AI-driven discovery systems on Internet openness; - concentration of power in platforms and intermediaries; - preserving user choice and agency in a world where navigation is increasingly mediated by algorithms rather than browsers.
Indeed, one could argue that the future challenge for At-Large is not the decline of typed URLs but the rise of algorithmic gatekeepers.
If users increasingly reach information through search engines, app stores, social media feeds, and AI assistants, then questions of openness, interoperability, transparency, and user choice become central public-interest concerns. *These concerns are directly connected to the Core Internet Values that At-Large has historically championed.*
The real issue, therefore, is not whether the DNS is shrinking in importance. Rather, it is *whether At-Large is sufficiently evolving its conception of end-user interests to reflect how people actually interact with the Internet today.*
In my view, the answer is that we must do both: continue protecting the stability, security, and inclusiveness of the DNS while expanding our attention to the broader ecosystem of digital intermediaries that increasingly shape users' online experiences.
That evolution would not diminish At-Large's mission. It would strengthen its relevance.
Cheers,
Pari Esfandiari -- Pari Esfandiari President *Global TechnoPolitics Forum <http://www.technopolitics.org/> * *Pario <http://www.parioconsultants.com/>- Architects of Ideas* info@TechnoPolitics.org <info@technopolitics.org> *Linkedin Profile <https://www.linkedin.com/in/pariesfandiari/>* *Tel: * US :+1-202*-735-1415* (Office) US : +1-310-435-0888 <http://voice.google.com/calls?a=nc,%2B13104350888> <http://voice.google.com/calls?a=nc,%2B13104350888> <http://voice.google.com/calls?a=nc,%2B13104350888> <http://voice.google.com/calls?a=nc,%2B13104350888>(Cell) Europe: +30-694-1607131 <http://voice.google.com/calls?a=nc,%2B306941607131> <http://voice.google.com/calls?a=nc,%2B306941607131> <http://voice.google.com/calls?a=nc,%2B13104350888> <http://voice.google.com/calls?a=nc,%2B13104350888>(Cell) UK : +44-731-210-4049 <http://voice.google.com/calls?a=nc,%2B447312104049> <http://voice.google.com/calls?a=nc,%2B447312104049> <http://voice.google.com/calls?a=nc,%2B447312104049> <http://voice.google.com/calls?a=nc,%2B447312104049> (Cell)
-- On Sun, May 31, 2026 at 6:35 PM Alan Greenberg via At-Large < at-large@icann.org> wrote:
I have never been one of the "Let a thousand flowers bloom" proponents. It was a foundational concept within ICANN and it obviously still is.
It may not help (or particularly hurt) end users - with a few possible exceptions.
Regardless, we are not going to stop this process, so I think that At-Large has correctly focused on a few areas of the program that do have importance to us: - IDNs - Community TLDs - Underserved areas - Contention resolution/auctions (which impacts ICANNs credibility and generates funds for "good things").
Alan
On Sun, May 31, 2026 at 8:45 AM Evan Leibovitch via At-Large < at-large@icann.org> wrote:
On Sat, May 30, 2026 at 9:22 PM <bzs@theworld.com> wrote:
One value of a domain is an identifier for a business or activity, like
a phone number or trade name. What would you put on an ad if not your domain along with name, phone number, etc.? Whether few transcribe that to a URL bar is not really the concern of ICANN, what if you were selling million-dollar sports cars?
I think you missed my point entirely.
I look from the focused perspective of the Internet user, who is neither domain buyer nor seller, where none of the above matters. So long as I can reach my intended target easily I'm happy. How I got there is simply not a big deal, and as I mentioned the number of people who get where they're going by typing a URL is an extremely small proportion.
Were I to care about the marketing of a brand I would think its primary question would not be "which domain should I buy?" but rather "should I spend my money on a premium domain or SEO or referral advertising?". But I *don't* care. This is At-Large and its role is not to sell domains or even the idea of domains; it's to advance that when domains *are* used they're in the best interests of Internet consumers who aren't anywhere on ICANN's revenue chain.
You're being too prescriptive. Either people buy domains or they don't, either the economics works for those involved or they don't*.
For the end-user the level of choice available to registrants is irrelevant. Indeed for the end user, most of ALAC's interventions -- notably on processes related to the domain buyer-seller relationship -- are irrelevant. That's why I keep asking for greater focus on the actual single global constituency ALAC is mandated to serve.
