Hacker News new | past | comments | ask | show | jobs | submit login
Price increase on .io domains on January 1, 2022 (Renewal: $55.00) (gandi.net)
286 points by leroman on Dec 1, 2021 | hide | past | favorite | 294 comments



How do we create an incentive model that kills the “parked domain” industry without out pricing people who want to host a legitimate non-commercial personal site?

At first glance I was happy to see the headline because it means it’ll be more expensive for squatters to profit, but then I do sympathize with people using these domains for legitimate non-commercial purposes.


In some countries, if you want to register a domain that's parked, you may appeal to the registrar, and they will take it away from the parker (unless they're legitimately planning to use it).

The conditions of use of the domain state that you will give them _some_ use. "Planning to resell it" it is not considered a valid use. It's a public resource after all.

As an analogy, imagine someone buys all the housing in town, just wanting to re-sell them at a higher price just because they bought them first. It's an obvious case for consumer protection to step in -- otherwise it's just angry mobs. Well, not in some countries.


Even if this is true, it is very easy to circumvent. Just host a blog with a few copy-pasted articles, a fake classifieds site, or a forum with registration closed, or something like that.


Too complicated, I'd like to start over with a new proposal.

Get rid of bulk discounts on domains, enforce a fee on bulk purchases.

First 15 domains per tax id are available at regular price. Additional domains have fees that escalate with scale.

The companies using automated systems to scalp 100,000+ domains should disappear over night.

This can attempted as an experiment for a new TLD first.


Perhaps there could be a grace period? If you're clearly not using it within 6 months then it reverts and you get a partial refund?


Which countries?


My thought is to require all domains be sold for the same set price (or at least constant for each TLD). If someone wants to register 1,000 domains let them, but don't allow them to profit off that. If someone breaks the rule, revoke the domain. I'd like to hear reasons why this is a bad idea so I can stop thinking about it every time I get priced out.


> I'd like to hear reasons why this is a bad idea

1. It is practically impossible to enforce. Maybe exactly the right amount of money changed hands in one transaction, but how do you know there were no other related transactions perhaps via anonymous crypto transfers or entirely off-net (the good 'ol brown envelope full of cash) or by non-monetary exchange (service rendered).

2. It punishes the final buyer, so the scalper in the middle doesn't care. It may put buyers off so reduce the competition and the prices scalpers can command, but it won't stop them completely. For every scalper that decides it isn't worth bothering because profits are down 50% (figure plucked from thin air) there will be a queue of smaller fish more than happy to take the remaining margin.

Look into attempts that have been made to control big-name concert tickets and how they have all failed - the problem is essentially the same here.


Or the simple method of having a company own the domain, and you sell the company.


If I understand your solution correctly this puts the risk on the buyer?

ie I naively buy a domain for a too high price, and then the registrar revokes the domain? The seller ends up with the money, and I end up with nothing? A smart seller could possibly even snatch back the domain from the registrar?


that seems like a nightmare to check. if 100 parties want to buy a specific domain from me (say: database.io), the incentive to sugar the deal will be huge. Why wouldn't I sell it to the party that'll gift me a Tesla when I sell them the domain for the fixed price?


Or just put on the invoice:

  domain transfer: $49.00

  totally unrelated consulting services: $49,951.00


That sounds totally unenforceable. ICANN doesn't have the resources to investigate in detail allegations that a domain was sold for more than face value. And certainly, nobody involved in such a sale will have incentives to report it, under normal circumstances.

Also, most desirable names would not be put to productive use since there'd be no relationship between what you can do with a name and how likely you are to get it, though this is a handwavier point.


Other folks have already pointed out how this is unenforceable, but I'll give a very tangible example.

Twitter (and Instagram) ToS prohibit selling handles and yet I've been on both sides of those transactions before. You just offer the person a "consulting" contract, part of which is delivering the handle


Why wouldn't the buyer just pay the seller and not bother with smoke and mirrors? It's not as if Twitter or Instagram know about the "consulting" contract.


Ship it!


What’s the next step? Communism for the Internet?


Communism is embedded in the core Internet protocols. And "we've" been doing our best to build a capitalist system on top of it ever since. :)


How is state ownership of the means of production embedded in the core Internet protocols?


Anarcho-communism, not Marxism–Leninism.


State ownership is merely one interpretation of the base communist concept of communal ownership. The internet is largely run on protocols built organically by the people using them and only later codified into standards by a committee of interested parties.


Pretty sure that's socialism. Communism is envisioned as a stateless society. Which is kind of what the internet is. There's standards bodies, but they don't really have any enforcement power. And ICANN, one of the few bodies with actual power, (by controlling domain name distribution) leads to problems like in this article.


> Pretty sure that's socialism.

State socialism specifically.


> Communism is embedded in the core Internet protocols.

Could you expand on that? People tend to use widely different definitions of "communism" so I'm not sure what you mean by that.


> How do we create an incentive model that kills the “parked domain” industry without out pricing people who want to host a legitimate non-commercial personal site?

The simplest solution would be to revoke domains with domain parking pages on them.

If the domain contains an offer to sell the domain, you lose the domain.

That wouldn't solve 100% of the problem, but it would solve 80% of the problem, and that's better than what we have.


This will only result in more sophisticated parked domain sites. That's what happens to phone numbers - companies buy phone numbers and put whatever trash phone services they can on them just to say "see we're using it for business we've got some pre-recorded phonesex running there".


Do it anyway. It makes their lives harder.

If someone can go to the domain and buy the domain, you lose the domain. It doesn't matter what else is there. If doesn't matter if nothing else is there. What matters is that something isn't: The ability to buy the domain.

That means Joe Schmoe can't go to the domain, put in his credit card number and enrich the scalper. Because if he can, they lose the domain.

If they come up with some convoluted scam to get around it, they'll get less business and scare away buyers who perceive it to be the subterfuge that it is, which is good.

The best thing about it is the low cost to honest people not trying to sell domains for money, who shouldn't be affected at all.


>The simplest solution would be to revoke domains with domain parking pages on them.

Parked domain pages are less bad than shallow fluff designed to try and improve pagerank.

>If the domain contains an offer to sell the domain, you lose the domain.

That could be used maliciously.


``` If the domain contains an offer to sell the domain, you lose the domain.

```

What about "We are looking for an acquisition/exit"?


What problem does this solve? Most domains are sold through exchanges


So prohibit listing domains on exchanges?

It's an offer to sell the domain that causes you to lose it, not where you put it. Putting it on the domain itself just makes it easy to find. But so does putting it on an exchange.


I believe we should be using the term "Domain Scalping" rather than "Domain Squatting". At the most basic level, it re-frames the problem.


This is the same problem as land value tax that was never solved.


It's solved in some [rather left leaning] countries. When you buy land you must put it to use, and if it's unused after X years, it's considered abandoned and it goes up for sale again.

TBH, I think it's a solid model -- buying a finite resource like land merely with the intent to deprive others of it and make a profit in the long term is as unethical as it gets. Though completely legal is most places, of course.


But how hard is it really to put a domain to use? Put up a hugo generated page that is even remotely related to the topic and off you go. Worse save a few related pages and reupload to a static site provider. A genuine entity wanting the domain will look to purchase even with "valid looking" content in there no?

Surely "put to use" doesnt imply profiting? If it does put up banner ads?

I am not saying this is ethical - just hard to make it discernible from ethical use. Atleast with land building something on it costs real money so it is easier to validate the "put to good use" clause! Sigh!


You make valid points, and, ultimately, what constitutes a "valid use" and what doesn't falls back on the registrar or the entity in charge. If it's a for-profit, you're gonna have a hard time, but if it's more of a non-profit, or government organisation, you might have better luck.

It's a huge "it depends" to be honest.

FWIW, if someone buys a huge amount of land next to the big city, and puts up some ads, I wouldn't consider that giving it "proper use", and if someone needs to build a house, I'd consider that land abandoned and unused -- but I can understand that opinions (and even laws on the subject) will vary greatly around the world.

