Stupid is as stupid does
We should be closing down any new tld domain applications (ie, including new .com and .net names), not making things even worse.
This is simply a gross cash-raising bid by ICANN.
The number of top-level internet domains is set to double over the next few years, after ICANN today approved the launch of a program that will let any company apply to run dot-anything. During a meeting here at the Raffles City Convention Center in Singapore, ICANN's board of directors voted 13-1 with one abstention, to …
1. come up with a name for a brand or company
2. search for vacant domain name
4. 95% chance of fail
The domain name system is a namespace with a finite set. We're already reaching the limit of that marketable usable set without spending $5k - $5m for a 'good' name and it can only get worse.
It'd be nice if they limited .co.* and .com to registered companies and registered brand names but unfortunately they don't and cyber squatters sit on everything that's a dictionary word or abbrev. and anything once owned by a legitimate user.
This extension to the naming system will significantly relieve the pressure and diversify the eco system. I can only hope it decentralises the roots out of the US too. Less there, the better. No reason not to spread them across the rest of the world.
Actually it's worse than that. By allowing a free-for-all in the top level namespace, they are almost certainly making it harder for future governing bodies to invent *recognisable* new TLDs for some truly useful purpose. (I don't count "dot brand" as useful.)
Our only hope is that the marketing executives of the world realise that there is no commercial value in paying a million bucks for a name that says "this website is a bit dodgy". I'm not optimistic.
do we have to have a new domain for every marketing fart from some company that already has a proper domain name to hang a subdomain off? The first time I came across that was quite a while back with Psion and their www.series5mx.com (I think it was called). No immediate recognition of its link to Psion (unless you already knew about that), and no easy navigation back and forth to their other products; I told them about these downsides as I perceived them, but they basically just shrugged. Since then this misuse of DNS as a product catalog has exploded.
But ICANN is happy with this, since every such fart fills their coffers.
>This extension to the naming system will significantly relieve the pressure and diversify the eco system.
No. it just means that the registered owner of 'Example-brand' will need to register
.example-brand as well as example-brand.com example-brand.net example-brand.co.uk example-brand.org etc.
It does nothing to expand the namespace.
Limit .co.* ? When I registered my .co.uk (over a decade ago), there wasn't really much of an alternative. .eu did not exist, .net had restrictions that nobody was following, .org didn't really apply, .com is something I never want to be (plus expensive). Other countries have .fr, .ch, .de, .es, etc but for some reason the UK doesn't have a .uk on its own, or even a .en if it is to go by country and you wish to be pedantic (plus the obligatory .ni, .sc, .wa...).
I agree with the general point, however. It seems to me that this new cash cow is just going to create an unholy mess.
Not really an improvement either though. If they'd stuck to only giving TLD names to people who should have had them (.com for companies, .org for non-profits, etc) then we wouldn't have a problem now. I dare say the internet would be a higher quality place to. Unfortunatly it's 20 years too late to implement that fix. This is a band-aid for a wound that calls for stitches IMHO, but it's better than nothing.
What "thing" is it that gets "worse" because of this?
As far as I can see, the only people who might have a serious objection here would be the likes of Google. They'll be losing out a bit when banging "Hitachi" into your combo address-cum-search-bar thingy goes straight to the default landing page at ".hitachi", as it should, rather than giving them the chance to punt a load of ads for people flogging Hitachi kit.
> What "thing" is it that gets "worse" because of this?
Er, we have no idea whether the DNS root will scale up by a factor of several. This is utter foolishness from an engineering viewpoint and very close to a swindle from a financial viewpoint. Of all the stupid things ICANN has done, this is definitely the stupidest.
pfft. yeah right buddy.
and lets not use electricity either because we have no idea if it will scale with the growing population.
maybe you want to live in a cave and eat moldy fungal spores and bark all day long, but I'll take my chances with betterment, ta.
"What "thing" is it that gets "worse" because of this?"
I'll head out and buy some various great domain names, like the following:
.m and .om (for obvious above reasons)
Perhaps I'll have paypa.lcom and amazo.ncom in my list. If people can't tell the difference between:
what makes you think they can tell the difference with:
What a f-cking stupid idea.
It basically amounts to a tax on businesses who will be anxious to spend a serious amount of domain to acquire names that they don't really need.
On top of that, arbitrary TLDs are likely to screw up all sorts of systems (web filters for example) which will probably need to be upgraded or replaced.
I suspect that the only people who really want this are ICANN themselves who are rubbing their hands together at the prospect of a huge pile of cash coming their way.
It's like taking the UK post code system and trying to make it user friendly. It completely misses the point: the postcode it the computer-friendly bit, you tack on at the end, to help with automation - just as the domain names aren't really there to do a top class job of abstracting the user away from the nuts and bolts of the transport layer beneath. A domain name is one step up the abstraction ladder from an IP address (which is a layer of abstraction up, from a MAC number).
Nowadays, they mainly exist to allow a given domain to point, flexibly, to a variety of potential IP addresses - either because a site is in the process of being moved from one physical system to another, or because multiple machines handle the traffic that can be requested via a given domain.
The fact that a human can sort-of-parse a URL, to guess where it points to, is like observing that - with enough knowledge of UK postal districts - a human can sort-of-parse a UK post code and guess where a letter is headed. Really, ICANN would have better served their time, by trying to think of ways that the domain name server infrastructure could reliably (and securely) abstract users *further* away from the transport layer.
Ultimately, no widely-used system for human interaction is going to have its future in dot-delimited hierarchies of text strings, no matter how many varieties of text string you offer to use with it.
...will only have a few hundred extra entries, according to that article. Even if it gets into thousands, it's still a small static file to be served.
That doesn't make it any less of a stupid idea though, except for ICANN and the domain registrars who will make a tidy packet from it.
Actually, people have been selling new TLDs (and making a packet from it) for many years - even though they don't work. See www.new.net. They will presumably register their 75 fake TLDs, and thus turn themselves into a legitimate business. The point is, even when it was selling snake-oil, it still made tons of money!
most people (shamefully me included sometimes) just type the name of the site into google and get the domain from there.
I remember when .ltd.uk was created, companies bought them and then it went quiet, because those companies also had .co.uk and everybody went to that domain by default (remember .biz?).
I would've been more in favour of a limited release (like .movie) rather than a free-for-all.
"most people (shamefully me included sometimes) just type the name of the site into google and get the domain from there..."
Auuggghh. My wife does that. Every time I try to tell her that "Google is NOT the Web" and that more often than not, a try at name-of-site.com will bring up the site you want, she just pitches a fit. I've pretty much given up trying to teach her anything about the Internet.
Take for example the John Lewis chain.
Do you guess at johnlewis.com, johnlewis.co.uk, john-lewis.com etcetc?
No, you just go to google (or use your address bar search accelerator), enter john lewis and the proper one is dumped at you as the first result removing any worries about going to a place that shares the name.
I used John Lewis as my example because up until recently johnlewis.com was a small time family owned company with no affiliation to the famous John Lewis, but it seems as though they have sold their domain. No doubt there are plenty other examples like this though, like PCWorld.
Since I'm pretty much ignorant about how these things really work backstage... Wouldn't it be possible to force them to move ($domain =~ s/\.com/\.xxx/ and etc. kind of thing), and them somehow punish/unplug whomever broke the "porn on .xxx only" rule? Why not, technically speaking?
Of course there would be some arbitrariness in defining porn, so that might be a problem. E.g. would Playboy be considered porn? (I wouldn't, but some people sure would)
Do you really think the porn and gambling companies are not already aware of that? Do you really think blocking .xxx would make any difference to anything?
I imagine there are some extremely morally sound porn companies that would only use .xxx and limit their adverts to same. Hmmm, I'm impressed I managed to finished that sentence without ROFLing.
I wonder how much they'd charge for certain four letter expletives?
Would anybody want to buy a TLD with a synonym for excrement? :)
Or the alternative way of describing the current Culture Secretary - a word the Daily Wail regards as "most offensive"? :)
Then again, pr0n providers might jump at the chance to buy an anagram of the abbreviation of French Connection UK :)
Any justification based on "more room for new domains" is missing the point. These random new namespaces will be quickly filled with typosquatters just like the old ones.
A finite number of new TLDs for industries which are already regulated (like .bank) seems to be feasible; where there is already an agreed impartial entity who can determine the validity of joining that TLD.
Allowing the .brand TLDs just moves the competition out of .com and into the TLD space. Exactly the same fight, but now you need another $200k to stay in the game.
ICANN have got this completely wrong again. The solution to the domain name mess is not more top level domains, but to do away with them, and restrict people to using .countrycode domains only. If a particular country, eg .us, wants .porn.us then let them have it, it won't affect the rest of net.
Sounds like a good idea to me but I see no harm in allowing plain .<foo>. if the applicant can provide resolvable proof of international business interests. google.search.uk and google.search are both OK.
I did wonder if you could just register .<word> and host your side at www.word?
I agree that this could quiet easily lead to a name explosion where even less logic is involved in a domian then there is at present!
> I did wonder if you could just register .<word> and host your side at www.word?
Why make it so complicated? http://word should work just fine.
(geeknote: just use a CNAME to map word to www.whatever.example.com)
Of course, it would apparently cost $200k for that, but who can begrudge the ICANN
Board its many fine lunches and dinners?
Wouldn't it be a good idea to implement compulsory authentication of DNS names as it would do a lot to eliminate online fraud. Click on URL in browser, browser does a DNS lookup + verification->DNS server returns IP + temporary key, browser uses key to communicate with server.
... because this will now effectively give them sole and exclusive control over their company domains in all territories at a single stroke.
Instead of needing to trust, rely or bully the local distributor who managed to nab their name with the local country suffix way back in the day, they can now insist that all their genuine info goes under their own 'vanity' domain - making changing distributor rather easier, should it become necessary.
I suspect that Canon and Hitachi will be the tip of a very, very large iceberg.
On the plus side, customers going to such multinational-suffixed sites will reduce the risk of getting themselves scammed by a local con artist, I guess... (which is no guarantee that the multinational won't scam them itself, of course!)
If a major corp spots you using brandname-cameras.co.uk on your site then they'll just take it off you via Nominet, I've had Samsung, Canon, Hiatchi and many more pull the same stunt with our customers.
What do you care if they have your brand name in the URL? They're selling your products, leave them alone!
In Other News: .......It seems that "ICANN will not consider applications from individuals or sole proprietorships.".....Although perhaps it's more the case that individuals & sole proprietorships will no longer be considering ICANN......
So......play the TLD game without spending $185K (plus potentially unlimited annual expenses). as anyone can now create their own set of Top Level Domains at no cost and without reference to ICANN, simply by registering new Dashcom (instead of Dotcom) Domains.
Dashcoms are memorable & relevant web addresses such as "sports-com", "live-music", "social-network" (available in any language, you can even use Facebook Emoticons like musical notes "♫♫-♫♫").
Totally outside ICANN's control and with users in over 90 countries worldwide, resolution is via an APP; although new ISP links are coming online to make that unnecessary.
Alternate DNS roots have been tried before. They have no market penetration, so the names registered with them have no value.
By an interesting co-incidence, this is why non-dotcom domain names are practically valueless too; everyone will either search for the company in google, or they'll try a .com/.co.uk/whatever is your local gLTD. Anything worthwhile ever popped up at .biz? Didn't think so.
The problem with the current set of TLDs is that the namespace is massively polluted with 'what you want, when you want it' ad-pushing valueless spam sites. I bet they outnumber real domains by a few orders of magnitude. If .coms were restricted to require your local jurisdiction's equivalent of a limited business or trademark the amount of domaining is going to drop just a wee bit.
Just because an idea may not have achieved critical mass in the past, doesn't mean it won't in the future.
Not-so-long-ago, people would have thought the Internet itself to be a complete waste of time effort and money. Surely a retailer, ANY retailer, (eg: a certain bookstore) would be mad to waste time, money and resources trying to sell anything over the Inter-Something? After all why would any consumer in their right mind even consider purchasing a vastly expensive computer, install/rent new phone lines, buy a modem (what's a modem?), buy an OS (what's an OS??), learn how to use it all.....Just to acquire a book???.....All they had to do was order one over the phone.
Your analogy is so invalid I'm falling asleep right here. Was there a perfectly functional existing system like the internet already in existence when the internet was invented? No, no there wasn't. So your analogy of your stupid "dashcom" alternate DNS system to "the internet" completely falls over.
Your sole argument being invalid, we're then still at the state of the previous poster being the last word on this discussion. These retarded ideas have been tried before, and failed, for the same reason this one will. Mostly: because they're retarded. There is no problem to solve (from the end users point of view, which is where the votes will be cast, as it is they who provide your currency (aka traffic)).
> Just because an idea may not have achieved critical mass in the past, doesn't mean it won't in the future.
oh for fuck's sake!
it is inherent to any hierarchical system that it has a single, unique root. these so-called alternate roots are nothing of the sort. please read rfc2826.
the current root is managed by icann. and although this is a badly broken institution, it is at least open and transparent. it gets participation from most stakeholders -- registries, registrars, isps, telcos, business, law enforcement, governments, trademark people, ngos, etc. they are able to make policy and to have their views represented. there are regular independent reviews of icann's processes and operations too. none of this is remotely true for the alternate root ding-a-lings.
now maybe one day someone or something will come along and kill icann. (please make it so, pretty please!) perhaps the flying spagetti monster will do this when our bunny rabbit overlords take over the planet. that will still leave us with a single organisation which manages the root of the dns.
the alternate root loonies are going to be seriously fucked by icann's announcement. a few hundred new tlds will almost certainly mean name space collisions with the bogus tlds in these alternate roots.
Making more TLD's is pointless nowadays, it used to be so you could tell 'what sort of site it was'. Surely people just type what they want to find in their preferred search engine. I'm not sure even I care what a site is called any more* as the domain names get more and more 'creative' not for creativity's sake, but because the domain name is not already taken.
*the only sites I remember are www.google.com and www.theregister.co.uk**
ICANN is fast becoming less and less relevant, they want the credit of "improving" what we in the IETF did going back to Vixie's original RFC on the DNS (Domain Name Service) by making the namespace commercial.
For the layman, the TLD's are like a country code, and the domain name like a phone number. Set by consensus originally based on ISO codes, these self-proclaimed policy makers are ruled by the Almighty Dollar Bank and Trust.
What these "so-called intelligent representatives" fail to take into account is two-fold. First, the ITU is about to launch a full-scale land grab to take over domains and IPv4/IPv6 addresses just as the do with telephone numbers/country codes. Second, as a pseudo-elected body, they have no real technical implementation skills as IETF members do.
These policy makers are idiots, pure and simple. We can have any TLD we want, it is just a matter of will the root servers (the master name servers) accept a TLD handed to them (yes and no) and will the domains propagate to other DNS servers.
There is going to be a HUGE backlash, where IT administrators (or their policy makers/legal/HR) departments instruct those IT staff to "block certain TLD's" in their company/ISP policy.
So just as you can block a phone number inbound/outbound, the same will happen to these new TLD's.
Net loss, epic fail (again!) for ICANN......
"...at the Raffles Convention Center in Singapore..."
Why am I not surprised by the location of this shindig? You couldn't see the ICANN boys at the Travelodge, Wolverhampton, would you?
I'm in the wrong job, all the conferences I have to attend are at shitty Hyatts an Hiltons in Northern California.
The whole concept of TLDs was to have a small - repeat small - forest of strictly hierarchical trees that could be easily parsed and verified. Instead we'll end up with a massive jungle of inconsistencies that will unverifiable, opening the doors to confusion, malicious abuse and plain old-fashioned error.
Even the basic task of parsing a user-supplied URL for validity on a web form will be impossible, as there will be no predictable extent or content for the TLD.
But what the heck? We already have the same problem with RFC 5322 - compliant email addresses - right from the start. The entire infrastructure is creaking to ruin as we bow low to the almighty buck.
I'm already seeing mail admins look at this situation and asking how to configure resolvers to only accept the existing set of TLDs (presumably while still getting updates to those, instead of just serving the root zone locally).
Imagine, your university's school of business uses "business.example.ac.uk" and people internally are used to mailing around to <fred@business>; what happens when "business." becomes a TLD? We know to not use existing TLDs for sub-domains, but when anything can become a TLD, all bets are off.
If you thought .biz got you high spam scores and wasn't worth trying to communicate with, you haven't seen anything yet.
Biting the hand that feeds IT © 1998–2021