Otherwise it's like someone, an industry regulator like ICANN, saying sorry
we have enough flavors of ice cream we don't need more.
Except ice cream actually requires physical ingredients. Allocate all your milk and sugar to some flavours and you're not able to make others. These are natural restraints. Once a gTLD is allocated out of thin air, it can then create a nearly infinite number of subdomains (also out of thin air) so long as it can convince enough suckers to buy. No restraints. I suspect that ICANN is not interested in the world knowing that no more than 7% of users (and likely even less) make use of hand-typed domains, let alone memorable ones, as it has a conflicted vested interest in maximizing revenue from applications.
On that note my feeling is that ICANN charges far too little for TLDs and this is why they flood the net with hundreds of new TLDs most of which are bound to become moribund and fail.
A relevance to end-users does exist here because stability matters. Bookmarking a destination URL, only for that bookmark to be invalid after some years because the TLD went away or changed policies decreases one's effective use of the Internet and it diminishes trust in the DNS. This stability/trust issue is probably the only element of gTLD expansion that should matter to At-Large. In that regard I agree with your assessment, and that ALAC should advocate for ICANN to charge much higher fees for gTLDs. I would go further and advocate for increasing the ICANN rental price of subdomains too.
It's become a high-level version of "domain tasting" if any remember that controversy. It encourages speculation in a product which should be providing trust and reliability.
I would posit that one of ALAC's only true advancements of end-user interests -- ever -- was its successful effort to eliminate domain tasting. Alan Greenberg deserves much praise for his role leading that.
* Criminality and abuse of the net aside, another area where new TLDs contribute
but that hasn't been raised here, yet.
That depends on what you mean by "here". I raised it specifically in my original post in this thread, paragraph 5. It's a wholly valid claim that ICANN and ALAC have been asleep at the wheel on this very important end-user issue, which even SSAC refuses to sufficiently address. Only ALAC can make this issue matter to ICANN.
- Evan
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
What was the effort that ALAC did for the support of underserved areas and communities Alan ???? ALAC was in the leadership of the Subpro WG that imposed a policy that ignored the underserved area for the benefit of small businesses and inter-governmental organizations. For underserved communities, only Indigenous community was considered. Of course, in the implementation phase, we couldn't do anything because the agreed policies couldn't be changed. Tijani Le dim. 31 mai 2026 à 18:35, Alan Greenberg via At-Large <at-large@icann.org> a écrit :
I have never been one of the "Let a thousand flowers bloom" proponents. It was a foundational concept within ICANN and it obviously still is.
It may not help (or particularly hurt) end users - with a few possible exceptions.
Regardless, we are not going to stop this process, so I think that At-Large has correctly focused on a few areas of the program that do have importance to us: - IDNs - Community TLDs - Underserved areas - Contention resolution/auctions (which impacts ICANNs credibility and generates funds for "good things").
Alan
On Sun, May 31, 2026 at 8:45 AM Evan Leibovitch via At-Large < at-large@icann.org> wrote:
On Sat, May 30, 2026 at 9:22 PM <bzs@theworld.com> wrote:
One value of a domain is an identifier for a business or activity, like
a phone number or trade name. What would you put on an ad if not your domain along with name, phone number, etc.? Whether few transcribe that to a URL bar is not really the concern of ICANN, what if you were selling million-dollar sports cars?
I think you missed my point entirely.
I look from the focused perspective of the Internet user, who is neither domain buyer nor seller, where none of the above matters. So long as I can reach my intended target easily I'm happy. How I got there is simply not a big deal, and as I mentioned the number of people who get where they're going by typing a URL is an extremely small proportion.
Were I to care about the marketing of a brand I would think its primary question would not be "which domain should I buy?" but rather "should I spend my money on a premium domain or SEO or referral advertising?". But I *don't* care. This is At-Large and its role is not to sell domains or even the idea of domains; it's to advance that when domains *are* used they're in the best interests of Internet consumers who aren't anywhere on ICANN's revenue chain.
You're being too prescriptive. Either people buy domains or they don't, either the economics works for those involved or they don't*.
For the end-user the level of choice available to registrants is irrelevant. Indeed for the end user, most of ALAC's interventions -- notably on processes related to the domain buyer-seller relationship -- are irrelevant. That's why I keep asking for greater focus on the actual single global constituency ALAC is mandated to serve.
Otherwise it's like someone, an industry regulator like ICANN, saying sorry
we have enough flavors of ice cream we don't need more.
Except ice cream actually requires physical ingredients. Allocate all your milk and sugar to some flavours and you're not able to make others. These are natural restraints. Once a gTLD is allocated out of thin air, it can then create a nearly infinite number of subdomains (also out of thin air) so long as it can convince enough suckers to buy. No restraints. I suspect that ICANN is not interested in the world knowing that no more than 7% of users (and likely even less) make use of hand-typed domains, let alone memorable ones, as it has a conflicted vested interest in maximizing revenue from applications.
On that note my feeling is that ICANN charges far too little for TLDs and this is why they flood the net with hundreds of new TLDs most of which are bound to become moribund and fail.
A relevance to end-users does exist here because stability matters. Bookmarking a destination URL, only for that bookmark to be invalid after some years because the TLD went away or changed policies decreases one's effective use of the Internet and it diminishes trust in the DNS. This stability/trust issue is probably the only element of gTLD expansion that should matter to At-Large. In that regard I agree with your assessment, and that ALAC should advocate for ICANN to charge much higher fees for gTLDs. I would go further and advocate for increasing the ICANN rental price of subdomains too.
It's become a high-level version of "domain tasting" if any remember that controversy. It encourages speculation in a product which should be providing trust and reliability.
I would posit that one of ALAC's only true advancements of end-user interests -- ever -- was its successful effort to eliminate domain tasting. Alan Greenberg deserves much praise for his role leading that.
* Criminality and abuse of the net aside, another area where new TLDs contribute
but that hasn't been raised here, yet.
That depends on what you mean by "here". I raised it specifically in my original post in this thread, paragraph 5. It's a wholly valid claim that ICANN and ALAC have been asleep at the wheel on this very important end-user issue, which even SSAC refuses to sufficiently address. Only ALAC can make this issue matter to ICANN.
- Evan
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
On Sun, May 31, 2026 at 12:40 PM Alan Greenberg <greenberg.alan@gmail.com> wrote: I have never been one of the "Let a thousand flowers bloom" proponents. It
was a foundational concept within ICANN and it obviously still is.
What it is, is a foundational flaw. In what ICANN calls "multistakeholder", some stakeholders -- the vested interests who seek to extract value from the DNS -- get to impose their will on the others. It's baked into the system. Two specific communities -- the providers and renters of domains -- have the ability to compel the Board (via the GNSO) to do their bidding over the wishes of governments and the public interest who can only advise. In most other fields of regulation it's the exact opposite; governments and the public set policy while industry advises but cannot compel. That absolutely nobody else has adopted ICANN's model for governance elsewhere clearly indicates that it hasn't lived up to expectations and isn't suitable for use elsewhere. It is also clear that because of this power imbalance, the Internet world had generally bypassed ICANN's chosen methods and followed standards set by groups that better reflect the public interest. There are numerous commercial and noncommercial entities that have served the public better than ICANN ever could hope to do. You've even offered a number of excellent examples: - IDNs
An utter joke; the world has left punycode behind. People everywhere can use search engines, commercial and non-commercial, to find what they want on the Internet using Unicode and its support of more than 170 scripts. ICANN has already recognized the futility of its wheel-reinvention, giving up on the "universal acceptance" begging campaign when it cut off UASG funding last year. Community participants who seriously care about improving global access to the Internet, whether physical or cultural, seriously need to look at other vehicles besides ICANN to accomplish this. UNESCO and ISOC are good places to start. - Community TLDs
This has always been a headache for ICANN, and it's an increasingly pointless one as alternatives develop. Such is the pervasive culture of exploitation, that well-meaning programs such as Applicant Support must erect so many barriers against abuse that they render themselves useless for their intended purpose. And based on Tijani's comments in this thread, not much appears to have changed in this regard since I left ALAC. The ICANN community has always operated under a number of unstated and undeserved conceits. Among them is that having a domain can bestow an identity or cohesion that did not previously exist. Another is that it is reasonable to extract a monthly rental fee from communities (and their members) simply to maintain this identity. These days communities now have access to wonderful tools such as Mastodon, Discord and Reddit for bringing together common interests; they support subgroups, virtual events and real-time communications. More importantly they are free to use, in sharp contrast to the domain world's demand of monthly rent. In this environment in which so many options exist for community building and development that are both cheaper and more effective, ALAC's efforts are not looking out for the best interests of communities. They are looking out for the best interests of domain landlords who want to extract value from communities. - Underserved areas
See above. Another conceit is that "underserved by memorable domain names" has any relation to "underserved by the Internet". Again, the net effect of ALAC's contributions here is to find ways to have the "underserved" pay for domains when less-costly, more-effective paths to Internet access exist. Again, this is not looking out for the best interests of the underserved, but rather for the industry that wants to find and rent domains to them. "When all you have is a hammer, every problem looks like a nail". Getting domains is not always the most effective answer to challenges of access or development; please stop acting as if it is. I await the first instance in which ALAC defines a use case in which better solutions exist to an access problem than getting a unique domain. - Evan
Hello everyone, I would like to thank Evan, Tjiani, and Pari for bringing this issue into the discussion with such clarity. For a long time, some of us have been questioning ALAC's strong focus on new gTLDs. gTLDs are fundamentally a business matter, far removed from the interests of non-commercial Internet end users and much closer to the interests of search engine companies. The expansion to more than 1,500 domain names has created unnecessary complexity, serving mainly to confuse users and generate millions of visits for search engines because people are often unsure which gTLD corresponds to what they are looking for. I believe our mission is different, as our colleagues have consistently emphasized and as Pari has so clearly articulated. As Internet users, we should focus on issues that help ensure that our participation contributes to an interoperable, resilient, multicultural, and secure Internet. We cannot spend years discussing only the business interests of entrepreneurs and companies in the sector. Without narrowing the digital divide (access) and developing competent users (rather than merely passive and unskilled users of web applications), it is impossible to build an Internet that contributes to users' data sovereignty. Best regards, *Sergio Salinas Porto**Presidente Internauta Argentina - LACRALO/ICANN <https://atlarge.icann.org/ralos/lacralo>**Asociación Argentina de Usuarios de Internet <http://www.internauta.org.ar/>/FeTIA <http://www.fetia.org.ar/>**FUILAC- Federación de Usuarios de Internet de LAC <https://fuilac.org>**facebook: salinasporto <http://www.facebook.com/salinasporto> **twitter: sergiosalinas <http://twitter.com/sergiosalinas>**Mobi:+54 9 223 5 215819**"Ojalá podamos ser desobedientes, cada vez que recibimos órdenes que humillan nuestra * * conciencia o violan nuestro sentido común" Eduardo Galeano* El dom, 31 may 2026 a las 16:38, Evan Leibovitch via At-Large (< at-large@icann.org>) escribió:
On Sun, May 31, 2026 at 12:40 PM Alan Greenberg <greenberg.alan@gmail.com> wrote:
I have never been one of the "Let a thousand flowers bloom" proponents. It
was a foundational concept within ICANN and it obviously still is.
What it is, is a foundational flaw. In what ICANN calls "multistakeholder", some stakeholders -- the vested interests who seek to extract value from the DNS -- get to impose their will on the others. It's baked into the system. Two specific communities -- the providers and renters of domains -- have the ability to compel the Board (via the GNSO) to do their bidding over the wishes of governments and the public interest who can only advise. In most other fields of regulation it's the exact opposite; governments and the public set policy while industry advises but cannot compel. That absolutely nobody else has adopted ICANN's model for governance elsewhere clearly indicates that it hasn't lived up to expectations and isn't suitable for use elsewhere.
It is also clear that because of this power imbalance, the Internet world had generally bypassed ICANN's chosen methods and followed standards set by groups that better reflect the public interest. There are numerous commercial and noncommercial entities that have served the public better than ICANN ever could hope to do. You've even offered a number of excellent examples:
- IDNs
An utter joke; the world has left punycode behind. People everywhere can use search engines, commercial and non-commercial, to find what they want on the Internet using Unicode and its support of more than 170 scripts.
ICANN has already recognized the futility of its wheel-reinvention, giving up on the "universal acceptance" begging campaign when it cut off UASG funding last year. Community participants who seriously care about improving global access to the Internet, whether physical or cultural, seriously need to look at other vehicles besides ICANN to accomplish this. UNESCO and ISOC are good places to start.
- Community TLDs
This has always been a headache for ICANN, and it's an increasingly pointless one as alternatives develop. Such is the pervasive culture of exploitation, that well-meaning programs such as Applicant Support must erect so many barriers against abuse that they render themselves useless for their intended purpose. And based on Tijani's comments in this thread, not much appears to have changed in this regard since I left ALAC.
The ICANN community has always operated under a number of unstated and undeserved conceits. Among them is that having a domain can bestow an identity or cohesion that did not previously exist. Another is that it is reasonable to extract a monthly rental fee from communities (and their members) simply to maintain this identity. These days communities now have access to wonderful tools such as Mastodon, Discord and Reddit for bringing together common interests; they support subgroups, virtual events and real-time communications. More importantly they are free to use, in sharp contrast to the domain world's demand of monthly rent.
In this environment in which so many options exist for community building and development that are both cheaper and more effective, ALAC's efforts are not looking out for the best interests of communities. They are looking out for the best interests of domain landlords who want to extract value from communities.
- Underserved areas
See above. Another conceit is that "underserved by memorable domain names" has any relation to "underserved by the Internet". Again, the net effect of ALAC's contributions here is to find ways to have the "underserved" pay for domains when less-costly, more-effective paths to Internet access exist. Again, this is not looking out for the best interests of the underserved, but rather for the industry that wants to find and rent domains to them.
"When all you have is a hammer, every problem looks like a nail". Getting domains is not always the most effective answer to challenges of access or development; please stop acting as if it is. I await the first instance in which ALAC defines a use case in which better solutions exist to an access problem than getting a unique domain. - Evan
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
Apropos your note, below, I wrote a piece (in 2017) that has thoughts similar to yours... "Domain Names Are Fading From User View" https://www.cavebear.com/cavebear-blog/fading-domain-names/ --karl-- On 5/30/26 2:53 PM, Evan Leibovitch via At-Large wrote:
Hi all,
As ICANN plunges itself into yet another new-gTLD feeding frenzy supported by ALAC, perhaps it should be reminded again that this ever-expanding universe of 'memorable' domains is chasing an ever-decreasing number of users.
Perhaps it is worth contributing some perspective and research to the ALAC conversation when considering what end users really need from ICANN, and never-ending expansion of the domain space is not among those needs.
Consider that the share of Internet traffic that is initiated by user-typed domains is in the range of *4-7*%. This is down from *9-14%* in 2017 and *13-20%* in 2010. This is clearly a downward trend that has not yet bottomed out. Statistics are hard to come by, because most surveys lump typed-URL traffic together with bookmarks (in which your browser remembers the domain name but you don't) and dark social into the category of "Direct Traffic" to differentiate from lookups done via search, QR codes, etc. Sources are linked below.
This is what everyone here is fighting over: mid-single-digits of how people get to their Internet destinations. Yet ICANN operates on a much larger budget than orgs like IETF or W3C -- which critically affect how the /entire/ Internet works -- because of all the money sloshing around from sources such as fees for new gTLDs that -- provably -- nobody needs except ICANN itself and domain speculators. That domains are really rented rather than sold obviously provides ongoing revenue; whether this model benefits end-users is certainly debatable.
This isn't Internet Governance. It's heavily-conflicted regulatory cosplay, overseeing a tiny (and shrinking in importance) piece of the Internet infrastructure, designed to make money for some while diverting attention from real issues affecting end-users. Such issues include domain abuse that goes well beyond the industry-created ICANN definition, that even its non-political technical bodies refuse to acknowledge. Where is ALAC on this?
I used to think that the ICANN community bubble of self-delusion insulated it from the broader Internet ecosystem. I now realize that it actually insulates the broader Internet ecosystem against being infected by ICANN; this is why absolutely NOBODY is considering ICANN processes for AI or other critical Internet governance. The term "multistakeholder" now appears toxic outside the bubble, though its good intentions are already being reinvented under different names elsewhere.
I strongly suggest that At-Large re-evaluate its role within this bubble, to redouble its advocacy of genuine end-user needs, starting with an honest evaluation of what those needs are (as opposed to what we wish to be or what other communities wish us to chase).
That is: Just how can ALAC make life better for that 4-7% of the world that still depends on typed URLs? And what else matters?
Sources: ACM <https://dl.acm.org/doi/fullHtml/10.1145/3473343> Sparktoro <https://sparktoro.com/blog/new-research-dark-social-falsely-attributes-signi...> Mental Momentum <https://research.mental-momentum.ai/r/dark-social-marketing-attribution-2026...> Averi <https://www.averi.ai/blog/57-of-our-traffic-is-direct-unknown-what-it-means> GR0 <https://gr0.com/blog/what-is-direct-traffic> Sitechecker <https://sitechecker.pro/what-is-direct-traffic>
-- Evan Leibovitch, Toronto Canada @evanleibovitch / @el56
_______________________________________________ At-Large mailing list --at-large@icann.org To unsubscribe send an email toat-large-leave@icann.org
At-Large Official Site:http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
Well, the author himself the reference! Thank you Karl! Evan's is a reprise of a series of conversations coming on 10 years old. Way back then, Karl directed me to this writing. The facts are hardly controversial. Carlton ============================== *Carlton A Samuels* *Mobile: 876-818-1799Strategy, Process, Governance, Assessment & Turnaround* ============================= On Sun, 31 May 2026 at 07:45, Karl Auerbach via At-Large <at-large@icann.org> wrote:
Apropos your note, below, I wrote a piece (in 2017) that has thoughts similar to yours... "Domain Names Are Fading From User View"
https://www.cavebear.com/cavebear-blog/fading-domain-names/
--karl-- On 5/30/26 2:53 PM, Evan Leibovitch via At-Large wrote:
Hi all,
As ICANN plunges itself into yet another new-gTLD feeding frenzy supported by ALAC, perhaps it should be reminded again that this ever-expanding universe of 'memorable' domains is chasing an ever-decreasing number of users.
Perhaps it is worth contributing some perspective and research to the ALAC conversation when considering what end users really need from ICANN, and never-ending expansion of the domain space is not among those needs.
Consider that the share of Internet traffic that is initiated by user-typed domains is in the range of *4-7*%. This is down from *9-14%* in 2017 and *13-20%* in 2010. This is clearly a downward trend that has not yet bottomed out. Statistics are hard to come by, because most surveys lump typed-URL traffic together with bookmarks (in which your browser remembers the domain name but you don't) and dark social into the category of "Direct Traffic" to differentiate from lookups done via search, QR codes, etc. Sources are linked below.
This is what everyone here is fighting over: mid-single-digits of how people get to their Internet destinations. Yet ICANN operates on a much larger budget than orgs like IETF or W3C -- which critically affect how the *entire* Internet works -- because of all the money sloshing around from sources such as fees for new gTLDs that -- provably -- nobody needs except ICANN itself and domain speculators. That domains are really rented rather than sold obviously provides ongoing revenue; whether this model benefits end-users is certainly debatable.
This isn't Internet Governance. It's heavily-conflicted regulatory cosplay, overseeing a tiny (and shrinking in importance) piece of the Internet infrastructure, designed to make money for some while diverting attention from real issues affecting end-users. Such issues include domain abuse that goes well beyond the industry-created ICANN definition, that even its non-political technical bodies refuse to acknowledge. Where is ALAC on this?
I used to think that the ICANN community bubble of self-delusion insulated it from the broader Internet ecosystem. I now realize that it actually insulates the broader Internet ecosystem against being infected by ICANN; this is why absolutely NOBODY is considering ICANN processes for AI or other critical Internet governance. The term "multistakeholder" now appears toxic outside the bubble, though its good intentions are already being reinvented under different names elsewhere.
I strongly suggest that At-Large re-evaluate its role within this bubble, to redouble its advocacy of genuine end-user needs, starting with an honest evaluation of what those needs are (as opposed to what we wish to be or what other communities wish us to chase).
That is: Just how can ALAC make life better for that 4-7% of the world that still depends on typed URLs? And what else matters?
Sources: ACM <https://dl.acm.org/doi/fullHtml/10.1145/3473343> Sparktoro <https://sparktoro.com/blog/new-research-dark-social-falsely-attributes-signi...> Mental Momentum <https://research.mental-momentum.ai/r/dark-social-marketing-attribution-2026...> Averi <https://www.averi.ai/blog/57-of-our-traffic-is-direct-unknown-what-it-means> GR0 <https://gr0.com/blog/what-is-direct-traffic> Sitechecker <https://sitechecker.pro/what-is-direct-traffic>
-- Evan Leibovitch, Toronto Canada @evanleibovitch / @el56
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
_______________________________________________ At-Large mailing list -- at-large@icann.org To unsubscribe send an email to at-large-leave@icann.org
At-Large Official Site: http://atlarge.icann.org _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
participants (9)
-
Alan Greenberg -
bzs＠theworld.com -
Carlton Samuels -
Evan Leibovitch -
Karl Auerbach -
Pari Esfandiari -
Sergio Salinas Porto -
Tijani BEN JEMAA -
Wolfgang Kleinwächter