The big question is: do you care about allowing people to have access to unused land to buy houses, or do you care about a corporation's right to own everything in sight?


This is effectively the justification that was used to repeatedly confiscate land from the indigenous peoples of the current US: "They aren't building visible improvements on the land in the way that settlers will, so move them off and give it to settlers".


Yeah, the justification is sensible to take land from organisations under your jurisdictions, that are part of your society, not civilizations that were there before you landed from another continent.

But the US has a mentality of "our jurisdiction is infinite", so I'm not surprised they'd do that.


With land, not using it is the equivalent to a privately funded conservation site. Its not necessarily a bad thing.

Many capitalistic countries also have adverse possession laws, where if someone uses the land for X number of years, it becomes theirs.


This loophole is solved by making it that declaring the land as such status means the land is commited to that use for a minimum number of years that makes the conservation meaningful. Something like a decade minimum with a high bar for converting it for other usage afterwards. Farmers in the UK for instance already have some incentives to leave some fields alone.


it was not solved by those that advocate for land value tax because it is not seen as a problem, just like many do not see domain "squatting" as a problem


At this point, there are many alternative gTLDs that can be used for personal sites, and quite a few are super cheap. Maybe the squatters will catch up and eventually buy them all up but for now there's a decent availability.


I just bought a throwaway domain for a little project- somehow we have exponentially more domains available than ever, yet nothing is as cheap as a 5-character .net was when I registered my first domain ~5 years ago...


The problem is that if you are already on .io (or another domain that increases the price) moving to another TLD means breaking all incoming links, email addresses, etc. Not really a good option.


You pay the max rent now, that buy 10 years, to price-lock, then begin the slog.


Can you suggest one? I'm interested in a gTLD that won't raise its prices suddenly and is of good reputation - so FB and co. won't block it.


.org is controlled by a non-profit and I believe it has reasonable controls on price increases.

(Private equity recently tried to obtain .org and jack up the prices, but the acquisition was blocked by ICANN https://arstechnica.com/tech-policy/2020/05/icann-blocks-con...)


> the acquisition was blocked by ICANN

Putting it that way is very charitable towards ICANN. They only blocked it after much public outcry and after the CA attorney general started an inquiry [0]. There is no reason to believe that they won't try again after the frog is sufficiently boiled.

[0] https://www.theregister.com/2020/04/17/icann_california_org_...


Don't Sell domains, instead sell a licence to use that domain.

You cannot sell the licence to somebody else, however if you sell the business/website etc, the domain transfers with it.

Sure, there will be a black market, however with the threat of having your domain licence revoked for breaking the licence agreement, I'm sure this will act as a sufficient deterrent.

Kind of like an OS licence, I can sell the desktop and the licence stays with the hardware (aka website), but I cannot sell my OS licence separately to somebody else.


Business: Unincorporated or incorporated? If the latter, just have a shell company and park some lorem ipsum, package some similar domains into the shell. This raises administrative hurdles, cost of enforcing, and makes it relatively easier for professional over a low barriers system. Domain name companies win.

Interest: Who defines an interest, community or other?

Personal?

Website? Domains are used for a lot more than WWW on port 80 or 443.


> business

The logistics of setting up as many shell companies and domains which are squatted seems unreasonable. The onus is on the purchaser not to illegally purchase a domain against the licence agreement or the domain is revoked. Both of these would likely massively reduce the financial incentives domain squatters enjoy.

> personal

They can still licence a domain, but when they're done with it, it returns to the pool of available domains for others to purchase.

> Websites

sure, and websites are the major driving force behind domain squatting though.

Maybe I'm explaining this poorly, and I know I haven't solved the worlds problems in a single post but it's fun to theorise, however if I purchase a software licence, I cannot sell that on. If a business purchases a software licence and that business changes ownership, the licence isn't always void.


> without out pricing people who want to host a legitimate non-commercial personal site

Don't buy a .io domain for your personal site - stick to the gTLDs or national TLDs that are run by trusted governments (for reference, I have a .ca TLD because I trust the Canadian government to keep the cost roughly competitive with gTLDs). ccTLDs that have been bought by private companies are essentially privatized in a way that gTLDs and government-run ccTLDs are not.


> ccTLDs that have been bought by private companies

It should not even be possible for companies to buy ccTLDs.


Why not? ccTLDs were for countries to do with as they choose. If they sold/leased-so-loosely-that-it's-basically-sold their ccTLD to a private business to operate, who is ICANN to tell them different?


Nuking 'domain tasting' would be a start.


I'd go one step further: Disallow corporations from selling public namespaces. Domain tasting exists because it encourages more registrations which is profitable for the registrars.


A related problem is, how do you make sure the domain goes to whoever values it most, as opposed to whoever thought of the name first? In the .com space a lot of good domain names are taken up by people who are not putting it to good use. I am not talking about parked domains, but people who upload an "under construction" gif and then don't update the page for the next 5 years. Maybe domain names should cost more.


> A related problem is, how do you make sure the domain goes to whoever values it most

Why should it go to whoever values it the most? And what do you even mean by thet? Whoever is willing to pay the most?

Domains are most useful as a stable human readable name for a server/service. If someone comes along later with a "better" use is not a reason to take the name away since that (or doing so indirectly by increasing the cost past what the owner can/wants to bear) would destroy the main value of domains in the first place.

And who are you to judge that an under construction gif is not a valid use of a domain. Maybe that gif is what the hoster wants to have there. Maybe there is more content that you don't know the path of. Maybe there are other non-HTTP services on that domain.


there was obviously some other issues with the whole scheme, but i think long-term the massive expansion of TLDs still holds promise to neutralize domain-parking.

the big problem is the scarcity of .com domains along with people's insistence on needing the .com because it is "better".


I thought domain squatting violated US law? Is there some loophole or is paying just cheaper/quicker?


Not as far as I know. I worked for a while for a company whose main business was squatting and throwing up ads on the pages. At least I worked for the part of the company that was trying other revenue models like lead generation although LG is of only slightly greater social value.


Only if the domain is violating a trademark.


For .io specifically, there are practically no legitimate non-commercial purposes since almost nobody lives in the British Indian Ocean Territories. Anybody with a .io domain is using it for branding rather than the original purpose of two-letter TLDs.

It's like renting a space on a fashionable commercial street — if rent goes up, there's not much point complaining.


How about hiding the address bar? Would make domain names irrelevant


Not all navigation is through links.


Like? All my use cases are covered by links and history search. And even then I sometimes go through google because I cant remember if it was com or net or org or some other cute domain name.

I was only half joking previously. Why require a solution when you can remove the problem instead?


What non-link navigation needs a domain name?

IP addresses work fine, like phone nmbers and street addresses.


How about entering in a URL from an offline mention, whether that's a print ad, being mentioned in a podcast or shouted from the stage of an indie band playing in a bar? When I started a band in the 90s, there was a great deal of consideration made in choosing the name for there being an available domain name (at the time there were about two dozen English words which had not been registered as domain names, of which exactly one was a potential band name (osteoclasts) but we decided against that—we ended up with the suboptimal dream-in-color.net since the name was unavailable with neither dashless nor .com/.org options available). One important thing was that the domain name would be something that a semi-drunk person might remember later.

Non-technical people will likely not recognize four three-digit numbers separated by periods as being an address for a website.

And phone numbers have been largely supplanted by contact lists on phones (I've occasionally forgotten my own number). Looking over my call history, out of the last fifty outgoing calls, six are not in my phone book and all but one of these were dialed directly from search results on the phone.


Things like cloudflare "host" thousands of sites behind the same IP. Old-school shared hosting providers are doing the same.


Good point. I've got about half a dozen websites that all share the same IP. I've been doing that for over 20 years now.


We are moving to ip6 not away


That's awesome for phishing.


It is. Would require an actual solution to prevent phishing - it's easy to overlook facebook.com.example.com

I highly doubt showing the domain prevents any meaningful amount of phishing attempts.


AFAIK, there are exactly two ways to avoid getting phished. One is using physical security keys (not implemented everywhere and we can't expect everyone in the world to buy one). The second one is checking the domain you're in.

Please, do elaborate on what other ways of phishing preventions you have in minds.


An app, perhaps call it ParkSurfing@Home that does nothing but silently visit parked domains in the background to burn up all their ad revenue without ever clicking on any links, making them completely worthless for advertising purposes.

Wouldn't stop the main issue but it might kick in the rear.


It would burn up your local resources and that of the adserver. Ads are loaded client-side.


That would have no effect. Ads are served from adservers, not content pages.


I believe if you kill the parked domain industry it will be harmful to legitimate businesses. Why? Because these domain squatters provide a valuable service - making sure domains are still purchasable by legitimate businesses, even if at a premium. Get rid of them, and people will go out and get various domains they like and might use one day (or redirect to the same spot), with no intention of selling them. More domains, not fewer, will languish.


You have to increase the yearly renewal even more (imo). $120/year is only $10 a month. That is two coffees at Starbucks. If that is too expensive for your legitimate non-commercial purpose, then you probably don't really need a domain. Yeah, some squatters may still try the long game but it starts to get cost prohibitive. The hope would be that if it is expensive enough, the squatters may only hold the most likely TLDs for a specific domain instead of holding all of them at almost no risk. I also think that we can extend this by adding lots and lots of new TLDs. Yeah, the single word domains will get snapped right up but the space becomes ever bigger and harder to corner the market. We could also make a rule that when there is a new TLD, that existing domain holders from the longest lived TLDs get first right of refusal. So if ".cars" is added, Ford would get the first shot at ford.cars

This squatting is the same reason we need a LVT on land to prevent people from squatting on property that could be developed to provide housing.


As the owner of a couple of domains (not .io, but these aren’t the only ones that go up in price in January) with no commercial use whatsoever, I am very happy to learn that you have decided that I should not get one. You sound exactly like those landlords who think it’s all jolly good when the average rent goes up at twice the inflation rate.

In the meantime, we got those domains because we use them, and not all of us actually get 2 Starbucks a month. The price going up 30% without any recourse and with a 1 month notice is a sad joke. We cannot seem to trust private interests with the management of public resources.


> coffees at Starbucks

I hate this argument. All the small fees priced in "Starbucks coffees" add up quickly and all of a sudden running a personal website to share one's hobby dries up the monthly coffee budget and then some. Then there's the fact that a significant part of the world lives on a single US Starbucks coffee per day.


That's a very US-centric view. There are many countries in the world where median wages are below $1000/month, and then $10/month is 1% of your income, which is a significant expense.

Increasing the price even further means that people from poor countries won't be able to afford domains (at least not a .io or .com), and this is bad for the open web, because it means that fan forums move to Reddit or Discord, small businesses move to Facebook and WeChat, and so on, until all content is inside these walled gardens instead of the open web.


Some of the best sites on the internet are run for less than $120/year.

> This squatting is the same reason we need a LVT on land to prevent people from squatting on property that could be developed to provide housing.

Yes because private property all exists to become housing.


> So if ".cars" is added, Ford would get the first shot at ford.cars

Actually .cars exists and ford.cars is indeed taken by Ford.


Your "solution" to squatters would be to make sure normal people can't afford domains at all, the only affect this would have on squatters is that they would be a little bit more picky about their inventory and raise prices to compensate.


Ted from Namecheap here. Donuts, who now operates the .io TLD, is only increasing the wholesale price by ~10%. We are still obviously not happy about this since it's always our goal to ensure that domain pricing is as accessible as possible to our customers.

It would appear that Gandi has just chosen to substantially increase their markup as part of this Donuts update. Our pricing will likely only increase a small amount, if at all.


Former Namecheap customer here, thank you for the transparancy. (I only left Namecheap because of the interface, I found that Dynadot was comparable in price and features, but with a nicer GUI. NC seems like a good company otherwise, so I'm always open to the possibility of switching back.)


Sorry to hear you had trouble with our interface. We do have some pretty substantial updates planned for 2022 and would love to have you back. Feel free to shoot any feedback to ted [at] namecheap.com.


Ted, thank you so much for participating in this thread! Namecheap provides excellent service, and I've been a happy customer for years.

As someone who owns ~30 domain names split between Namecheap and GoDaddy, it would be incredibly beneficial if you overhauled the primary domain list view. As it stands right now, your domain list UI is bulky and clunky and requires several clickthroughs to show all domains or sort effectively.

The first view you see should show a glorified table / spreadsheet view of all domains on the same page by default. A reasonable pagination limit would work for extremely large accounts, but in general I think many people would want to see everything in one place if possible.

It would further be fantastic if the domains were immediately sortable on multiple dimensions (name, expiration, tag/category) and had an integrated search box.

Tags are also really important. The tag system you have in place right now is rather useless, because you can't leverage it easily from the main view. You have to navigate away to an entirely separate product area, and even then they don't provide much assistance.

I tag my domains to group them by project. (eg. vocodes.com, vo.codes, fakeyou.com, fakeyou.app, etc. are grouped together.) It'd be really great for this to feel first class and be integrated into the primary domain view as an additional set of filters.

Godaddy used to suck in this department, but they've really upped their game. Their domain list is top notch and gives Namecheap's UI a run for its money, which is a real reversal. You used to have the best in the business.

I really abhor GoDaddy since they allowed an old high school friend of mine to break into my account and steal strategywiki.org ~15 years ago. (For those interested, I've posted the story on HN before and it should be searchable.) The only reason I have any business with them at all is because of domain marketplace purchases. Grudges aside, their product has evolved over the last few years and is really good. I'd use it as a reference point.


Passing all of your helpful feedback to our Domains PM. Thank you. As I mentioned in another comment, we're developing a new dashboard experience among other things, which we're really excited about.


Any chance you could support .pm and .it domains? Would love to move over my remaining domains to you


Will pass on the request.


Fantastic, thank you so much!


Does that update include the ability to dynamically update a host's IPV6 address via a request to dynamicdns.park-your-domain.com? The guide [0] mentions that

>It is only possible to dynamically update IPv4 addresses (A records) at this time. Currently, IPv6 is not supported by our Dynamic DNS.

Is there any particular thing blocking this feature from being available for IPV6?

[0] https://www.namecheap.com/support/knowledgebase/article.aspx...


Navigation! It’s a mess. I would very much like to see separation of “get the job done” navigation (and fields) from “upsell” navigation (and fields).

Also, could you unhide the login so it’s easier for password managers to autofill and for me to submit?


The interface may be a little clunky, but they provide better support than any other domain or SSL vendor, which always gives them the advantage in my book.


Support is core to our business and always our main focus but so is UX. We are admittedly a bit behind in improving our dashboard but can promise you that we have already started developing a radically updated experience and will be ready in the nearish future.


I agree, I was a big namecheap user but now it is a dark pattern trap to trick people into addons. Just changing your DNS is a confusing jumble.


I just registered a domain with them the other day, and there didn't seem to be any egregious dark patterns. Are you talking about anything in specific?

Of course they present options during the checkout process, which is slightly forward, but reasonable as a business.

However, (1) the defaults for all add-ons are do-not-buy, (2) the buttons to navigate to the next screen are clear (not swapped, minimized, etc). But that may have just been my settings / environment?


Isn’t the wholesale price of domains public information?


TBF to Gandhi, IIRC they tend to give favourable prices when you buy domains or deposit funds in bulk. Their opening tier of prices are indeed pricey. Their API is excellent though.

I've always found Namecheap to be price competitive and their customer service is pretty good also. API was always a bit basic (think that had a lot to do with enom implementation)


I'm sure they do indeed have special pricing for bulk customers. Although, not sure how much has changed with them since they were acquired by private equity a couple of years ago.

I do admit that our API needs some love - fortunately we have a lot of major backend updates in the works and a new API will follow at some point. Unrelated but we did roll out a new API (entirely separate from the API you're referring to) for our Auctions platform: https://documenter.getpostman.com/view/12017783/TW6wKUfy. Curious to get your feedback if you're interested in taking it for a spin!


I'm not sure how giving favorable pricing to bulk purchasers (i.e. mostly squatters) is a positive thing for domains.


Hi Ted, this seemed a bit outrageous, so I asked the question to Gandi on Twitter. It seems what you're talking about is the "promotion price", see: https://twitter.com/gandi_net/status/1466358344998965253

And since I wasn't the only one to ask, we now know that the actual bulk price will increase 50%: https://twitter.com/gandibar/status/1466313609475198981

Which means that Gandi is in fact reducing its relative margin. I'm curious about you opinion on this. In particular, how different are bulk and promotion prices ?


Thanks Ted for the update. I bought my io domain from you guys many years ago when it was $70 or so. Thanks for keeping the price low over the years and for the update.


Thanks for choosing us!


Love it that you report on details here. Happy namecheap customer, may GoDaddy burn.


coupon code "BYEBYEGD" works for .com transfers ha


No real feedback other than I've been a happy Namecheap customer for years, thanks for the great service.


Music to our ears.


What about the future increase of .sh? I own one for personal stuff, and it's apparently going to increase from ~40€ to ~73€ which is a bit hard to swallow for personal use.


Not familiar with any .sh price increase but it's certainly possible..

If you can swing it, you can add years to your registration(s) now to lock in the current price before any sort of increase.


We left namecheap bc it lived up to its name: we were harassed by defective copyright notices that were copied to namecheap. Instead of forwarding them to us, you sent us an ultimatum that you would drop our DNS. We even had our General Counsel draft you a letter, only to have it repeated. Then we understood that we should just drop the vigilante registrar Have been 7 years with networksolutions and very happy.


I use Namecheap to host more than 10 domains and some shared hosting, and found customer support to be awesome! Just wanted to mention it. Wondering if you have a special promo plan for guys like me that depend on Namecheap for all their personal and business domains?


Check out our tiered pricing. If you qualify, support can set you up! https://www.namecheap.com/domains/tier-pricing/


Thanks for the transparency and great service! I have all but one of my domains there now (register.com had a deathgrip on until I got fed up with their process and let it lapse. Now it sits available for $370, which I'm not willing to shell out)


Thank you! Bummer re the premium cost on the name that expired.


Thanks for the great service. I used to have all my registration through Dyn (nee DynDNS.org); when they got bought out by Oracle and started moving everything over I moved to Namecheap. Been perfectly happy since.


Woot!


> it's always our goal to ensure that domain pricing is as accessible as possible to our customers.

Then why not charge $30 for an .io domain like CloudFlare does instead of $39.98/yr?


We would love to and we are always striving to get as close to wholesale as possible but we do have operating costs — domains are essentially our core business and we have 1,000+ CS agents + a large risk/abuse team that works around the clock to focus on the customer experience. Anyone who is selling domains with no markup at all is taking a loss in some form and/or treating it as a loss leader for other products.


Ted, you guys are awesome. Been a happy customer for the last maybe 7 years. Keep up the good work!


So happy to hear that!


Taking this opportunity to ask industry professionals.....

What the hell is happening to ".Web"?


Still in limbo but looks like Verisign will be able to launch it at some point: http://domainincite.com/27180-verisign-boss-talks-web-launch...


>Still in limbo

Thx. They won it in 2016 and we are fast coming to 2022. Sigh


I cannot pass up the opportunity to personally thank you for NameCheap.


Thank YOU!


Hey Ted, will Namecheap go public anytime soon?


No plans for that. We have never even taken any investment and luckily have been growing substantially over the last few years as a bootstrapped company.


Oh I would love to read the NC journey as a bootstrapped company.


It might be helpful to look at this as a percentage change- it looks like they were currently charging $42.18/year as the standard price (1), so this is a 30% price increase.

(1) https://news.gandi.net/en/2021/10/enjoy-our-io-promotion-bef...


As another commenter pointed out they already raised the price a year ago - sot this is a 30% YOY increase.


Well that is one way to get 30% YoY company growth... I wonder how many new domains got registered and what the total revenue diff is.

Actually, could it be that the .io hype went down and this is their attempt at compensating?


I used to work for a registrar and had the job of connecting up to the registries, who are the entities that ultimately allocate the domain names. (Yeah great naming choice, no chance of those words being confused with each other.) Registrars talk to registries via EPP, an xml based specification. While there is a spec, the registries variously mess it up in odd and interesting ways, all of which I got to discover as we did these integrations. .io was by far the worst. I remember that you could renew domain names in units of either 1, 2, or 5 years. It was like a developer had done the dynamic programming problem with coins and decided that they would put that in their api.

Considering that .io seems to have been adopted as a signal of "geeky tech cred", I always laugh, and think really it stands for .ironic.


The .io infrastructure under ICB was indeed quite slow. Everything moved to Afilias (and now Donuts) many years ago though and EPP is quite performant now fortunately.

We deal with a few independent ccTLDs these days that are still horribly unreliable though. Not going to name any names but some of these are quite popular and still either don't run on EPP and/or run on an old outdated server in a local DC.


Btw, would you mind sharing what registrar you worked for? No worries if not but it's a small industry!


A reminder: just because you don't see any meaningful activities happening on www., don't assume that the owner is domain squatting. Folks make use of their domains in a variety of ways, not just for hosting a website on www.


Exactly. I use it for email


Can namecheap.com do a package deal for hosting a Gemini site with down the line add-on for VPS? I don't want the Wordpress stuff.


The .io TLD is owned by Ethos Capital. The same private equity company that tried to buy the .org TLD a while ago. Go figure...


From their website:

"OUR ETHOS Drive prosperity for all stakeholders

We believe that long-term success is best accomplished through a thriving ecosystem in which prosperity is shared amongst all stakeholders." [1]

Bullshit, they exist to squeeze every little penny left in a company out and then dump the hollowed out remains on the stock market to die.

[1] https://ethoscapital.com/about-us/


Well they obviously mean that those 'stakeholders' are their shareholders.


The stake that is held is driven through the heart of the company.


[flagged]


>Who exactly are you quoting?

They are quoting the page they linked, which is the Ethos Capital "About Us" page, where they have written the above quote under the heading "Our Ethos".


At the time of my comment, their post did not include that link, only a random quotation and "that's bullshit, here's my take".


I think it's pretty sad how the rights for this get sold around, but the territory to which it belongs gets little to no revenue from domain sales.

There's an ongoing complaint seeking repatriation of the domain, and I really hope I'm wrong in speculating that nothing will be done about it.


Man, that is early-Web levels of sordid.

https://en.m.wikipedia.org/wiki/.io#History

FYI, apparently this is the current state of UN affairs: https://www.lexpress.mu/node/388292 (previous and recent decisions referenced)


I’m always mind-boggled at the level of acceptance in the industry of clear misuse of ccTLDs. Is having “input/output” in your domain really so cute that it’s worth tying your entire business to an incorrect ccTLD? Especially when we have so many new cute, built-for-purpose geek-cred gTLDs like .dev.


It’s much worse than that. The Chagosian islanders who lived in this territory were all exiled by the British to make room for a military base on their land. They have been fighting in the legal system for decades just for the right to go home.


Ethos later acquired a controlling interest in Donuts (Jan 2021), who acquired Afilias (Nov 2020), who is the service provider for the .org TLD. The "operator" is still the Public Internet Registry, but from my understanding Ethos now controls the company that effectively runs the .org infrastructure.


Wasn't it owned by Internet Computer Bureau (ICB)?

Regardless, the British government profited off of the .io domain while it rightfully belonged to the Chagossians of British Indian Ocean Territory.

https://gigaom.com/2014/06/30/the-dark-side-of-io-how-the-u-...


As far as I can tell, the British government has officially claimed not to have been paid anything from the .io registrar.

Despite the registrar claiming it deposited some amount into accounts for each territorial party they were ostensibly operating on the behalf of.

So, someone's lying.


Ethos Capital has a mortgage operation. My experience with them lead me to the conclusion that their name does not imply "ethical".


The sunnier the name, the darker the reality.


The inverse of this doesn’t hold unfortunately. Blackwater sounded like a load of bad guys and behaved like it too.


Cerberus Capital have entered the chat...


What is scary is that Ethos Capital is made up entirely of ICANN affiliates. In other words, these are the same people that govern our domain names.

Pretty fucked up if you ask me. They are also going around buying up basically every major TLD.

Lease with ICANN and be subject to their price gouging forever or own with Handshake, the only thing ICANN is afraid of [1], and never pay again.

[1] https://www.icann.org/en/blogs/details/buyer-beware-not-all-...


Handshake, what's that? Oh, I see... blockchain... "Handshake coin (HNS)"... PoW... the way you're plugging it, I thought it had that kind of smell.


Oh, excuse me. I thought that smell was Ethos Capital.


It’s possibly worth noting that this is through Afilias, who seem to operate at least eleven other countries’ registries, including Australia’s (.au), and such of the older-but-not-original gTLDs as .info and .aero.

https://afilias.info/ has a list.

This seems to be to be a fairly rotten space, with power tending to consolidate around parties that have acted in fairly demonstrable bad faith.


Irony Capital should hostilely buyout Ethos capital to complete the joke


According to their website, they do not own .io

https://donuts.domains/what-we-do/top-level-domain-portfolio...


IANA has it as Internet Computer Bureau Ltd [0], which was bought by Afilias in 2017 [1] and is operated as a subsidiary, and Afilias itself was bought by Donuts in 2020 [2], and Ethos Capital acquired a controlling interest in Donuts in 2021 [3].

So it's accurate to say that ICB (inasmuch as it exists) operates .io, as a subsidiary of Afilias, a Donuts company, controlled by Ethos Capital.

Simple!

[0] https://www.iana.org/domains/root/db/io.html

[1] https://en.m.wikipedia.org/wiki/Internet_Computer_Bureau

[2] https://en.m.wikipedia.org/wiki/Afilias

[3] https://en.m.wikipedia.org/wiki/Donuts_(company)



[flagged]


How do you deal with trademark infringement and phishing with Handshake domains? I'm sure some bad actors are salivating at the idea of registering trademarked domains on the blockchain so they can extort businesses anonymously.

No one using their domains to build things is going to want to buy into a system with no legal oversight. It's only squatters, flippers, and scammers that want that system because they can extract value for doing nothing.


Handshake is nowhere close to overtaking ICANN.


   yet.
Even Namecheap seems to offer Handshake domains. But I always get the feeling that some folks on HN seem to be consistently wrong with their projections.

I wouldn't rule out these blockchain-based domain guys that quickly; since there is utility there. Unlike the other useless shitcoins I keep seeing.

[0] https://www.namecheap.com/domains/handshake-domains/


Shame on Namecheap and Porkbun for scamming unsuspecting suckers.


I like how the .nl tld in the Netherlands is managed by the non-profit SIDN [0]. They are dedicated to an open internet, and they make sure that I can very easily move my domain between (hosting) providers for example (and indeed it was easy to transfer from transip to vimexx for me). They are truly independent. It's how it should be if you ask me. When I read about the .org issues and now .io, I'm happy with my .nl domain, for personal use.

[0]: https://www.sidn.nl/en/about-sidn/what-we-stand-for


> and they make sure that I can very easily move my domain between (hosting) providers

Is it so easy that I could "do it for you"?


Hehe, no, at least, not without my credentials (and I turn on 2fa everywhere). You need a key from one provider and provide it to the next, for example Transip (one of the biggest Dutch providers) has the procedure to import a domain outlined here: [0]. To move away from Transip do this: [1]. I bet it's similar for other tlds, not? I have no experience with others. I went through this, no downtime, no issues, DNS was propagated on the same evening I initiated the transfer.

[0]: https://www.transip.eu/knowledgebase/entry/24-transfering-yo...

[1]: https://www.transip.co.uk/knowledgebase/entry/28-how-transfe...


If you are not changing any DNS records (A/MX/CNAME... though you can change the NS records if you move the entire zone correctly, and the old NS servers keep it for a while too), there should never be any downtime related to a domain registrar change.

DNS propagation is not completely under your control (there can be caching DNS servers that ignore the TTL of records), so you should always give it at least 24h by serving the same zone from both (if you've changed NS records). I'd instead avoid that worry by using non-registrar provided DNS servers.

I would hope that TLD registries that fail to provide this are very few if not non-existent.


It seems, that .io is rising the price every year this time around.

Just about a year ago: https://news.ycombinator.com/item?id=25534498


That's because it is owned by Ethos Capital, and is the reason they shouldn't own .org


What's the reasoning or history behind why domain prices A) vary between TLDs and B) are allowed to be sold at any price (think huge markups from domain speculating)?


Once you have enough users, you can raise the price and they can't easily flee.

For example, I (regrettably) have a .io domain with my own email, and moving away from it would be super painful. I'd have to make sure that I update my email *everywhere* (or risk my accounts getting hijacked), and even reach out to people with whom I communicate very infrequently.

TBH, I _should_ do it, but I can image that taking around a year.


> TBH, I _should_ do it, but I can image that taking around a year.

And even then, it will probably feel bad to let it expire just because you might have forgotten an account. That's why I still have my very first email. And it's a lot of effort for saving a few bucks a year (at least in the US and the EU, for poorer countries the story might be different). Unfortunately, that's exactly what Ethos is exploiting.


Yeah, it's not purely about the money, but also knowing where it's going.

I might spend that much money on takeout in a week, and it's fine cause the money goes to the kid who biked it over, and the chef who cooked it, etc.

But the same amount of money going to a corp trying to squeeze as much money as possible giving no extra value in return, and depriving the original owners of that revenue... that does tend to make me feel dirtier.


If you have a good .com that's tech themed you basically have to eat the cost on the matching .io for brand protection even if you don't use it for anything.


There were always different entities in charge of different TLDs. ICANN, for example, has no leverage over CCTLDs. Basically because "no central authority". Then, within a TLD, some allow resellers, so competition.


Any perspective on the best TLDs for startups?

.com is obviously the preferred choice, but from my perspective these are the next best picks:

- .io for B2B, SAAS, etc. startups. Occasionally makes sense for B2C (itch.io, etc.) Seems to be falling out of fashion, though.

- .ai is the top choice for ML-based startups. These are really hot now.

- .dev for the occasional SAAS or library. Doesn't make sense B2C.

- .app if you're linking to a mobile app or have a SPA-like functionality. Not super popular, though.

AFAIK, the other TLDs seem like spam/noise. That might just be personal perception and I could be totally wrong. Am I missing anything?


Isn't that mostly a risk and marketing question? How seriously can you expect the TLD to be managed in the future and what do you want to communicate.

My own personal bias makes me a bit wary of "novelty" TLDs since you can't really know how they will be managed in the future. Would you use a banking service that has a novelty domain? Perhaps not. Would you use a piece of software that is hosted on the same TLD? Probably.

I have no idea if my own personal bias is shared by any significant portion of the internet population. Nor do I think my bias is consistent (for instance I can't really decide if ".io" is a novelty/fad TLD or not. Some days I think it is).


Anything other than .com, .org, .net, you're basically at the mercy of the owner who owns that domain extension. They can price-gouge at will.


Pure opinion here but the .tech tld seems a good choice as well. I've used it.

And the business specific TLDs are probably good. I know someone who recently launched with the .financial (which I don't like as it's too long)..


Not for a startup, but .codes is pretty nice for personal or portfolio websites.


I used that TLD for word play on vocoder with https://vo.codes


Ha, this is great. I cracked up at the SpongeBob ones.


My personal website uses .one


This is basically treating the TLD like the next buzzword, but in the end: does this have any practical effect apart from being cool in the eye of the owner? I mean I've been there and done that, but all things considered it doesn't actually seem to matter - there's still the domain as well. Though I might be missing something.


I remember YC's general rule of thumb. If your customers / users are consumers then you need .com.

Otherwise for SaaS for B2B it doesn't matter as much ( if at all ).


I've been falling back on .app because they're relatively affordable, there are still a lot of names left on that TLD, and I do happen to build websites which are more like PWAs or SPAs.


I'd steer clear of .icu. They've had such quality problems that it's not uncommon to see them completely blocked in mail filters and the like.


ICU is very strongly tied to "Intensive Care Unit" for me.


I thought it was synonymous with the Unicode library for any sysadmin, db admin, or C/C++ programmer: https://icu.unicode.org/


.so and .co are popular when .io and .ai are taken.


Somalia? Isn't there a risk of the government collapsing (further)?

Though British Indian Ocean Territory has a risk of the territory disappearing. Colombia and Anguilla should stay around.


The domains at this point aren't really tied to the goverment.


I think ".co" doesn't look too awful and most people "get it." If it's a B-to-B you have more flexibility


I wouldn't want a .co if I didn't also own the .com variant, though.


Considering using the ".tech" TLD for my future projects


The stupidity of thinking what "the best TLD for your startup" is.

In the meantime all respectable startups and big companies use .com like normal people.


That's an odd comment to me. Different TLDs have different pricing, different restrictions, and so on. Giving the topic some thought isn't stupid. And .com is a pretty crowded namespace.


>different pricing

.com is the cheapest

>different restrictions

.com has the fewest

Anything else?


>.com is the cheapest

It isn't, either for initial purchase or renewal, and that's pretty easy to verify.

>>different restrictions - .com has the fewest

How so?


Country code TLDs (ccTLDs) [1] often have rules set by the country of origin. These rules vary widely, but some of them include country of residency requirements (you can't buy as a foreigner), ban on pornography, or strict adherence to religious requirements.

Popular domain hack ccTLD .ly is required to adhere to Libyan and Islamic/Sharia Law [2]

(edit because my posting is rate limited: )

There are plenty of non-ccTLDs, though, including thousands of gTLDs.

You can register a .horse domain if you want, and as far as I know, there are no restrictions on use.

You just need to research what you're buying.

[1] https://en.wikipedia.org/wiki/Country_code_top-level_domain

[2] https://en.wikipedia.org/wiki/.ly


Yes, but singling out .com as if it were the only non ccTLD doesn't make sense.


Likely because 'respectable startups' (with money) can afford to buy the .com.

https://www.arbo.works/ is a new, and very respected startup. They don't have a .com (yet).


Discord couldn't afford to buy discord.com, so they bought discordapp.com. They didn't think of buying discord.io or discord.app or some crap like that.


Ah yes, branding is "stupid" now. Gotcha.


I think the GPs point is that anything other than .com is stupid branding. Not that branding is stupid.


yeah, and that's the reason why most of them have stupid names


Build an amazing app/platform/service. Host it on whatever tld is cheaper/convenient/some_other_criteria. No one will care what the extension is - especially with current and upcoming consumption models (mobile apps, vr/ar, and so on).


> No one will care what the extension is

Until you try to send emails from your random TLD domain and they all end up in spam.

Or when your shitty TLD has a NIC-level DNS outage. (.club had one a few weeks ago for instance, experienced it firsthand.) Or no outage, but resolves slowly.


Company: oefrha-galactic-corporation.com

Project: oefrha-super-app.${something}

One owns the other. Useful for running multiple experiments at once, though shifting to a different structure is better (imho) if the experiment is succesful.


No one will care what the extension is

Only on HN. In the real world, it matters quite a lot.

I tried to get a healthcare company to pick up its .health equivalent. The marketing department included the possibility of using the domain in its annual customer survey. The result was that a very large number of patients believed it to be a spam or scam domain.


Even as a competent user myself, I still hesitate when I see a .biz or .info domain. I wouldn't say I trust .com's more, but it's certainly built into my mental checklist when evaluating if something is legitimate. For decades the unconventional cheaper domains were used for spam purposes.


Yep, for those of us who were around in the earlier days of the internet, .biz means low quality at best, spam, popups, and malware at worst.


It's not that simple. See @substack's example. He has the .net domain, and now a new company with unlimited $ bought the .com.


StartCom refused to issue free certificates for free .tk domains, "back in the day", for spam-related concerns. Emails from these domains may be blocked, too, I guess.

So the tld matters a bit still.


You can always find a suitable .com domain, which is at the top of the squatting pecking order. Just need to be creative and flexible


Is there any objective reason (performance, reliability, etc) to avoid certain TLDs?


.com/.net/.org - your name is probably gone

New gTLDs - The registry can pretty much arbitrarily set prices in the future

ccTLDs - Politics might make you have to change domain (e.g. UK owners of .eu)

Pick your poison, I tend to stick with the original TLDs and my own country's ccTLD


How is .me? It's used a lot by students for their portfolio, I bit the bullet but wondering if I screwed up.


IMO .me is one of the safer ccTLDs. It's from a fairly stable country that hasn't been keen to hike up the prices or impose restrictions on who can have their domains so far; it's not often associated with spam; and it's even used by large companies like Facebook (fb.me).

When it comes to pricing, though, you never know what will happen down the road. I honestly don't know who I'd like to trust more: Verisign or the government of Montenegro?


> When it comes to pricing, though, you never know what will happen down the road. I honestly don't know who I'd like to trust more: Verisign or the government of Montenegro?

Since .me is more targetet at / suitable for individuals rather than a hipster startup domain like .io, the prices will hopefully remain lower. Or at least that's what I like to tell myself.


.me is absolutely fine and, as an employer, I would not mark down an applicant for using it.


Some country tlds are risky because they don't necessarily check residency requirements when you purchase, but might later if your domain catches their attention. Not all have residency requirements, but some do.

There's also some TLDs that are well known for web spam, phishing, or email spam, sometimes because of free or cheap subdomains. Those might have issues getting indexed in Google, or issues with SMTP delivery. Google once deindexed the entirety of *.co.cc, for example.


A couple, but they're all "human" reasons and not really technical. The TLDs records will be served off the gTLD nameservers like everything else, you can change your own nameservers from there, and DNS is slow as hell anyway (by modern standards) so it's already not like it's tightly optimized usually, and performance as perceived by the end user is greatly predicated on how warm their cache is. So any performance difference between nameservers probably barely matters unless you're like Amazon or something.

Anyway, non-objective reasons I can think of:

Lots of people don't know anything but com exists.

Com is definitely still dominant, so someone sitting on yourthing.com when yourthing.biz.no is your real site will probably be getting some chunk of your traffic. Imagine if Amazon.co.uk was the real Amazon and Amazon.com was like, a news website dedicated to posting all the bad stuff Amazon does. Tbh that'd be nice for society but you can probably understand it would be undesirable for a company.

.io is one of the only ones that's managed to kinda read com-like status, and it's only for a niche (though a lucrative niche!) audience.


> Imagine if Amazon.co.uk was the real Amazon and Amazon.com was like

This is actually an interesting example, because Amazon.com is a US/Canada specific site, while Amazon.co.uk is the UK/Ireland site (though I'd imagine they'll eventually either set up an Amazon.ie or change the redirect to Amazon.de with the tax/import brexit implications of that). So while all domains are Amazon owned, I wouldn't say Amazon has .com primacy over regional domains.

Ditto for google search, though the rest of their products are on google.com for me (I think gmail used to be googlemail.co.uk in the UK and googlemail.de in Germany because of local companies who had the gmail name first, until Google bought out the names from the local owners).

All the web 2.0 or newer companies are .com rather than regionalised though.


The $1 TLDs like .top .xyz .fun etc will be blocked in a lot of places due to heavy spam and malware distribution.


I remember reading a blog post that I can't find that one company saw a massive uptick on email deliverability when they stop using .xyz and moved to a more standard one.

It makes some sense, .xyz are so cheap that I've bought mines for $1 so just buying a domain every day isn't outside the realms of possbility for spammers. Personally, I just use my .xyz domains for internal stuff where I still want things connected to the internet and available.


While there are some performance differences [1] between TLDs, I don't personally think it should factor very heavily when choosing one.

As for reliability, there have been a few ".io" outages and I remember ".club" going down a few years ago but DNS makes complete TLD outages pretty rare.

I'd say the biggest reason to avoid certain TLDs is the human factor. Humans are used to typing ".com" or ".org" or ".net". I don't have any data to back that up though, so it probably isn't a satisfying answer!

[1] https://blog.nameshield.com/blog/2020/04/30/choosing-the-rig...


ccTLDs are perhaps marginally less reliable given who has overriding control over them (.ly is Libya, for example), but beyond that it's probably pretty low on your list of things to care about.


This is not true if you want email to come from you main domain and you choose a remotely exotic TLD. I learned this the hard way. Don't be like me, choose a boring TLD, worry about less stupid things - there are so so many already.


That's a good point, I was thinking of it from a web perspective.


Long time buyer of random domains here. Over the years I've heard reports of people trying to get back their ccTLD because they weren't a citizen of that country and had to fight back-and-forth to reclaim their domain, oftentimes handing ownership of the domain over to a trusted citizen of that country so they can keep their domain.

This is why I don't build important stuff on a ccTLD and go for generic ones like .club .party etc. In my country, I have a single ccTLD that I registered with a passport scan and had to verify my legal name with the registrar, which is good because I don't have to fight to get it back in a rug-pull scenario.

I don't have anything too important on it, just a few links to projects I worked on, and a contact page. The domain price could be hiked at any moment so I don't take the domain too seriously. (I've often thought of just not renewing it, and retiring it).


The issue is you will always be leasing versus owning unless you use Handshake. The same entity of interest, Ethos, owns most of the generic top level domain names since they own Donuts [1].

[1] https://donuts.domains/what-we-do/top-level-domain-portfolio...


While I like the idea of using blockchain-determined name ownership; there's a few problems. First, you have the problems that you get just because of how blockchains work[0]. But moreso there are specific chicken-and-egg problems with migrating away from standard DNS:

- Because DNS is well-established and everyone uses it, most websites want to be registered in DNS, and blockchain names provide little to no value to them.

- Because blockchain names see little usage, most ISPs see no need to resolve them in addition to or overlapping standard DNS names.

For a blockchain name system to actually succeed, you need buy-in from large network operators. Most network operators, for better or for worse, have just decided to trust ICANN and ICANN alone. At the very least, you would need IANA and ICANN to agree to set certain TLDs aside for competing naming systems so that network operators could be confident that resolving them wouldn't cause security problems.

[0] specifically the whole 'immutability' and 'no central authority' thing legalizes certain kinds of theft


I think big tech and entertainment will actively fight against blockchain domains if it comes to it. Their brands are massively squatted on from what I've seen. Here's a single ETH wallet that's squatting on a solid 40 .eth domains with high value trademarks:

https://app.ens.domains/address/0x140eF05a724CA161d49ffE8D97...

Do you think `microsoft` or `disneyland` or `amazonprime` or `barclays`, etc. are going to be onboard with a system where their global brand names get zero legal protection? Not likely IMO.


You're not wrong, but I'm getting tired of explaining to cryptobros that possession of a cryptographic private key is not equivalent to legal ownership of property.


All of the top domains which make up the majority of web traffic were reserved to the domain holders on Handshake [1].

[1] https://dns.live/top.html


That's not a good replacement for actual trademark protections though. They reserved 100k TLDs in a pool of unlimited TLDs and there's no solution to stopping 2nd level domains from infringing.

For example, what happens if the owner of the HNS TLD `techsupport` decides to sell subdomains and sells `microsoft.techsupport`? That's an amazing phishing domain if HNS is part of a mainstream resolver somewhere. How about `stripe.payments` or `disney.movies` or `google.search`?

What do those reserved domains do for me if I'm not in the Alexa top 100k?


The phishing issue not an argument against HNS domains. Phishing already exists without handshake domains. With ICANN domains you can also be creative and place misleading domains that harm for a while. I would rather say the ICANN system as it is today is not addressing the problem properly and fast enough. There are plenty of ways to make it faster and more secure working with a decentralized system that is not dependent on the reaction time and judgement of one institution.

If you are today not in the Alexa Top 100k you have plenty of room to find a great domain. And there is a lot of opportunity to get better names. ICANN domains are limited, intentionally made scarce and often blocked and not used. And you need to rent them instead of owning them. What is the added value from renting them?


I've been using a .io for several years for primary email and side projects. I'd have about two dozen critical emails to switch of I got booted of .io. I had no idea it was a ccTLD.


Same thing with .sh Domains among a few others. .sh in particular is going up from 40-something to 80-something, renewal.


A while ago I went on a shopping spree buying domains with the name of our project in various TLDs. I did consider .io, but decided to pass at $29 registration price. Now I'm glad I did, because while the money aren't great, I would have had to drop this domain on a principle. You know, to stick it to them!


I think raising the price of TLDs is the only way to stop domain squaters. When even the lowest price for a domain you're squating on is 2k you can sit and pay 50 a year for 50+ years and still come out on top.

Kinda sucks since I have some .io domains and I could only renew them as far as 2025.


> I think raising the price of TLDs is the only way to stop domain squaters

That does not make any sense in any world. If a squatter successfully makes money on a single domain, for example, they can reinvest on dozens or hundreds of domains next and therefore squat as much as they like just the same. And early, lucky squatters always exist. This does not solve anything. You are just pricing out people with legitimate domain name use and not much money, as usual.


It makes sense in this world, you just dislike it.

I can buy a domain that costs $50 a year and I can squat on that 40 years until I would be in the loss margin for a $2,000 resell value and $2,000 is minor money for resell values. You make it $200 and then it's only 10-years, 10-years is a long time but not that long if you look at how long some people have been squating on domains. You increase the risk of not making a profit.

This talk of outpricing people, people are buying domains for thousands for their pet projects, so let's remain in the world of here and now.


> This talk of outpricing people, people are buying domains for thousands for their pet projects, so let's remain in the world of here and now.

The world of "here and now" would be acknowledging that "thousands of dollars" outprices a significant majority of the world. Even if you take a US-centric view, that's going to outprice the majority of the population.

If you're working at McDonalds for $12/hr with a pet project on the side, a $1,000 domain would take 83 hours of work to pay for. That's two weeks and change of full-time work. You're not going to be able to pay for that while also paying rent and other bills.


> The world of "here and now" would be acknowledging that "thousands of dollars" outprices a significant majority of the world. Even if you take a US-centric view, that's going to outprice the majority of the population.

It's significant for me. But I would still be willing to spend it to work on something i cared about. Just like I was willing to spend money on forming a company to protect myself legally.

I wouldn't be surprised if a large portion of domains that are owned but not being used aren't truly being squated on for profit but have people sitting on them for side projects they never started/finished and the cost of holding the domain is so little that they'll pay out $20-50 a year. I say this owning about 40 domains and I am not selling domains or even advertising domains. I just bought a bunch in the last year for projects and then decided I didn't like that domain so bought another one. The cost is so low that I can do that without worrying and without thought.

> If you're working at McDonalds for $12/hr with a pet project on the side, a $1,000 domain would take 83 hours of work to pay for. That's two weeks and change of full-time work. You're not going to be able to pay for that while also paying rent and other bills.

If you're working at McDonalds for $12/hr you gotta bigger problems like being able to afford an apartment to stay in. Saying they can't afford luxuries such as domain names when they can't afford the basics such as somewhere to stay, food, etc. is pointless. They can't afford a domain just now.


Does ICANN already have rules around what prices registrars can charge for domains? It seems like that's the only solution in the current TLD system - ICANN needs to step in and prevent anti-consumer practices like year-over-year price increases that vastly outstrip inflation.


My .com renewal was $14, namecheap.


I missed moving my personal portfolio .co from GoDaddy to Namecheap when it renewed last week and got hit with a $55 bill.

EDIT: I'd really like to know why I got down voted on this. My renewal this year was twice the price. Everything at GoDaddy seems to be double what every body else charges and it's fair to be unimpressed with still having legacy products stuck at their company even if I didn't notice/it's my fault.

I had a $55 charge from them on my credit card. The page that told me what it was was buried under six clicks. It's shady.


GoDaddy will also be able charge your old expired credit card.

A few years ago my dad had mistakenly signed up for some dark pattern "premium" SEO/marketing service on GoDaddy for one of his domains. It was over $600 a year and was near useless.

Cancelling this "service" was behind multiple clicks. He gave up on cancelling mistakenly thinking that since his old card had expired he would have a chance to cancel at at a renewal.

Of course GoDaddy was able to charge another $629 on an expired credit card...

Moved all the domains in my family from GoDaddy to Namecheap with no regrets.


The problem with .com is that it is both popular and heavily squatted. You'll potentially need to use a non-English word or two or more words to describe your business or website.

.io is still squatted, but much easier game. And it makes sense in our field.

As an example, I wanted a common English word name for my startup. The .com is owned by a squatting firm that balked at a $1M dollar offer. They want eight figures, no less. The .io, on the other hand, is listed at mid-five figures. Considerable delta between those two. Once you start doing business and begin to grow, you can potentially negotiate to lease the .com and redirect. Set up a lease to buy if you exit.


Is there really much incentive to pay big bucks for the .com these days?

Most people finding your site will do so through google or other links right?

As for myself I hardly consider the real URL these days (except in some cases to check it’s right before entering credentials).


Wait, people lease domain names? Is that common? Doesn't that violate the TOS you agreed to with your registrar?


Perfectly legal. I know $100 billion dollar companies that lease domains. (I don't think I'm allowed to disclose further.)

It would be foolish to do anything except redirect though.


But if you re successful you ll inevitably need to buy the .com . Why not skip this unnecessary trouble


What’s the incentive to pay the hit bucks for the .com these days?

Most people finding your site will do so through google or other links right?

As for myself I hardly consider the real URL these days (except in some cases to check it’s right before entering credentials).


Cloudflare Registrar is "at cost". Hard to get cheaper than that.


Hmm, it made me thinking that there is one more way to get it cheaper than $8 - use "value of the currency". There are many EU countries with inflation problems, where you can buy domains using their local currency - for example PLN. Currently, 8 USD is worth 32.56 PLN [0]. There are few registrars where you can get .com for 30.99 PLN [1], which is a bit cheaper than $8.

[0]: https://www.google.com/search?client=firefox-b-d&q=8+USD+to+...

[1]: https://www.ovhcloud.com/pl/domains/tld/com/


Is it a good idea to have your domain and dns controlled by the same company though? Say something bad happens at Cloudflare - you're kinda stuffed.


If their DNS service goes down, I can transfer to another DNS in about 2h. I tested that. The bigger problem is if Cloudflare's AI bans my account and I lose access to their Registrar.

My hope is that Cloudflare is small enough that it still has human beings in control of their AI. Unlike Google's AI which can’t be bargained with, can’t be reasoned with, which doesn’t feel pity, or remorse, or sympathy. I have friends at Google, but I don't have any confidence that they will be able to help when their Terminator strikes.

I'm in the process of de-Googlifying my digital life as much as possible. Getting out of Google Domains into Cloudflare was the first step. Setting up backup accounts and workflows that will continue work after a catastrophic Google AI termination can be a bit tricky.


Cloudflare renews ".com" for $8.57 USD/year (if anyone is looking for cheaper; not sure if they are the cheapest). Namecheap seemed to have the lowest price for ".io" for a while.


With the proviso that you must use their DNS service.


The Base Cost of a .com is $8 then there a $4,000 annual ICANN fee for the Accredited Register such costs are spread over how ever many domains under their management of course name cheap has over 10 million domains under management so that fee is nothing per domain. It can price out small competitors though


That's because the .com registry charges <10$/year.


Which is still ridiculously high.


I understand the initial design of ccTLDs but can we just admit it's mostly a failure?

People use them as vanity URLs and there is little to no consistency in how they are used. .io, .ai and .tv are good examples. Then you have .ly where the local warlord can decide to ban your domain for no reasons.

At least .gov, .edu, .com and .org are somewhat consistent.


I have my .io domain registered until 2023 which gives me enough time to look for an alternative and migrate all my self-hosted services over.

What are the best TLDs these days to own that aren't owned by large PE companies just wanting to make a quick buck?


Related: I strongly prefer to renew yearly, so that I have a working routine for renewal.


This works: Make use of the "renew yearly" strategy, but make sure that at any given time you're already paid up for the next 2–3 years. This will give you a wide margin for error (and a comfortable timetable to fix them).


I bought my .io domain for like 5 USD. If renewal is 55, I might let it go.


Please cancel if you can. My .io domain is going to be extremely difficult to migrate off of so I feel a bit stuck.


Thanks to their last price increase I looked in to the previously not available .com and .fr for my side project and got them both for less than the .io I was using.


What is the groups preference for "best" registrar? Good balance of usability, available TLD's, pricing and security?


More reasons to move to a decentralised p2p web. 55 usd is a ridiculous amount for renewal.


The last time I suggested domains should be NFT's managed by a DAO i got a boat load of downvotes. But you have to admit, ENS's community governance model is superior to this. Its nice that domain owners have a vote in decision making process, where as here the community has no say at all.


> ENS's community governance model is superior to this.

ENS is no different and it is just you registering a .eth subdomain (collides with Ethiopia's three letter reserved TLD) in a decentralised blockchain but is on a centralised root with the key-holders sitting on a DAO.

The fact that 'trust' is involved is the vulnerability with these DAOs.


FYI Cloudflare currently charges $30 per year for .io domains.


Probably only for first year. Most TLDs are discounted significantly for the first year to get you in the door, then when you've built your business on the domain they hit you with the real fee when the renewal rolls around.


No, I confirmed that it is for all years. They deliberately don't make money on domain name registration. The use it to attract/retain business for their other services.


Chances are they'll have to follow this price increase, at least, if they don't want to start actively losing money on each domain purchase.


Sure, I was just pointing out that their current rates are cheaper than gandi, especially for renewals. So if someone wants to renew for multiple years it may make sense to switch.


This is why I only use .com and .{country code of stable country, like iceland}, as the likelihood of shenanigans on these is so low.

.io definitely is one of those cool CS street cred domains, but 2x- or 3xing the price is just ridiculous :(


For me, not having good domains available of hoarding is a bigger problem than paying $55/year. I'm glad there is a TLD that is useful and expensive.


OT, but I gotta say I love this retro-design (it's more visible on their homepage). A nice break from every website looking identical.


Why .io domains are so popular?


Because they are cool! Also identifies something as having developers or technical folk as the target audience. And because of it's popularity, it's one of the few generally trusted top-level domains (by both humans and mail daemons)


For me personally, it came down to the fact that $firstname.io was available, but .com, .org, .net, and whatever else I checked wasn't. Kind of regret it now, not so much for the price but because of the questionable political situation with the continued availability of the TLD doing forward.


I/O is used frequently in tech as an acronym for Input/Output. I suppose that's why it's popular for techy stuff like SaaS providers and such. It's also a moon of Jupiter which adds to the nerdgasm.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

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

Search: