Hacker News new | past | comments | ask | show | jobs | submit login
The .blog Bait and Switch (chrisschidle.com)
305 points by cschidle on Nov 17, 2016 | hide | past | favorite | 178 comments

This seems to have become the de facto pattern with the new gTLD system. A company swoops in and buys a namespace for $200k (or more if there is a competitive bid), and they effectively turn into the cartel equivalent of the domain squatter.

You can register your new gTLD for anywhere between $1 and $100, depending on what I suppose they expect their target demographic to be willing to pay. Dot horse is more expensive than dot webcam, for instance, and I imagine dot lawyer is more expensive than dot lol.

However, if your domain name uses a word or phrase that happens to be in their "premium domain" dictionary, then they'll charge an elevated price based on heuristics similar to what yesteryear's domain squatters used. Some simple and concise domains can run hundreds of thousands of dollars.

  cheap.horse is $25/yr
  tiny.horse is $100/yr
  internet.horse is $1000/yr
  free.horse is "ask us"
Controlling the TLD is the ultimate form of domain squatting.

I wonder what will happen when some of these companies go under because they can no longer afford to pay off their debts? Will those that registered domain names lose them outright? Can someone else buy the gTLD and take them all over?

I wonder how much to buy "my.horse".

Then I can setup a sole mailbox "look", and have it auto respond to any messages with "my horse is amazing".

I wonder how may will get the joke.

I own amazin.horse using this very concept with the "my" subdomain as my professional email address. Closest I could get. :)

I have a .horse domain - I tried every permutation horse puns and in-jokes (my.lovely.horse, etc) but they were all taken by squatters or reserved for stupid prices.

Eventually I settled for sheep.horse, which is easy to remember but not much else.

When I was first building m.reddit.com, I put it under `reddit.horse`, mostly because I found it hilarious that `.horse` existed.

So disappointed that when I visit reddit.horse now it doesn't show a badly photoshopped picture of Snoo riding a horse...

I'm glad that someone snagged hoarse.horse and is doing more with it than I would have.

not.horse is available, too, if someone wants to tell that joke via a TLD.

And of course, bad.horse is exactly what you expect it to be.

Don't forget to traceroute bad.horse.

17 bad.horse ( 274.138 ms 272.734 ms 271.442 ms

18 bad.horse ( 281.595 ms 346.926 ms 277.790 ms

19 bad.horse ( 321.686 ms 486.714 ms 288.551 ms

20 bad.horse ( 280.425 ms 281.488 ms 314.627 ms

21 he.rides.across.the.nation ( 296.891 ms 297.092 ms 297.880 ms

22 the.thoroughbred.of.sin ( 380.358 ms 305.077 ms 312.953 ms

23 he.got.the.application ( 308.228 ms 373.024 ms 334.579 ms

24 that.you.just.sent.in ( 305.447 ms 452.741 ms 306.166 ms

25 it.needs.evaluation ( 342.778 ms 317.674 ms 313.240 ms

26 so.let.the.games.begin ( 387.610 ms 310.964 ms 482.971 ms

27 a.heinous.crime ( 320.908 ms 455.232 ms 351.113 ms

28 a.show.of.force ( 327.957 ms 328.891 ms 494.652 ms

29 a.murder.would.be.nice.of.course ( 406.810 ms 379.742 ms 406.548 ms

30 bad.horse ( 341.390 ms 341.207 ms 531.487 ms

31 bad.horse ( 415.444 ms 412.884 ms 345.962 ms

32 bad.horse ( 344.182 ms 345.273 ms 343.255 ms

33 he-s.bad ( 539.652 ms 352.428 ms 354.169 ms

34 the.evil.league.of.evil ( 354.151 ms 368.464 ms 409.535 ms

35 is.watching.so.beware ( 412.046 ms 397.060 ms 377.321 ms

36 the.grade.that.you.receive ( 369.120 ms 472.142 ms 369.329 ms

37 will.be.your.last.we.swear ( 450.581 ms 584.539 ms 369.829 ms

38 so.make.the.bad.horse.gleeful ( 443.618 ms 380.086 ms 430.089 ms

39 or.he-ll.make.you.his.mare ( 409.878 ms 409.638 ms 376.659 ms

40 o_o ( 394.767 ms 392.315 ms 394.865 ms

41 you-re.saddled.up ( 470.925 ms 574.203 ms 390.227 ms

42 there-s.no.recourse ( 399.117 ms 398.627 ms 418.105 ms

43 it-s.hi-ho.silver ( 614.293 ms 407.711 ms 410.689 ms

44 signed.bad.horse ( 419.605 ms 410.318 ms 538.911 ms

What a waste of precious ipv4 addresses! :)


Even more geek than setting up joke SSIDs for your neighbours to enjoy.

NSA SurvVan 83/B1-GER

That made my afternoon. Thank you.

Oh my. That is commitment. I love it. Thanks for pointing this out!

I am very disappointed that https://correct.horse/ does not have a strong password generator on it.

Let's also note that dead.horse exists, though I'm ambivalent as to whether it's as good as bad.horse.

When are people going to stop mentioning that dead.horse?

The Northernmost part of the Internet reachable by road.

I bought myey.es (Spain TLD) a while back so I have look@myey.es. I'm a hobby photographer, thought I'd use it.

Never did. If there are any photogs here who want it, let me know.

Hello! :)

Email me - my name is John Noble and my email address is:

[first initial][last initial]@[first initial][surname].net

Hi. Sorry for the late reply. You want? Can't figure out if I can PM here on HN...

Nice! I bought jungle.horse and put my Donald Trump text to speech engine there. These domains are so ridiculous.

I recently registered butt.chat. $30.

I think you'll find that the universe pretty much covers everything.

The ghost of Kurt Godel would like a word.

I had to pony up about $200/year for belinda.horse for a friend. Definitely not $20.

This is worth it at any price.

surely my.lovely.horse :-)

Ah ... Father Ted! I was thinking the same :)

Tangentially related from an Irish perspective:


haa, when this came out I reg'd horseoutsi.de and set it to play the video fullscreen/loop.

also this email address:


You Sir, win the internet for today. :)

Thank you for the chuckle.


My money on the first big causality of the new gTLD program is http://Rightside.co

Looking at ntldstats.com, it would appear that they can't manage to break +250 new registrations per day for about 40 domain extensions that they own and operate. Even worse, it looks like they have been going negative on new registrations now: https://ntldstats.com/registry/Rightside-Registry

The user interface has also gone to shit on their site Name.com. I'm a developer and can't for the life of me understand their backend anymore. Google Domains crushes them design-wise.

Agreed. Name/Enom support has gone to complete sh*t. They are putting all the money into marketing crap new TLDs.

They stole a domain from me too. :(

I was trying to get internet.party for a browser extension I'm working on and I discovered that any gTLD with internet in the name is like $60,000. Absurd.

That seems about market value to me. A lot of it is anybody's guess, but that passes my sniff test.

There is a process when a failed TLD is handed off to a stable domain registry. It's only for a few years though.

Many of these new TLDs will disappear. There is no demand.

this is why I ended up with spherical.fish - memorable, my first name is buried in it, and cheap.

also, there's no .cow gTLD so I can't do spherical.cow. Spherical.horse is squatted already.

I hope your name is Al and not Eric.

Why not Erica?

it's al.fi

I bought Free.TV and it was only $100,000 - per year!

If anyone wants to use it, let me know

I've never understood why some random company is allowed to arbitrate the domain registration and make $$$ out of it. Feels like it's kinda not in the spirit of the internet.

It is a complete state-sponsored racket. There's no way around it.

You pay an upfront fee to ICANN and probably some kind of recurring tax and you get to create a TLD. You then sell this TLD for higher prices than ICANN and run a very simple service that processes the transactions.

You then take anything lower than a 5 or 6 letter domain that is a word in the english language and define it as a "premium" domain and charge obscene prices for those domains. The fee you charge for the domain based on this system is recurring. So instead of buying milk.com for $10k and selling it for $12k you already own milk.blog by default and if someone wants to register it you charge $500 a year.

NOT TO MENTION this whole system is like a slap in the face to a certain kind of "domainers." People who buy domains based on their perceived value, park them, and then try and resell them for more money. There are many millions of dollars wrapped up in this other, also stupid racket. These companies got so bold as to get their domains "valued" and then bet on the value of their company based on their holdings like some type of stock filled with mortgages. This new system is like cutting directly into that strategy but I guess ICANN gets some of the money this way.

I have a last name that's extremely uncommon. MyLastName.com was available for many, many years. I thought about buying it a few times in the last 20 or so years but I don't have any use for it so I didn't. Two years ago it was bought by a domain squatter who was probably iterating through the white pages. Now that I can't have it I want it (I guess human nature) but they are asking like $2k for it. It seems somehow "unfair" that "my" last name is being "held hostage" by some faceless company that's just trying to make a quick buck.

Start a business with your name, trademark it in a certain domain, file a claim with ICANN. Long process, but might be cheaper. :)

Almost as expensive (UDRP claims start around $1500), and more importantly, it'll fail. WIPO isn't that dumb -- they can look up the registration date of your trademark, check WHOIS, and determine that the domain was registered before your trademark existed.

> and determine that the domain was registered before your trademark existed

It's a shame they don't determine that the domain was squatted before your trademark existed. But it seems in their interest to give the benefit of the doubt to the one buying many, many domains yearly.

This is, in general, about as sleazy as the squatter.

Exactly. And definitely not guaranteed to work. Just pay a visit to nissan.com and read about the scumbag efforts of the car company to swipe that domain from some random dude who's owned it -- and isn't a squatter! -- for years and years.

mylastname.com was already gone in 1999, so I had to settle for mylastname.net. Too bad it didn't occur to me to secure it in 1996.

If Tucows is willing to sell they certainly aren't publicizing it.

Spare a thought for us Cooks of the world. Four letters, a noun and a verb...

After the gTLD explosion I thought I could finally snare a decent one but they all went into the $1000+ category immediately, excepting niches like cook.republican, cook.accountant etc.

Luckily some did eventually come down, I managed to get cook.run for a more reasonable $50/year

I managed to have a bit of fun with .party and .science.

firstname.party is great to have.

Welcome to rentier capitalism.

Sounds like ICANN needs to replace the fee + tax scheme with Dutch auctions.

So the organization that provides an essential service to the internet gets decent funding and a bunch of squatters (sorry, "domainers") get shot down.

Sounds great to me.

There has got to be a hedge fund who buys up domains

DNS is a rotten system. From censorship, to ddos attacks, to squatting, to privacy issues with whois, to messed up financial incentives. It's a shame that DNS is a critical part of the internet infrastructure. We should work on getting rid of it.

I guess it's two distinct issues to me; the tech stack that is hobbling along, and the people that control access to it (ICANN, gTLD registries et al I guess).

The latter is the really corrupt, broken thing here, and we should be able to swap that out.

what's the alternative?


Namecoin is probably the largest, and provides .bit. IPFS has IPNS, but that is just subdirectories of a user hash, there is no pretty naming there.

There is also tech like what is behind .onion, where you can generate names until you get one you like, where you can devote more resources to getting prettier names at the cost of electricity and hardware to do so.


GNS (GNU Name System) or something like it.


"Unlike DNS, GNS does not rely on central root zones or authorities. Instead any user administers his own root and can can create arbitrary name value mappings. Furthermore users can delegate resolution to other users' zones just like DNS NS records do. Zones are uniquely identified via public keys and resource records are signed using the corresponding public key. Delegation to another user's zone is done using special PKEY records and petnames. A petname is a name that can be freely chosen by the user. This results in non-unique name-value mappings as www.bob.gnu to one user might be www.friend.gnu for someone else."

A Blockchain.

Poe's Law, Bitcoin sub-clause, strikes again

It is a distributed ledger; calling blockchain here is certainly not crazy-talk.

I mean, I was being a bit facetious, but there are actual proposals for blockchain backed DNS.

[1] https://github.com/okTurtles/dnschain

Yes, this is an application where the blockchain wouldn't necessarily be stupid.

I don't see it as necessary, to be honest, and given the frequency with which people lose access to their wallets/have them stolen, I'm not entirely sure the bulk of domain owners would want bearer instruments to be the sole way to manage (or lose) their domain.

It's not much different from people hoarding dictionary domains and selling them for insane amounts.

But I understand your point and the whole system is rotten. First come first served is no longer viable.

What is viable though? First come, first served enables domainers. Jacking up annual domain prices can be harsh on businesses in poorer countries.

Public auctions for each new and released domain might have merit, but will still attract domainers and preclude many fresh companies.

I never said I have a solution I just know that the current situation is not good either.

Anyone can run their own DNS root, but good luck getting people to use it. And ICANN has a history of deliberately colliding with TLDs used by competing services.

Browser vendors and OS makers should do an end run around ISPs and ICANN. Just agree between themselves ( what are they like 10? tops) and fork the current registries. Then issue domains themselves.

What are the plebs gonna do? Go the the non existent alternative?

Google already controls chunks of every platform involved, some of them are very large, (network, domain registry, OS, browser, search engine, web hosting/content delivery) and eats away the importance of domains bite by bite. Non-tech people do not type in domains anymore, they search for them. URLs are getting hidden away further and further.

The DNS resolver is configurable, not hard coded. Forcing a switch would require OS vendors to ignore the resolver issued via DHCP, which would not work on many corporate networks that block outbound DNS not via their resolver.

Maybe then just prefer if reachable? Soon enough it's the new standard.

Yes I think any new system would have to see the old one grandfathered (give the current incumbents time to find something useful to do instead of jetting off to conferences all year).

Browsers/OS would be reasonably easy in most circumstances, but there are embedded devices, load-balancing configurations, other esoteric uses for DNS (text records, mail etc) that'd have to be considered as well.

An attack surface is ISPs and Governments strongarming them to use their own DNS roots which can then be plugged into the "new" root servers conforming with the theoretical browser vendor, os vendor alliance.

Pick your favorite:


There are others. But no, they don't become the new standard.

Say you owned ensiferum.com. Should you be allowed to sell subdomains and make money out of it? That's really all they are doing.

Because running a global TLD is a non-trivial and costly public service to provide, so the TLD sponsor is allowed to monetize it.

It really isn't that hard (source: designed and ran the development team that built the initial .name platform, before we outsourced it to Verisign; Verisign now owns it outright), and you can contract any number of operators to do it for you.

There certainly is the issue that most of these TLDs aren't very interesting to most people, though. With low demand, I certainly understand why they want to be able to milk the few names that people find attractive.

Here is a message I received from Gandi.net regarding the .blog domain:



You are getting this email because you have performed a pre-order with your handle {redacted} for the domain name {redacted}.

Unfortunately, this order is currently in error, because the domain name in question has been defined as Premium by the registry. This decision is of course beyond our control and frequently corresponds to a list of common and/or connotative words that allow registered to apply special rates to them.

This means that in your case, we have placed the order in error, because the price that you initially paid is no longer the price that the registry is charging for it.

If you would like to know the price, in order to eventually purchase it, then reply to this email, so that our support team can obtain that for you from the registry.

As long as you are considering this, do not cancel your order, since this will cause you to lose your place in the waiting list.

If, however, you no longer wish to register the domain, you can cancel it from the link provided below, and your prepaid account will be refunded for the order:

Good on Gandi for providing a straightforward response. Living up to their motto.

This is very common with these new TLD's. At least they didn't give it to you for a year then jack up the price. I had jacki.party as a fun domain for my wife that she didn't end up using, but after a year the price went from like $29/yr to $2,495/yr. because they decided it was "premium." There was nothing my registrar could do because it was the domain registry who decided that, so I just let it lapse.

Why is a very specific spelling (note the lack of "e") of a not terribly common name so expensive? No clue, but I'm guessing the same forces are at work (i.e., $$$).

What registry was this? That's not supposed to be possible -- registries are only supposed to be able to designate domains as "premium" if they aren't registered yet.

I suspect that any domain that gets used becomes "premium."

The problem is that it only takes a few hits at $2500/yr to make it worth it to screw a bunch of people out of an already overpriced domain.

So if you serve 15,000 people with .party domains at $29/yr each, you only need 175 people to pay $2,495 and you've already made more money.

I don't mind paying more for a domain, I might even pay a measurable amount of money every month for the right one, but trying to insist that jacki.party is worth $210/mo is bananas.

This is the bit that really gets me. $2,495 for a domain I can understand. If somebody is willing to pay that, more power to them. But once they've bought that domain, pricing should be at the registrars listed rates. It's unconscionable to charge a premium price for renewal, and even more so to jack up the price after it's already been purchased.

Yikes. Just extended my lastn.ame domain hack to 2026.

That makes me feel very good that I locked in .party for 10 years and $3/year with Gandi.

That sucks!

Was aware they could do this but totally BS!

They are doing this a lot with more and more of the new TLDs. Defeats the purpose of even bothering with them. Create more, to open up the domain space, make it cost prohibitive, so we're back to where we started.

> Create more, to open up the domain space...

Adding TLDs doesn't open up the domain space; it closes it. The domain name system was designed to expand downwards, by adding more hierarchies. Downwards is where the space is. Expanding upwards is limited.

The racket ran out of space selling keywords inside .com, where this was originally meant to be confined. Once they finish selling keywords at the root, that'll be it.

I don't know. I don't think nested hierarchies is a thing. I mean, I guess you have the "Go Network" go.com example (disney.go.com). But I think the ship has sailed for deeper hierarchies. I really don't want an email address that is: taftster@foo.bar.baz.example.com

I didn't make any judgement on whether it is a good or bad thing. The fact remains: don't delude yourself into thinking that adding a TLD free-for-all opens up the space. It doesn't. It closes it.

The current "problem" stems from the fact that .coms are full. Once the TLD space is full, the same problem will remain. Everyone who wants one cannot have a memorable name in a flat namespace. There are too many people and there is too little room.

When that happens, I fully expect that as the value of available domain names go up, people will invest in expensive domains (blah.com) in order to treat them as "2LD"s and sell its 3LDs (xyz.blah.com). But it won't happen while there are 2LDs available for the exact reasons stated above (nobody wants a subdomain when they can have an actual domain).

I've also got stood up with fashion.blog. Full refund and no invitation to auction. I haven't even bothered to ask for details, because that was their way of assessing and freezing the domain names that people would be interested in buying.

> because that was their way of assessing and freezing the domain names that people would be interested in buying.

Indeed. Seems like it would be easy for them to use the auction to automatically see "300 people tried to claim this domain, it must be a good one, let's make it premium."

They wouldn't need to spend a penny doing market research to select the premium names themselves.

That's the feeling I got as well. The answers were really dodgy too. But such an ethically challenged practice would reflect poorly on the WordPress brand and you wouldn't want your low level employees to know about it since they could easily leak it. So maybe the CS reps weren't dodgy so much as trying to keep customers happy with very little information available to them.

Maybe one way you could do it is create a tiny subsidiary to manage the reservations/registration. The subsidiary can be created under the pretenses of shielding the parent company from legal liability so you can both limit the team size, the scope of their purpose and isolate them from the parent company.

Build the backend to provide information on reservations but accessible only to subsidiary financial management/officers and parent company employees whose wealth is tied to the success of shares in the company. Officially only share the total number of reservations and financials with the parent company via something documented like email. Then the backend reporting is decoupled from the reservations that come in for early access. If you setup a distinct founder program and establish enough time between the reservation announcement and when you give normal employees early access you don't even have to worry about your employees snatching up the most profitable ones.

The first the CS reps hear about it is when a reply/complaint comes in from one of the cancellations. When asked their supervisor only knows that the name came over as one of the early access reservations from the parent company. The CS reps don't have enough information to draw conclusions because they don't get access to numbers and an early access employee program provides plausible deniability if the reps start asking questions.

That was kind of a fun mental exercise. I've seen some real cloak and dagger stuff from people but those were always at small companies. I have no idea how plausible any of that would be, especially in companies this big. But it would be interesting to see what would happen if we had a fairly unique but uninteresting domain that wasn't obviously gibberish and a a lot of us put in a reservation for it at random intervals to create a perception of interest and popularity.

As a fellow very-common-name sufferer, I long ago gave up any hope of getting a vanity domain name (how I wish I had paid the $25 back in 1996). Most of the new TLD seem pretty scammy to me, the costs in obtaining and running most obscure TLDs far outweigh what could be got by paying customers actually waiting to run sites off them.

I suspect that a lot of them are just hoping that enough large companies buy up kfc.blog, walmart.porn, etc to make a return on investment. Fleecing people who actually want the domains is just icing on the cake.

I am somewhat of a hypocrite, I run my blog off https://sheep.horse/, but only because I thought that the .horse TLD was silly enough to be worth supporting.

At a USD200,000 price for a new TLD, it takes 2,000 domains at $100/year to break even. Of course there are other costs associated to running a TLD, but they probably don't scale linearly.

So on a 5 year term, it might even be a small profitable business.

> As a fellow very-common-name sufferer

When I started reading this post, I clicked the get.blog link first and wondered if chris.blog might be available. I assumed not and didn't check before going back to the article and seeing the topic domain, hah.

That's exactly it. Brand protection companies like MarkMonitor will buy these up as a matter of course to protect their clients, and they'll do so at the inflated Early Access rates to boot. By the time the domains are released to the public, these companies have already made their profit.

At one point even their clients will realize that the value of owning {name}.{everything} is more finite than the number of possible values for {everything}. When this happens, TLD racketeers will be sitting on a well oiled money burning machine, because only few of those who are attracted by that kind of business are good at keeping expenses low in times of plenty. The end game for many squatting-TLDs will be an existential pricing standoff between the dying registrar and the very low number of users that were foolish enough to build a long term identity in one of its namespaces.

I registered "steve.org.uk" back in 1999 and it cost me way more than $25. In fact I hate to think how much I've paid for it over the years, it's been a while now!

(Recently managed to register "steve.fi", as it expired just a few months after I relocated to Finland. That was a happy coincidence. I guess that means I have two vanity-domains!)

> only because I thought that the .horse TLD was silly enough to be worth supporting.

Not sure how silly it is but there are some silly amounts of money involved in horse racing.

This seems like it has gotten out of control - it is a basic key/value lookup, and we've created a rent-seeking, global series of registries to further suck good money from people.

I worked briefly at a registry - it is money for jam.

I don't know if the DNS protocol and all the related ones are fundamentally broken or need fixing, but it seems to me the problem is ICANN and the web that spreads out from under it.

Why can't we just create an alternate ... distributed naming system, except without the rent-seeking/government ownership attached? I mean, this focus on how important the last bit is is ridiculous, the namespace doesn't have to be artificially restricted, beyond what the protocol might require.

I guess I have a simplistic view and probably lack a bit of history, and there must be something preventing it. Do the top-level servers collude, and help to create something like ICANN, or was it taken out of people's hands etc?

> except without the rent-seeking/government ownership attached?

Because people like money. This TLD business literally invented a new economic activity, and one with great margins. It doesn't really hurt anyone, has little to no cost, and makes a pretty penny. People With Money love stuff like this; if tomorrow they could invent a marketable TLD system for physical addresses, phone numbers or anything else, they'd jump at the chance.

This is just how the capitalist world works: stuff exists to make money, everything else is a bonus.

Agreed, I guess the way that works in general is just disappointing in that it doesn't produce anything meaningful or advance anything (to my mind). It's spinning wheels and burning energy for no appreciable reason, to make some people rich.

What is the end game on that though? Everyone for themselves, buy a shotgun, and hick USA?

Legal/Govermental/Regulatory issues, someone has to deal with it. No single organization is going to deal with every country and their various legal issues. Registries and Registrars help abstract a lot of that (ICANN definitely doesn't want to deal with it). I don't disagree with the ease of storing a few million records or more. That is not the real overhead and is in reality only a fraction of whats involved.

Yes the added verification etc on top of the key/value lookup is the meat of it I guess. A complicated problem, I guess it'd be nice to somehow remove the money factor from a critical function of the net.

I was keeping a close eye on .family for quite a while - I'm getting married next year, so mysurname.family would have been a good domain to have both for wedding planning and personal email accounts.

When it finally came out, I eagerly went to my favourite registrar to try and buy it, and found that it's listed at over £1000. It's not a particularly common name either. Out of interest, I tried a few other surnames off the top of my head, and it seems that pretty much any domain that could be somebody's surname has been bumped up massively.

Of course, they have every right to do that, but the people buying these domains are individuals - very few of these domains are ever going to have a business use. To me, it looks like they've done nothing more than price themselves out of the whole point of their TLD.

Instead, I'm going to wait out the expiry of the .com, which isn't being squatted as such, but it is owned by a company for a single software product of theirs which looks like it was written in the 90s and requires you to send a cheque if you want to buy it. The last update was helpfully reminding visitors that the app wouldn't be compatible with the upcoming OS X Lion. One day, they'll stop renewing that domain...my time will come.

My family name is $20/year!! Now I need to decide if it's worth supporting this cartel...

I believe this new gTLD registries will manage it to finally sink their own boat. They are behaving so greedy that this new gTLD will never become popular. But then it wont take long until the first registries will close shop and this will be the end of this whole ngTLD scam. Because nobody will invest money to market a domain which might not exist anymore in a few years.

There are some proposed safeguards built into the rules to prevent this meltdown caused by failing registries. But looking at how the industry currently behaves I don't think the will to stand to the rules and grab as much money as they can and leave the sinking ship asap.

When the .beer tld came out I tried to buy "california.beer" as soon as domains were available to the public. The site wouldn't let me buy even though it was listed as available. It was finally registered by a squatter months later. I wouldn't be surprised if the good names were kept in reserve and sold through special channels.

I'll tell my TLD story. I was following .family for a long time, since my current "primary domain" is horrible and hard to say. <family name>.family sounded pretty awesome.

There was that whole pre-registration BS but it is so convoluted and expensive I just waited till it went live. It did, and when I checked on my last name, it was set at $1200/year for a normally $30/year domain. This was when I was introduced to the idea of reserved domains in this new TLD system.

So I emailed them and put forth my argument, that this TLD in particular is a bit unfair, considering I don't control what name I'm born with. I didn't think it would have any impact, because $$$.

One day I got an email and they offered me to buy it for a one-time price of $1200 and then the renewal would be set at $30. I said no. There are so many TLDs now that it's easy to get a common-word, easy-to-say one.

> that this TLD in particular is a bit unfair, considering I don't control what name I'm born with

To be fair, it would also have been a bit unfair to the next guy with your last name if you had taken it.

I signed up too and then saw the ridiculous prices and policies. Happy to stick with .com for 10 bucks.

Like the author I would also have no problem in paying a few of several hundred dollars as an initial registration one time fee, for a domain like chris.blog.

I would really prefer using chris.blog instead of buying a .com domain like this-is-chriss-development-blog.com, just because any shorter .com domain is already taken.

Right on. While I said "I appreciate a good name", I'm truly a sucker for a good name. Branding is huge, even when it's a personal brand.

So yes, a one-time fee of $250-$1,000 feels ok to me, especially considering what Automattic paid for the TLD.

It's still possible to come up with vanity domains using common TLDs. You just need to have some wiggle room (: I replaced a single letter in the shortened version of my own name and got a .net for it just a few months ago.

What is the long-term security/availability of those new tlds?

If I register a .de domain today I am pretty sure that it will last a long time. But some of the new ones have a pretty sketchy vibe to it and I am not sure if I'd trust them enough to build a business on top of a .fancy domain

The Registrar's policy language makes so many exceptions for themselves that you'd have better odds playing roulette:


"We are not bound to select a winner among RFP respondents and may hold the Domain Name for future allocation."

"We may activate and allocate to ourselves or third parties certain Domain Names (plus their IDN variants, where applicable) deemed necessary for the operation or the promotion of TLD."

"the Registry makes no guarantees or warranties as to whether any applicant will be successful in registering any Registry Domain Name."

In 2011 I paid for and registered o1.io (I really wanted 01.io, but back then it was not available to register domains containing only numbers on .io).

The .io NIC web interface is really horrible and I ended up in an inconsistent state after making the "horrible" mistake of clicking the back button. Even though my Paypal account got charged, I received an email confirming I was the owner of the domain, and so on, I couldn't access my account.

Next step I took was contacting them. I got ignored and after 7 days later they made a transaction reversal on Paypal and I never heard a word back from them, even though I have tried to contact.

Some time later they made the o1.io domain a reserved one and so if I wanted it now I would have to pay thousands. What a shame.

At this point, this looks ever more akin to charging people to have their phone numbers listed in a phone book. If you don't know what a phone book is, it was the DNS-equivalent of the 1960s. You know those plastic-wrapped bricks of yellow paper that sometimes appear on your doorstep, that you just drop right into the recycling bin? Those things are actually phone books.

The phone book is rather simple. Your individual given name or your business dba name is the dictionary key, and the phone number is the dictionary value. Your dictionary entry is automatically included, unless you paid to be "unlisted". That would keep your phone service provider from printing your name-number pair in their phone books.

This approach is still viable. Every state has a database of corporate names and dba registrations. The secretary of state for Illinois, as an example, could make aaa-plumbing.corp.il.us resolve to the server address for "AAA Plumbing, Inc.", if that business chose to disclose it, or to a landing page otherwise. Likewise, zzz-towing.llc.il.us could resolve to the address for "ZZZ Towing, LLC".

The US Patent and Trademark Office could do the same. Let lego.028.tm.us (and lego.009.tm.us, lego.041.tm.us, etc.) resolve to LEGO Juris A/S Corp's servers, so you can see a site about plastic building blocks, and let lego.005.tm.us resolve to American Lego Group Inc's server, so you can see a site about soy protein supplements.

If you are Foobar LLC, you really shouldn't have to register "foobar.com" and "foobar.net" and "foobar.cc" and "foobar.co.uk" and "foobar.biz" and "foobar.info" and "foobar.sucks" and "foobar.pooky" and "foobar.booger". It isn't really elucidative, and does not help discoverability to have so many TLDs. The problem is that all the simple names have been mined out of the original TLDs, and there is no mechanism to resolve name collisions between Foobar LLC out of the US and Foobar GmbH in Germany. Whoever grabs "foobar.com" first gets to keep it, and the other is stuck trying to make some DNS entry that somehow relates to the company name.

Adding a bunch of niche TLDs helps a bit, but not if the registrar can charge $1k/year for names. That's just ridiculous. You're not going to end up with all the Johns on the Internet getting their own john.something domain name that way. You just let the richest John on the Internet to have all the john.anything domain names.

"If you don't know what a phone book is, it was the DNS-equivalent of the 1960s. You know those plastic-wrapped bricks of yellow paper that sometimes appear on your doorstep, that you just drop right into the recycling bin?"

You mean monitor stands?

To sum up, the .com TLD should likely be split up into separate TLDs, corresponding to the Nice classes for international trademarks.

   1 .chem     16 .book     31 .farm
   2 .paint    17 .flex     32 .bev
   3 .clean    18 .skin     33 .booze
   4 .oil      19 .build    34 .smoke
   5 .drug     20 .home     35 .ads / .biz
   6 .metal    21 .ware     36 .fin
   7 .mech     22 .rope     37 .con
   8 .tool     23 .yarn     38 .tel
   9 .sci      24 .tex      39 .mov
  10 .med      25 .cloth    40 .mat
  11 .env      26 .lace     41 .edu / .fun
  12 .wheel    27 .floor    42 .tek / .law
  13 .gun      28 .toy      43 .hos
  14 .bling    29 .meat     44 .doc / .eco
  15 .music    30 .food     45 .soc / .sec
The clear correspondence between domain and Nice category would make it easier to evict name-squatters that are not engaged in the relevant type of business.

Look up the rules for `.ltd.uk` :).

Also, for that matter, .plc.uk, .ac.uk, .sch.uk, .nhs.uk, .police.uk and .gov.uk. All of them are available only to organisations that already exist in a particular context, and only with their 'real' name.

That doesn't prevent some collisions around characters that are legal in business names but not in DNS, of course, and it also doesn't mean that anyone actually uses the .ltd.uk and .plc.uk domains: I don't think I've actually ever seen one in the wild. I've seen all the others, though.

The ICANN whois also does not give any information, there is just a copyright message from Nominet UK


Maybe the domain got bought during the sunrise period by a copyright holder!?


The reply I received from get.blog support (last image in the post) stated that "chris.blog ended up being part of the list of reserved domains that are not available for registration". So I imagine they'll either keep it internal or sell it as a premium domain.

Yeah right, you had mentioned in your P.S. two other names dave and matt, I did a whois for those two and got some information:

https://whois.icann.org/en/lookup?name=matt.blog https://whois.icann.org/en/lookup?name=dave.blog

So yes it seems it got reserved. I wonder why too.

P.S. I'm a chris too ;) that's why im so interested in this, I own chris.lu but as it is a country tld this is not so good from a SEO point of view. I want to change if I find a better one, but haven't found one yet. Have fun with your new chris.lol :)

I wanted a domain for a tax industry blog. I really wanted digital.tax, but it was held back as premium. Then, one day I was browsing 123reg and noticed it was available. I quickly looked at other registrars and it was still labeled as either taken or reserved. I placed my order and got the domain! I'm not sure how common these discrepancies are but I guess they do occur, so perhaps all hope is not lost.

Fwiw I snagged lech.blog at list price (so $30 + 220), I wouldn't pay a cent more, but my main problem is that I still have no access to the domain and, unless I am provoven (or refunded) otherwise, my domain registartion runs from 28th of October. Some email were sent from an unverified, unencrypted, unrelated domain - I nearly sent them to spam. They really need to sort out their internal processes.

Same happened to me. They should expect a class action lawsuit as even though the list was not finalized, they should have said something.

I had something like this happened before with WordPress.com - I have nikolay.wordpress.com, and then an employee took it way from me. After making a long discussion on TechChrunch about it, the employee decided to give it back to me.

So, my point is, there's nothing new under the sun here - obviously Matt is the old Matt having double standards, and not knowing what's acceptable and what is not in the digital world.

Also, he's going against his investor's interested, because some of these domains could've brought in millions of dollars, but he wants to do his buddies favors, and not honor what was promised.

Edit: Maybe if there are several of us, we should try to see if we can have a class action lawsuit. Our hundreds of dollars were used by the Automattic for 3 months, that's a lot of money collectively, and they should not get away - there was no mentioning of any Founders when they took our money!

> Perhaps it's not fair to call this bait & switch. Really it was bait & refund

The title of this article was a bait & switch.

I was being generous when I wrote that line. But I maintain that clearly stating it will go to auction then pulling the auction is a "switch".

In my case, my surname TLD exists like XXX.{surname}. So I bought it w/ my first name ==> firstname.surname . It's cool and I will keep it as long as I can.

I agree with the OP. They seems to gather the popularity info by accepting pre-register requests.

Is there any policy to ban this behavior? Shouldn't the TLD be publicly opened?

technically its their property and theirs to do with as they please. though i disagree with the behavior.

nothing is stopping you from reselling *.whatever.com and so forth.

i wish i had my whole name as a domain, thats pretty awesome for you!

"Perhaps it's not fair to call this bait & switch. Really it was bait & refund"

The article title was a bait and switch! Honestly, I don't know much about the domain registration business but I had always imagined it worked even suckier than the way he describes it working in this case.

When you hand over $250 and they hold it for 3 months with the promise of giving you a shot at an auction, then they cancel the auction... I still call that a "switch".

Still haven't received the refund, by the way.

I had this happen to me as well. I registered a domain directly with get.blog on the day that they first started accepting pre-registrations. I got an email a couple of days ago saying that I would be getting my money back, nothing about an auction. I asked for details because I too was expecting an auction. My response also came from the same guy ("Ran") who said that it wasn't available when my "application" was processed.

The twist in this case? The domain I applied for is now pointing directly towards WordPress.com. So get.blog, owned by parent company Automattic, took the domain which had been made available for the public, and gave it to WordPress.com. I wonder what ICANN will have to say?


Who wants to help me create an .arse TLD? Who's with me!

I've long felt the only sane approach is to define country-level TLDs (via the UN or something) transition everything under them, and then let each legal jurisdiction handle (or mishandle) their own shit.

While it's far less-convenient for global companies and "ignoring borders", it decentralizes the system. Compartmentalizing failure and allowing some kinds of competition.

That would be lovely, but I don't see any way of such change actually happening. The current hierarchy is extremely entrenched and heavily supported by the most powerful entities on the net.

I wonder if using the validation of the DNS record and all the surrounding nonsense is really a useful function though.

I mean, if we created a new distributed system, and grandfathered the old one, how can the current incumbents stop it? The control over a name pointing to an address could be done with a blockchain, I'm hearing.

So we move to a new naming convention. An arbitrary prefix on the end of a name shouldn't mean $$$$.

I'm saying we (techies anyway, not money-grubbing VC's), can make the decision for them, en masse. When they say "But, we need to make money from this!", we reply, "No, we need a naming service for these internet addresses. Go get stuffed."

The finer details of how it all works can be left till later.

> I mean, if we created a new distributed system, and grandfathered the old one, how can the current incumbents stop it?

That's a weird way of putting it. Anyone can create DNS roots based on any hierarchy they want, no-one will stop you from doing that. The meaningful question is how do you get anyone to use your fancy new roots?

You mean, like .co.uk, or .us???

Basically, it'd mean the (virtually impossible) task of deprecating the non-country TLDs with X years before they stop working, and saying: "Okay, .us is managed by whatever infrastructure the USA sets up, and .uk is managed by whatever the UK sets up, etc."

So the US might choose to allow cocacola.us, while the UK might have on a grouping scheme like cocacola.co.uk, and meanwhile North Korea refuses outright or insists on cocacoala.obeys.dear.leader.nk, whatever.

Countries can break up, dissolve, or form too. I'm not sure that making all domains depend on things that exist in the real world is a good idea. They should depend on abstract ideas that are less liable to change and can be resold easily to different operators.

> Countries can break up, dissolve, or form too.

Not that often, and even when they do it often means very real issues in terms of governance of internet infrastructure.

> I'm not sure that making all domains depend on things that exist in the real world is a good idea.

But domains already depend on things which exist in the real world! Not only are they leased by real-world entities, but even the names of domains already refer to countries, companies, brands, products, activities, clubs, and people.

The real difference with what we have now is that it's a weird global monopoly that has shown it's willing to erode organizational principles in order to make a buck.

Splitting it up by sovereign nations is actually moving back closer to the decentralized principles of the internet, avoiding a single point of (political) failure, and creating subsections which can evolve separately and cover for one another.

Ex: Turkey's emerging dictatorship refuses to let anyone register erdogansucks.net.tr ? Fine, go register erdogansucks.net.de , and advertise with that instead.

There is .su (Soviet Union) gTLD.

As SapphireSun said, it's because .su is based on "abstract ideas that are less liable to change"!

What will it take me to register the Holy Roman Empire or Ottoman Emprire (.hre/.ot or .oe) TLD?

I fail to understand the purpose of these non-standard TLDs. Nobody is going to remember or type in such a domain, so you stuff is only going to be found by a web search - where a standard domain would be better. So why not just register a .com - which are also cheaper?

There's another serious problem with newer TLDs: there are millions of websites out there that "validate" email addresses against hardcoded lists of known/valid TLDs. Anyone who spends $1000+ on a fancy domain is going to be extremely frustrated when every other week or month they stumble on another site that refuses to accept their email address as valid.

The only safe domains for everyday use are .com, .net, and .org - and the well-known country codes that have been around for at least 10 years. Anything else is going to cause you headaches, and your "investment" for a short memorable domain name is not going to feel worthwhile.

Plus, the registrars are scum with these TLDs. Just stay away.

Yup, and whole companies are blocking new TLDs. Too much spam.

> So why not just register a .com

Because the name you wish to register is not available as .com domain.

> Nobody is going to remember or type in such a domain

I don't think this is true, I live in Europe and we have tons of country tlds, people use and remember them.

Also is chris.blog really so much longer to type then chris.com or a lot harder to remember? I don't agree.

At first it might feel strange to type something else then .com but as I said in europe we have country tlds and it works well.

I may be wrong and just speaking from experience, but Americans were really trapped in the .com/net/org for a long time. I think, because that's all we were exposed to... I didn't realize the depth and usage of CCTLDs until I moved to Europe, and saw them everywhere. When I gave my mom my first work email she added a .com after the .at at the end, and wondered why it never got to me.

> Also is chris.blog really so much longer to type then chris.com or a lot harder to remember? I don't agree.

It depends on the audience. If you have a non-technical audience or you are e-commerce a .com is basically a necessity still.

Also in Europe, but I've seen adverts using .travel domains (which are a bit older than most), like egypt.travel and ecuador.travel.

City domains, like sadiq.london.

Admittedly, I can't remember any publicity of the weirder ones, like .lol or .xyz.

Have you ever tried to buy a .com domain dude? There's hardly any useful names left.

All 60+ domains I own are .com although I have a .net and a .org for a couple of words. I only paid premium price of $2000 on one of them. My company name is Sendlinks and getting Sendlinks.com kind of makes sense. ;)

In particular, I have a 5-character easy to spell meaningless name that I am reserving for some large scale service I will build in the future.

Getting an unregistered .com is tough, but many are available at a reasonable price on secondary market. I define reasonable price as under $1000. When I decided on the Sendlinks name, the seller originally wanted $5000. I made the decision to buy it at that price once I was ready to re-brand.

Since no one else would likely buy the name, I wasn't too concerned about someone taking it. Fast forward a few months later, I was at the point where I had money and so I decided to just buy it then and cross that item from my launch checklist. Luckily for me, the price dropped to $2000.

I own the .com domains of my Internet nickname and my last name among other properties.

That's not true. I just registered a six letter domain a few months ago (minops.com).

There's still plenty of good ones left. I created a program to help me find them: https://github.com/jedberg/wordgen

My point is that you're better off with a "bad" .com domain (bad really just means more than 10 characters long) than a "good" .weird domain.

Can someone think of a better way to do DNS? I remember when the internet was just a hosts file you had a copy of on your computer, so it is not outside the realm of possibilities to come up with something different. Something better.

Are gTLDs like this even worth it as a primary domain? In my limited experience, these kinds of domains tend to confuse people who are not technically oriented.

For people who aren't technically oriented, anything that isn't .com/.net/.org (and regional variants like .com.au) are confusing, but there's no good domain names left there. If you're starting a website you're stuck with the choice of a terrible .com domain with shudder hyphens or something, or you can get something like yourbusiness.shopping or yourhandle.blog.

It's gonna be a case-by-case thing whether one option is better than the other, but them's the breaks.

I thought about buying coffee.blog but the fee was 100k per year! Insane!

What are the rules around changing annual registration fees for gTLDs?

The new domain companies can raise them whenever they want and however much they want.

Don't give these scum your money, moral of the story is.

I exclaimed 'wow 30 bucks?' when I saw 30. But then was shocked at 250. Fainted at the auction.

No one cares.

Applications are open for YC Winter 2024

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact