Talk:Domain name

Intercapping / Caveat Emptor again

If you click on the links in the "caveat emptor" section, it becomes clear that only two of the five sites are genuine. The genuine ones are www.whorepresents.com and www.therapistfinder.com. Of the other three: Expert's Exchange is now at www.experts-exchange.com; http://web.archive.org/ proves it used to lack the hyphen, though the hyphenated version predates it by 2 years, so this is arguably somewhat an urban legend too. penisland.net is clearly a hoax if you read the section about "long and skinny pens, thick pens, even black pens", and notice that most of the internal links don't work. www.powergenitalia.com isn't a working website, and there's actually a Snope's page up about how it's not connected to the power company Powergen.

All five are good examples of phrases that can read differently when they're converted into domain names, but I'm going to rewrite the section so that it gets the facts surrounding them straight.--71.198.157.236 18:37, 8 January 2006 (UTC)

On secound thought I just removed penisland.net and powergenitalia.com from the section, because I couldn't think of a short way to summarize their fictional status. I think we only need one or two examples of this phenomenon in order to illustrate it, so they may as well be examples that actually occurred. If we started putting down the (non-notable) hoaxes as well, we could wind up with an endless list, because it's not that hard of a word game to come up with these ambigous phrases. For example, when I see the URL www.letthemeatcake.com, I can't help but read it as, "let the meat-cake". --71.198.157.236 18:49, 8 January 2006 (UTC)

<years pass...> The Intercapping info is non-obvious and valuable, so I undid the last edit, which removed it.--Elvey 22:10, 11 November 2007 (UTC)

RFC 1034

RFC 1034 provides a good, authoritative explanation of domain name lingo. e.g.: "A DNS query names the domain name of interest and describes the type of resource information that is desired. For example, the Internet uses some of its domain names to identify hosts; queries for address resources return Internet host addresses.

This article should be short (i.e. disambiguation size) and prominently refer to the already thorough DNS article, which should incorporate some of the content here (e.g. TLD list, most other stuff). Duplication of Effort, otherwise. Elvey (later modified) — Preceding undated comment added 20:02, 17 December 2005 (UTC)

I disagree. I think the DNS article should focus on the technical aspects while this article should focus on the commercial aspects of the "dot com". —Preceding unsigned comment added by 200.114.201.109 (talk • contribs) 14:28, 17 December 2005

Actually, I don't see how creating/maintaining such a split serves the average reader. That's my PRIMARY reason for supporting relocation of most of this articles' content. Non-technical aspects (e.g. history, the domain name industry, and trivia about domains, are all about the DNS. Elvey — Preceding undated comment added 20:02, 17 December 2005 (UTC)

On the contrary, the average reader wanting to know more about "domain names" does not need to know the technical aspects of how the whole system works. This is analogous to an average reader wanting to know more about cars, without being inundated with the technical aspects of how a car actually runs. —Preceding unsigned comment added by 24.232.217.51 (talk • contribs) 21:47, 21 December 2005

Citing this, and other RFCs, as my authority, I'm re-working the first bullet point of the introduction. A hostname is a particular kind of domain name. It is not improper usage to refer to a hostname as a domain name. --kop 18:48, 28 February 2007 (UTC)

---

I'm glad for the reference to 1034 here. I went to the article to find out why special characters are not allowed, and only through that RFC did I find the answer:

---

Node labels which use special characters, leading digits, etc., are likely to break older software which depends on more restrictive choices.

---

I think this is important enough to add to the article. But I'll wait for someone more familiar with these internet articles in wikipedia to decide that.

(Note: Any theoretical software that this would break would have to be some very, very... almost deliberately malformed code ... but it's possible. But that is a discussion for a different place.) Robbiemuffin (talk) 18:58, 2 March 2008 (UTC)

This subject is explained more in both the hostname and domain name system article, both of which are mentioned in the leading paragraph about where to find more detailed information. I do not think that every article needs to discuss all issues, so I would say it would be bad to add this stuff in yet another place. Wrs1864 (talk) 21:07, 2 March 2008 (UTC)

Deleted content

Serenade, Ltd.

I removed the following paragraph from the section "Commercial resale of domain names". It didn't seem to fit with the article, especially since the linked Google search doesn't seem to validate what he/she's saying. --216.195.163.105 21:57, 24 Jun 2005 (UTC)

United Arab Emirates based company Serenade Ltd. is an example of 'warehousing' thousands of domain names for future profit. Each of the thousands of sites displays a page titled "World News Network" standardized news gathering array is presented as the main page on each of thier thousands of sites. Numerous lawsuits have been mounted by individuals, small business, and even local government to gain control of these domain names, all of which have failed. Google "Serenade Ltd" and see for yourself. All the news comes from Reuters and the like, and no company details are found easily on any of the sites. —Preceding unsigned comment added by 216.195.163.105 (talk • contribs) 21:57, 24 June 2005

I bet there's an article where content like that would be more appropriate than this one, e.g. DNS Elvey — Preceding undated comment added 23:20, 6 November 2005 (UTC)

Unique identification

The page says that a domain name distinguishes any one computer on the Internet from any other, which seems incorrect to me.

A domain name is simply a name which can hold a number of Resource Records. These RRs need not correspond to an Internet host in any way (they can, for example, be Hesiod data, SRV records, TXT records, etc.), nor need they name one unique host.

For example, look at www.wikipedia.org:

$ host -t a www.wikipedia.org
www.wikipedia.org has address 207.142.131.202
www.wikipedia.org has address 207.142.131.203
www.wikipedia.org has address 207.142.131.204
www.wikipedia.org has address 207.142.131.205

As you can see, the domain name "www.wikipedia.org" identifies 4 different Internet addresses. Also, look at, for example, _service._dns-sd._udp.dns-sd.org or 4.0.41.198.in-addr.arpa, and you'll find them only containing PTR records (no host data, in other words).

If noone objects to this, I'll try and change the article after a few days. It is equally possible that I'll forget it, however, so if I haven't, please do so for me. ;-) Today is 2005-06-30.

--Dolda2000 30 June 2005 17:42 (UTC)

Dolda2000 is correct, I cannot come up with good verbiage to replace the existing false one but it is propagating too (I came here after going Huh? on the Turkish one). —Preceding unsigned comment added by 85.101.251.76 (talk • contribs) 18:15, 18 August 2005

No Answers Here

I came here looking for answers but did not get them...

1. the bit about the final dot did not make sense to me
2. the idea that an IP address can refer to many Domain names and a Domain name can refer to many ip addresses is confusing if it is true. How then would you ensure that two packets taking different routes to the same domain actually arrive at the same buffer? —Preceding unsigned comment added by Bschmidt (talk • contribs) 20:58, 26 August 2005
   1. I do not understand what you are talking about here. An IP address does not at all refer to a domain name, and domain names have nothing to do with routing. Would you mind clarifying what you mean? --Dolda2000 16:36, 13 December 2005 (UTC)
3. co.uk translates to "company" "United Kingdom" - the co is more similar to com than to a country. Bschmidt 21:04, 26 August 2005 (UTC)
4. IANA should get mentioned here. Bschmidt 21:04, 26 August 2005 (UTC)
5. I wanted to know about .gov - what is up with that? —Preceding unsigned comment added by Bschmidt (talk • contribs) 21:08, 26 August 2005

Would it be okay per Wikipedia to challenge somebody at INTERNIC to write this page?Bschmidt 21:04, 26 August 2005 (UTC)

1- The ending dot is required by some protocols (DNS query), to mark the end of a domain name. You should see it as the root of domains: i.e. if you search the nameserver for TLD "com" or "ch", you ask the root nameserver (at the level of the .<void>), and this root nameserver give you the nameserver responsible to the asked TLD. You see? There are a level above the .com (this is named as the final dot). But is not necesary, is more an internal view of name resolver, so it is optional. The final dot is used also in some programs to distinguish a fully qualified domain name from a relative domain name (resolvers, by default, search first in the local domain and than full qualified) or other non domain item.

2- This is true, and it should be documented in an IP address article. Anyway some examples:

a-: one machine, more IP: a computer (normally a routers) can be connected to two or more nets, so one interface is on the net 10.0.5.x (i.e. the net the link other nets), so it needs an IP address in this net (i.e. 10.0.5.182). The other interface is on an other net (local net, LAN) 192.168.205.x, so it need an other IP address (i.e. 192.168.205.12). How to pass packets to one net to the other, if you don't have a machine with more IPs? (note that IP is linked to an interface, not to computer).

b-: one IP, more machines: You find it usually in NAT, but also on some redundant machines. In NAT you have, from your ionternet provider few IPs. But you have more computers. So you build an internal net with a private IP net (and addresses). Your router makes NAT, translating public connection (to public IP and a port to/from external IP and port) to a valid internal connection (internal/private IP and por to/from external point). So from internal side, the machines behave normally, and from external side, other see your net as a single machine. There are some limitation because your net (from extern) should behave as a single machine, and you should configure carefully your NAT, to see what internal machines should process external connections.

-- Cate 08:38, 27 August 2005 (UTC)

3- I think it is an error. Corrected Cate 13:51, 27 August 2005 (UTC)

4- IANA: yes it should be mentioned, as normal root server

Someone need to explain or give a link to alternative root servers, but in a way NOT to confuse the readers. —Preceding unsigned comment added by 212.72.197.166 (talk • contribs) 21:39, 20 October 2005

5- ?? check .gov

-- Cate 13:51, 27 August 2005 (UTC)

.GOV is USA Government. When domain system was developed, they think it will be used in USA only. This is explanation of COM/NET/ORG/EDU/GOV. Because .GOV can be registred only from USA government organisations, some countries made own government subdomain. Turkey is very good example for it - .gov.tr (Turkey copy the TLD idea and they have also .com.tr/.net.tr/.org.tr/.edu.tr/.k12.tr/.gen.tr) —Preceding unsigned comment added by 212.72.197.166 (talk • contribs) 21:36, 20 October 2005

Is there any truth to mtpf.gov as being the only non-US entity to ever have had .gov (if it existed, it's gone now as the former "Metro Toronto police force" is now simply torontopolice.ca) --66.102.80.223 04:47, 21 October 2007 (UTC)

"Registered Domains"???

What's the term for the set of domains that includes example.com, example.co.uk and example.fr but excludes their subdomains, i.e. just the domains that were registered with a normal registrar? "Registered Domains/RDs"? Do we need to coin a term for this? If we do, I think we want a term that generally excludes free domains (Otherwise, we can expect, e.g .info in a lot of CSV-authorized spam.) (We would include domains that are free, but regulated in a way that introduces an adequate non-cash cost, e.g. domains in .edu may fall into this category.) (Domains in cjb.net would not be included, even if cjb.net was well-regulated; if it was well regulated, then cjb.net would be the RD set). Perhaps domains that return a valid result from a good whois client such as GNU jwhois (which supports a result cache) or something that uses its config file (/etc/jwhois.conf) would be a good tool/proxy for identifying this set ('Whoisable Domains'? ). (An (expensive) whois lookup by a reputation service would be reasonable, if done once by services in the process of determining the reputation of an ?RD?.) It's about defining a Responsible Entity's Domain... It would be expected that (all? most?) reputation services would consider all subdomains of an RD to have the reputation of the RD. Some accreditation services would probably be willing to accredit subdomains of an RD separately. - Elvey 18:45, 7 October 2005 (UTC)

Commercial resale of domain names

Link removed

This was added to external links (and I removed it):

• TheRecordBreakngDomain.com - Attempting to Break the Guinness World Record for Most Expensive Internet Domain Name. —Preceding unsigned comment added by Just zis Guy, you know? (talk • contribs) 09:44, 8 November 2005

No one is going to buy this domain, so no money is going to go to the owner (other than via content (e.g. ads) foisted on visitors) or charity. Har har.

Someone had removed the parentheses mentioning TheRecordBreakingDomain.com's attempt to sell for more than business.com to break the world record. As silly as the attempt may seem, it has been legitimized by Guinness World Records and the media, and is therefore relevant within the context as a side-note to the most expensive domain. I agree with comments below that it should not be included in the External Links section as it is too specific in relation to the article as a whole. —Preceding unsigned comment added by 200.114.201.109 (talk • contribs) 14:28, 17 December 2005

Someone continues to remove the side-note mentioning TheRecordBreakngDomain.com. Can someone please give their reasons for doing this or where the side-note lacks relevance and impartiality? —Preceding unsigned comment added by 24.232.217.51 (talk • contribs) 21:42, 21 December 2005

"thefreakinrecordbreakindomainname.com isn't notable - it may be, once it works, but before that you're just looking for publicity. Removed." - that's the comment left by the last remover. I was going to remove it too, but couldn't think of a good explanation, and it's a lark. The remover provided one though, IMO. —Preceding unsigned comment added by Elvey (talk • contribs) 04:49, 22 December 2005

Was re-added(anon IP) and re-removed(Elvey). — Preceding undated comment added 04:48, 26 December 2005 (UTC)

If there's a next time, this goes on Talk:Spam_blacklist === TheRecordBreakingDomain.com === Keeps being spammed to http://en.wikipedia.org/wiki/Domain_name - from different IPs. Yes, this is a judgement call; see comments at http://en.wikipedia.org/wiki/Talk:Domain_name —Preceding unsigned comment added by Elvey (talk • contribs) 05:17, 26 December 2005

Happy to join the discussion, I think this is pretty relevant if you read through the author's site. If an article on Houses mentioned the most expensive house, wouldn't it be relevant to mention the most expensive house currently for sale? I don't know, just my two cents. —Preceding unsigned comment added by 201.217.142.161 (talk • contribs) 00:07, 30 December 2005

As already said by someone else: "thefreakinrecordbreakindomainname.com isn't notable - it may be, once it works, but before that you're just looking for publicity. There are plenty of domains on offer for more than than this one. (I'm offering one for sale for a price defined as twice the then current price of thefreakinrecordbreakindomainname.com. Oh and you have to pay me $1mm just to find out the name of the domain. Silly? So is the argument that thefreakinrecordbreakindomainname is notable.) —Preceding unsigned comment added by Elvey (talk • contribs) 05:42, 10 January 2006

How about including a list of the 5 or 10 most expensive domain name purchases? —Preceding unsigned comment added by JMS1225 (talk • contribs) 18:21, 27 April 2007

Go ahead.   — Jeff G. (talk|contribs) 18:11, 17 May 2007 (UTC)

Caveat Emptor

Someone added Gina Shoes (www.ginashoes.com) as an example of a domain name which is open to (potentially comic) misinterpretation. I'm obviously having a senior moment because I can't see why that would be funny. - Just zis Guy, you know? 10:02, 2 November 2005 (UTC)

I think they're interpreting it as "Gina's Hoes". *Dan T.* 13:04, 2 November 2005 (UTC)

They might need four candles then. - Just zis Guy, you know? 13:08, 2 November 2005 (UTC)

Isn't Pen Island a joke site? samd 03:25, 28 November 2005 (UTC)

Why the www subdomain ?

I would like to know, and I think it's of common interest, why many domains need to be prepended the famous www. to be accessible over http.

It seems like it comes from history, where all the ressources where distributed on different computers, and thus, every would gain a name according to it's service (eg. smtp, pop, ftp, ...).

In the common age, where the www has gained so much momentum over other protocols (eg. nntp), does it still make sense to have this www. in front of every domain ? Also, lots of websites use subdomains to separate their websites in logical manners so it doesn't have much to do with services anymore. Many big websites are using reverse-proxies and round-robbin, so a domain doesn't really represent a machine anymore.

I would be happy to have more informations about this and integrate it in the article in a nice manner.

Cheers, zimbatm — Preceding undated comment added 00:29, 1 December 2005 (UTC)

The widespread usage of the www prefix is indeed for historical reasons. However, if you consider an entire network domain that happens to have an HTTP server, it makes sense to name that server www.yourdomain. Of course, HTTP-only domains don't actually need the www prefix, but in the early days of the web, there were no HTTP-only domains. Therefore, it can be said that the www prefix exists mainly for the sake of convention. --Dolda2000 16:51, 13 December 2005 (UTC)

There's no official standard that I know of that requires any service to be at a particular hostname, so it's just a matter of tradition that Web sites are normally at "www." hostnames, FTP servers at "ftp.", and mail servers at "mail.", and so on. Not all sites conform to these traditions, and it's entirely under the control of the system administrator at each site. These days, most Web addresses work identically with or without the "www", and some sites (especially ones with subdomains, like Wikipedia) prefer to cite their addresses without it. *Dan T.* 01:25, 1 December 2005 (UTC)

No. Every computer has a name, but than to distinguish what computer serves some services, you create some aliases (CNAME) or other DNS records (MX for mailbox). Some RFCs (but with status "informational", not "standard") propose some name schemes (www, ftp,...). BTW domain name are introduced not force users to learn/remember difficult IPs, so there is also no need (from the begining of domain names), to force user to remember also the machine names. Cate 08:45, 1 December 2005 (UTC)

I don't agree that it is simply used for "historical" reasons, I believe there are underlying reasons, many of them I idenified in an article I wrote on this very subject here: no www? - I did attempt to sit on the fence, but as you can see I am pro-www. Feel free to offer feedback on my article.--Hm2k 17:59, 13 March 2006 (UTC)

A few comments on your article:

"Having www helps clarify the purpose of the url. ie: www is world wide web, webmail is for web based email, etc" - Technically webmail is still the WWW, so using the "webmail" subdomain isn't really correct. It's the "http://" that really tells your application which (other) app to start, so it doesn't matter, but that still sort of invalidates this argument. Of course behind the scenes you should always have the protocol, even if you don't render it, for instance <a href="http://www.whatever.com/">(www.)whatever.com</a>. This also applies to: "example.com is often not detected as a URL"

"The major websites still use the www. prefix." - The majority of websites (which probably includes "the major" ones) uses bad HTML, but that's not a real reason for everybody else to do it. I understand the argument of course, I'm not saying that it doesn't make sense.

"www has been used since the dawn of the world wide web why would we stop using it now?" - That's not an argument on its own. If I'd discover I've been paying too much for something, I won't care for how long I've been doing it, I'd still stop doing it rightaway.

"When communiticating URLs its shorter, easier and quicker to use www.example.com vs. http://example.com/" - Just "example.com" will be recognised as a URL by everyone with an internet connection soon enough. Some website names don't even resemble their domain names, so this won't always work, and if you really want a functional URL, it's not that much trouble to type out the "http://" and trailing "/", especially in printed media you mention: "Using the www. prefix helps readers of printed advertisements define what it is.". Also "domains with unusual TLDs may not be idenfied" is pretty much the same thing again.

"Removing the www. would not make sense, it would make more sense to change the prefix subdomain to web. allowing this to be the new standard." - This seems more like an opinion. This still leaves all the problems associated with "www.", like why not mail to "somebody@web.domain.com" instead of without the "web.". If I were a proponent of the "www.", which I'm (currently anyway) not, I'd say changing it now would be more trouble than it's worth.

"People still use and want the www. prefix the same way they would use the ftp. prefix or wap. prefix." - I know a bunch of ftp://www.domain.com/ addresses, which clearly indicate that it's the FTP method of accessing the same site. This renders the www useless. Again, the "ftp://" tells people what the URL is. Besides, I don't recall any applications that turn "ftp.domain.com" into a clickable link of any kind. "wap." makes sense, since those pages are actually different pages from the "www." ones (although they may use the same database for content). As for people wanting to use it, that seems more like an opinion to me.

"Sometimes the content of example.com is different than the content of www.example.com" - True, but I haven't seen this happen often, so this should be easily changable.

Like I mentioned, I'm not fully decided on the issue, but I'm leaning heavily toward not wanting to use the "www.". Does anybody else think Wikipedia deserves a whole article dedicated to this issue? It seems like it's big enough to warrant one. Retodon8 15:45, 27 November 2006 (UTC)

In fact, it is for historical reasons. Load balancers, reverse proxies, netcache appliances, spanning and especially NAT made the "www" subdomain mostly technically obsolete. To distinguish different services, you can use the ports anyway. For different types of access (e.g. WAP), mime-types are a preferred way. however, it allows an easy and classic possibility of service seperation over different hosts in the lower- and midrange. it's good the way it is - enables both and there's no practical need for an extreme. I don't think that an explaination would lead to anything else than a discussion like this. "www" is neither wrong nor as needed as it used to, but it still is a fully quantified hostname that's up to the owner. there's no rfc or standard that wants us to reserve "www". There are domains that don't promote or don't want the "www" like slashdot.org, digg or del.icio.us Dork. 22:07, 1 April 2006 (UTC)

I've discovered more so recently that is more of a media thing. I noticed that certain companies are dropping the www. from their "url", in much the same way that people began to drop the "http://" and the trailing "/" on urls for shortness (the "http:// is usually added by the browser, and dropping the trailing slash makes your web server work harder). However, although this means its shorter to write, it can sometimes be difficult to identify that it is infact a domain name, a good example of this is "del.icio.us", it can also be noted that "www.del.icio.us" does not work, however "www.slashdot.org" redirects to "slashdot.org", visit "google.com" it redirects to "www.google.com, same happens with yahoo, take from that what you will. Another example of this was where I saw "manchester.gov.uk" written on a parking ticket, yet if you visit that in your browser, it actually does not work, you are forced to search for it or add the www. prefix. The no www. thing works in media if you have a common tld (.com|.org|.net) otherwise you're expected to understand the internet at an advanced level. I think using the www. prefix helps people to idenify that they are actually taking about a website rather than something else. Fact is there's many ways to write the same thing... but so people understand what we're talking about its best to use www.example.com as opposed to example.com unless you use http://example.com:80/ or something, which just ends up longer. There's still lots that could be said about the www. prefix, like the fact that keeping it for historic purposes is a good idea, as everyone is fimilar with that. --Hm2k 16:38, 2 April 2006 (UTC)

The dropping of 'www.' from a domain used for HTTP is a very old thing - that was happening nearly ten years ago because the browsers then were able to automatically add '.com', but not always the 'www.'. But in any case, this highlights an error on the page - in the URL http://www.example.com/, both 'www.example.com', and 'example.com', are domain names, technically. 'www.example.com' is also a hostname. (But, of course, so is 'example.com', most likely.). For what it's worth, a hostname is merely a domain name that can be used to obtain an IP address, usually via an A record, or a CNAME pointing to another domain name with an A record. That is, of course, assuming that the hostname is a domain name at all. Outside of DNS, hostnames usually are not domain names, and domain names are usually not hostnames. Finally, 'example.com' is a subdomain. As is 'com'. Just in case you thought that 'subdomain' was at least safe. All the TLDs are, of course, subdomains of the root. —Preceding unsigned comment added by 217.155.137.62 (talk • contribs) 23:49, 22 April 2006

Technically not, a qualified domain is like: "example.com", therefore a subdomain would be like: "sub.example.com", your use of subdomain is used loosely, but is incorrect terminology for this technology. But yes, the www. prefix would represent the hostname for the web server, which is why it should be used. --Hm2k 13:43, 7 June 2006 (UTC)

The .mobi domain guidelines indicate that a site should respond to web requests on the second-level (example.mobi) domain without requiring www. - this is because the target market is mobile devices. Otherwise, browers like Firefox do add www. themselves and retry if they don't get a response on just-plain example.org --66.102.80.223 04:53, 21 October 2007 (UTC)

Merge (?)

While there is a clear difference between a domain name and the domain name system, I'm very unsure whether the difference is large enough to warrant an entirely own page for only domain names. I would say that this claim is at least partially warranted by the fact that this article contains a rather large amount of fluff. Therefore, I would at the very least like it discussed if this article shouldn't be merged into the domain name system article. --Dolda2000 17:41, 13 December 2005 (UTC)

I think there's a distinction, however; the DNS refers to the technical mechanism whereby domain names are resolved, while domain names themselves have many social, cultural, economic, legal, and political aspects that don't relate to such technical issues. Thus, there's justification for separate articles, although this topical separation may not be consistently observed in the articles as they currently stand. *Dan T.* 17:53, 13 December 2005 (UTC)

Indeed, there is a clear distinction, but there will also be a very large overlap. For example, it wouldn't be unreasonable to include such things as the finer structure of domain names, rooting (last dot) of domain names, FQDNs and lots of other stuff in both this article and the DNS article. If this article should remain, I think there should at least be a very clear notice that it only deals with the social issues that you mentioned, and that all technical details are discussed in the DNS article. --Dolda2000 19:07, 13 December 2005 (UTC)

I have registered to participate in this discussion specially, but have found that *Dan T.* has already given the most important argument contra. The second one is size of both articles and probable size of the resulting article. I'll probably be able to add something after I read both articles again. Stan0 11:11, 17 December 2005 (UTC)

Judging from the amount of data in both articles I think that alone warrants enough reason for each of them to have their own page. Also judging by the amount of backlinks to the domain name article, I could NOT see a good reason to merge this article into domain name system, if anything it should be the other way around. However, I do not believe they should merge, as to the average user a domain name is simply what they use to get onto a website, DNS is far too technical for them to understand. Further more most people in the industry will tell you that there's a clearly defined line between a domain and DNS, or are we to propose that we merge this IP_Address with the DNS article too? --HM2K 12:35, 10 January 2006 (UTC)

One is more technical, the other is about an industry in some ways. —Preceding unsigned comment added by 66.68.148.220 (talk • contribs) 05:07, 2 February 2006

I agree that they should -not- be merged together. There is a very distinct difference between a domain name and DNS. Readers who are looking up "what is a domain name" will not be concerned with the naming system and how it traverses across the Internet. The same applies in the vice-versa situation. I think that HM2K and Dtobias have more or less summarized what my contributions to this discussion would be. TNLTRPB 10:07, 6 April 2006 (UTC)

Should define "domain parking" and "pointing"

The commonly used Cpanel has these features, but I can't get a clear defintion for for them anywhere. —Preceding unsigned comment added by Jonfernquest (talk • contribs) 04:56, 27 December 2005

Taken from cPanel itself: Domain pointers allow you to "point" or "park" additional domain names to your existing hosting account. This will allow users to also reach your website when entering the "parked" or "pointed" domain into their browsers. Also see: Parked Domains cPanel Documentation. In simple terms it usually means that the domain your point/park will have exactly the same web content as the original domain. --HM2K 12:24, 10 January 2006 (UTC)

Sponsored/Unsponsored

Say what Sponsored/Unsponsored mean. —Preceding unsigned comment added by 210.201.31.246 (talk • contribs) 21:48, 16 February 2006

Never heard of it, referances please. —Preceding unsigned comment added by Hm2k (talk • contribs) 12:40, 14 April 2006

See sponsored top-level domains and unsponsored top-level domains. In short: unsponsored: every one can register a domain, sponsored: there are some rules to apply (topic, domainname) Cate 16:42, 16 August 2006 (UTC)

Graph of cumulative registrations

Does anybody know where to find a graph of the number of cumulatively registered domains, from say 1990-2006? I think it would make a really interesting addition to this page - however, I'm not sure where to find the data for it. Nicholasink 04:47, 21 March 2006 (UTC)

The .orb top domain?

A few years ago, there were discussions about implementing the ".orb" top domain for space stations and other objects in orbit. I can't find any references to it here. Additional information on this topic would be interesting! 11:47, 26 September 2006 (CET) —Preceding unsigned comment added by 213.100.17.254 (talk • contribs)

Numerical domain names?

The article currently says:

"Later it became permissible for labels to commence with a digit (but not for domain names to be entirely numeric), and for labels to contain internal underscores, but support for such domain names is uneven."

So, domain names may start with a digit now, but that isn't always supported (server-side? browser-dependent?). Fully numerical names obviously start with a digit, therefore fall under that category, and from what I gather from the article aren't even allowed. However, http://www.4815162342.com/ seems to work just fine. Is such a name actually allowed at this date or not? Thanks for any info! Retodon8 14:10, 27 November 2006 (UTC)

Technically it is supported (but some application bugs). But some domain authorities have some restriction on second level domain names. It is common to forbid small numbers (and one or two letters long domains). But some old/existent domains brake these riles. In short: technical possible, but some time restricted. Cate 14:25, 27 November 2006 (UTC)

Certainly it does exist, 411.ca and 411.info would be prime examples. What can't exist is a domain name where *every* portion is a number, such as http: 192.168.0.1 - those are treated as IP addresses, not domain names. Keep the numbers out of the top-level domain (for instance, .org is valid .000 is not) and the rest should fall into place.

Internal underscores, however, are not valid in ASCII domain names. Only letters, numbers and embedded hyphens are accepted, with restrictions on hyphens due to the standardised use of prefixes like xn-- for special purposes. --carlb 23:02, 9 October 2007 (UTC)

Avoiding example.com

Ahoerstemeier, why would you want to avoid example.com anywhere?

Also, I noticed that you edited someone else's comment for clarity, spelling or grammar. As a rule, please refrain from editing others' comments without their permission. Though it may appear helpful to correct typing errors, grammar, etc., please do not go out of your way to bring talk pages to publishing standards, since it is not terribly productive and will tend to irritate the users whose comments you are correcting. For more details, see Wikipedia:Talk page guidelines. Thanks, -- Jeff G. 18:24, 12 January 2007 (UTC)

Because if we minimize the number of pages having a weblink to example.com, then this List will be a tool to find and clean editing experiment. If there are however 500 pages which contain that link for whatever reason it is unusable as such a tool. I don't think there is any harm in exchanging example.com with example.org. andy 17:38, 15 January 2007 (UTC)

I still don't understand exactly what you're trying to accomplish; what "experiments" are you trying to "clean up"? But I do prefer using example.org or example.net myself anyway, in order to do my part to combat the common "dot-com-itis" whereby people seem to think that everything is, or should be, a .com domain even where this makes no sense. The RFC on "dummy domains" permits the use of example.com, example.net, or example.org, as well as the .example dummy TLD. *Dan T.* 00:47, 16 January 2007 (UTC)

Thanks, you've convinced me. Please see my recent edit to this article. -- Jeff G. 23:41, 16 January 2007 (UTC)

I don't think there's any need or point to obfuscating non-link references to example.com as you did, when there's actually a need to specifically refer to it. But I do agree that places where you're just seeking an arbitrary name for example purposes can be done perfectly well with the .org and .net versions. *Dan T.* 00:40, 17 January 2007 (UTC)

I was using Ahoerstemeier's original section title as a guide for munging, rather than his List linked above (as andy). If he is comfortable with unmunging the dots, he is welcome to do that and I would not object - I am just not familiar enough with the contents of the editing experiments he's been finding to make that judgment call at this time. -- Jeff G. 00:51, 17 January 2007

This section is referenced by User talk:Shadow1#Shadowbot biting newcomers.3F. -- Jeff G. (talk|contribs|links|watch|logs) 06:20, 21 January 2007 (UTC)

.bzh - Proposed for deletion

Hi,

I am proposing the article .bzh for deletion. Please see Talk:.bzh for explanation. Your input is welcome. --Amir E. Aharoni 07:23, 20 October 2007 (UTC)

Dash or Minus Sign or Hyphen?

After looking up RFC 952, we learn that it is in fact a "minus sign", and after looking up hyphen we learn that they are not one of the same, and apparently it's not a dash either. So which is it? --Hm2k (talk) 21:00, 4 February 2008 (UTC)

US-ASCII has only one character (hex 2D) that is used for the minus-sign, dash, and hyphen, so they are all interchangable. RFC 952 mentions both the "minus sign" in the text and "hyphen" in the ABNF. In rfc 1345, which describes the character sets used on the internet, the name of 0x2D is HYPHEN-MINUS. Wrs1864 (talk) 21:30, 4 February 2008 (UTC)

References?

This entry has footnotes but they are not listed in the Reference section at the bottom of the entry. Can this be corrected?Nwjerseyliz (talk) 17:00, 14 February 2008 (UTC)

Clean-up of Resale of Domain Names

I did a little clarifying of the appraisal section, don't think I changed anything important. I did not know what was meant by the part: (or any sum resp. division etc.) so I left it. HorusFlight (talk) 14:09, 16 February 2008 (UTC)

Multiple Meanings Nonsense

Domaine name has only one meaning - it is a name for a resource record, which accidentally could be IP address. It is only application and registration which could be different. Especially wrong sentence about "product". It is just matter of domain name delegation which could be monetized. "Names used for other purposes" phraze also wrong - email's domain part associates with some resource records (MX) which should be interpreted by application. --67.207.96.242 (talk) 18:58, 28 February 2008 (UTC)

Wikipedia is a general encyclopedia, not a technical reference. While I can see your point, in the general population "domain name" does have several meanings, some of which contradict what is spelled out in the RFCs. Wrs1864 (talk) 21:14, 2 March 2008 (UTC)

I still don't get what domain means!

Johnnyappleseedrocks (talk) 12:40, 22 March 2008 (UTC)

What is a domain?

The article should start off with a simple AND CLEAR explanation of what a domain is, with an example. It is unclear how much of a URL is a domain. Please re-write for non-insiders. —Preceding unsigned comment added by 211.225.37.48 (talk) 10:08, 29 May 2008 (UTC)

Unicode Domain Names

Why are they not implemented yet? Are there sources/information concerning it to be included in the article? It's a disgrace that in my local country, presenters on tv say "visit something.xx" when their language is completely different. --Leladax (talk) 16:47, 11 June 2008 (UTC)

Internationalized domain name is mentioned right in the intro. What more do you think needs to be there? Wrs1864 (talk)

Free Domain Names

If there is such thing I beleive it should be included. Melab-1 (talk) 22:37, 6 July 2008 (UTC)

Difficult to understand

Will Somebody Please rewrite the "domain name" article so it can be understood by someone who isn't already an expert? Is there a simpler explanation? This has got to be the most confusing article I've ever seen. The first paragraph is just horribly nebulous and complicated. i.e., Assume the reader has an IQ of 100, not 190 or 60. Perhaps give some examples.

Pardon the boldface. I'm obviously not skilled in this code either, it's because I've never posted anything before and just don't know how (not bad ettiquette). I'm just a regular guy who was trying to find out what a "Domain Name" is and was horribly confused by this article. I hope someone who knows, and can explain it well will fix it. Olseshaa (talk) 20:53, 17 July 2008 (UTC)

I agree, this article is extremely difficult to understand. I am a 19 year old college student and I consider myself fairly computer and web saavy, but when trying to learn more about domain names after reading this article, I still have almost no idea of how they work/are created/registered/controlled, etc. Encyclopedias are to used to convey a basic knowledge of a subject, quite unlike this article. —Preceding unsigned comment added by 204.248.24.165 (talk) 17:07, 23 July 2008 (UTC)

Un-mangled

Here is an ammended version of the awful rambling over-blown summary that someone added to this article, thus burying it. If someone wants to rummage through it and re-insert some of it please be mindful of the stated scope of the article:

In computer networking, a domain name is a name given to a collection of network devices that belong to a domain, an administrative space managed according to some common characteristics of the members. In particular, the term domain name is best known in connection with the Internet where it describes the regions of administrative authority within the Domain Name System.

Internet domain names are used in a variety of contexts for identification, reference, and access to Internet resources. They can appear as components of a Web site's Uniform Resource Locator (URL) Web-address (e.g. www.wikipedia.org) or in an e-mail addresses after the customary '@' separator (e.g. example@wikipedia.org), or an X.500 directory such as LDAP or Active Directory.

Domain names are created out of a naming space and methodology that was first defined by Paul Mockapetris in IETF publication RFC 882 and RFC 883 (1983) and used in the first expansion of the ARPANET, a predecessor of today's Internet. The model prescribes a tree-like structure of named nodes starting from an unnamed root node (cf. DNS root zone) that was only designated by a full stop (period, dot, "."). The complete domain name of each node is the string of names of nodes leading to the root node, each separated by a dot. The sequence is written from left to right with increasing order of scope, e.g., node-d.node-c.node-b.node-a. When the full name path of a node is specified, the domain name is said to be fully-qualified (cf. Fully qualified domain name). This condition is often, particularly in the technical aspects of DNS), indicated explicitly by appending a dot at the end of the name (to indicate the root domain).

The DNS methodology confers a unique name to every resource or service participating in the domain name system. This name is referred to as the domain name of a device or Internet host. However, not all nodes in the tree system denote a specific device, rather they are parent labels of an entire collection of subordinate nodes. Such nodes are the domains of the Internet. They represent the spaces of autonomy that are delegated by a group of service providers, called domain name registrars.

These registrars are accredited by the Internet Corporation for Assigned Names and Numbers (ICANN), the organization charged with overseeing the name and number systems of the Internet. In addition to IANA, each top-level domain (TLD) is maintained and serviced technically by a sponsoring organization, the TLD Registry. The registry is responsible for maintaining the database of names registered within the TLDs they administer. The registry receives registration information from each domain name registrar authorized to assign names in the corresponding TLD and publishes the information using a special service, the whois protocol.

In this context a domain name is sometimes referred to as a 'product' sold by domain name registrars. However, the rules of assignment specify that no legal ownership is conferred with such transactions, only the right of exclusive use and the authority to the name space. Once assigned, a domain name becomes part of the pool of registered domain names and is no longer available for use by anyone else. Colloquially, marketers incorrectly refer to domain names as "web addresses", however, a web address is actually a fully specified World-Wide Web resource locator, such as http://www.example.org, actually pointing to a web site.

New domain names are usually registered through the registrar for annual terms with a minimum of one year. The maximum length of prepaid registration is often 10 years, but varies depending on the policies of the sponsoring registry of the top-level domain under which registration is sought. Registration periods may be extended, usually at any time, until the end of a grace period after the registration expiration date.

Domain names may be transferred between parties or advertised for 'sale'. This is often called the "domain name aftermarket" (see below). After a domain name registration and the grace period expire, the domain name is either returned to the pool of available names, or receives special treatment by the registrars and could possibly end up in the 'aftermarket'.

—Preceding unsigned comment added by Anon lynx (talk • contribs) 2008-10-17T21:00:24

Clarifying the 'www' issue:

Point which needs correcting is that the 'www' prefix is NOT part of the registered domain-name. It is in fact the hostname of the site's webserver.

The reason for having the 'www' prefix is historical, in the early days of the Internet domains served multiple purposes -for example ftp, gopher, http- with equal priority. The Web (http) aspect has come to dominate the scene to the extent that it is now customary to treat any unqualified DNS request as a Web request, and route it to the webserver's IP address.

For a better understanding of this, refer to how DNS A-records work. --Anteaus (talk) 20:37, 30 May 2009 (UTC)

Private TLDs

About this line removed: "When Internet domains are liberalized next year, businesses will be able to use their own name in place of .com or .net, for example. The change would allow a company like "Wigets" to have a domain such as services.wigets or bargins.wigets.(ref)"

This is important and no longer a WP:CRYSTAL issue. The event is very notabe and almost certain to take place. Like it or not, this is "almost certain" and people should be aware of it. Even if it wasn't almost certain, a discussion about the "potential change" should be in the article. If it appears elsewhere, it should also have been here because, pro or con, this is important. Whoever removed the above could have, at least, taken the time to reword it or discussed citing a credible source expressing that it is not almost certain. 172.129.45.211 (talk) 21:58, 13 June 2009 (UTC)

It's a proposal, not an event. Something that is 'almost certain' to happen is not an event. It would be ok to discuss important proposals in proper language to document historical developments at ICANN, but WP is not a medium to follow the market places tick by tick. Let it happen and then cite the sources for the facts. Liberalization of domain names? what's that? Seems sensationalism at best. In reality, the obstacles to even 'liberal' availability appear substantial, not the least is the cost of around US $200,000. Only a small minority of companies today is spending that to buy their domain name, so the given example of what companies might do is highly naive. Kbrose (talk) 02:42, 14 June 2009 (UTC)

Who is being naive? Are you going to keep deleting any mention of this? Are you going to delete or rewrite the other article?: Unrestricted generic top-level domain. If this is not a "private domain name" then what is? It is not "just" a propsal. If you didn't like the word "liberalized" then you should have changed it. Including a sentence or two here and a link to the unrestricted TLD article is very appropriated. 172.165.37.62 (talk) 18:55, 14 June 2009 (UTC) Notice: don't mention private domain name or unrestricted generic top-level domain it will be deleted by above user. 172.129.118.252 (talk) 03:04, 26 June 2009 (UTC)

I removed the paragraph again, for several reasons. First, the attached sources all parrot the ICANN release, so there's really just one unique source here. Second, it clearly states that it "approved a recommendation", not an implementation. Third, there doesn't appear to have been any activity since this approval. Fourth, the release states that these gTLDs would be available by Q2 2009, which will end in several days, yet there's no mention of such domains being available. Right now, it appears to be nothing more than an abandoned project, which doesn't merit inclusion here. It isn't "almost certain" as you claim. Mindmatrix 17:01, 26 June 2009 (UTC)

I agree that the removal is warranted on these grounds. In addition the text proposed a usage pattern in non-encyclopedic language that is nothing more than speculation, referred to previously. The ICANN process in itself is notable, however, and deserves mention. It's not clear if it really warrants inclusion in this general article about domain names, which already contains too much controversial 'OR' about domaining. However, I rewrote the TLD section to include this. Any more detail should be left to the TLD article, which had this process mentioned for some time. Kbrose (talk) 18:07, 26 June 2009 (UTC)

First of all, the purpose of wikipedia is for people to learn and find information about something of interest without first knowing some specific jargon of the subject. I was interested in private domain names and could not find any hint in the domain names article. When I finally found something outside of wikipedia, I came back to add it. In your deletions, you are defeating wikipedia's stated goals of allowing anyone to contribute. You have found reasons acceptable to yourselves, but put yourself in the position of someone trying to find info about private domain names.

Wikipedia's filled with thousands of articles that could be deleted by your standards of inclusion. Put back some mention of private domain names that would at least lead someone to another article about this subject. If you just keep deleting it to prove your opinion of "non-encyclopedic language" or other pejoratives, you could probably delete 80% of wikipedia's content. Either you must have some other agenda or you want to be the gatekeeper of content in your "turf" area.

Where is the rewrite of the TLD section that hints of private domain names. In your obvious expert qualifications on what should and should not be mentioned in wikipedia about domains, you could be commended for providing a public disservice. Put something in at least about the other articles that I or anyone else would be able to find without knowing the jargon. You can't expect general audiences searching for "private" domain name to have to search for "unrestricted generic". But you know that, so put back and rewrite the section you deleted so someone can use wikipedia as it is meant to be used !!!! 172.165.17.156 (talk) 22:24, 26 June 2009 (UTC)

Merging Fictitious domain name

The article is small enough to fit within this one, and I don't see it expanding to a point where it'll get over 5KB in size. TTN (talk) 20:00, 26 September 2009 (UTC)

This Article is a Non-Encyclopedic, Unsourced Catastrophe

... of opinion, personal notions and tangential ravings that are completely inappropriate. I'm actually shocked to see a topic with this much relevance in such a shabby state. Apparently, the "Editorial Culture" of this entry is in sore need of some guidance as far as what does and doesn't constitute appropriate entries into a Wikipedia article. Referring it to admins for review, personally volunteering to help clean it up and raise it to standard. Seeking opinions or thoughts before slashing and burning anything. --LoverOfArt (talk) 10:17, 24 November 2009 (UTC)

yes, it would appear that characterization is somewhat less subjective than most of the article. The article has long been taken over by 'domainers' once arguing that more technical and encyclopedic content belongs in Domain Name System, if I recall a comment made a long time ago (perhaps I don't). Kbrose (talk) 19:04, 24 November 2009 (UTC)

A discussion of abuse is certainly encyclopedic IMO. Given the dependence of phishing on domain name abuse alone, which affects everyone online, it seems clear. See my edit summary and inline comment. Perhaps a See also: DNS belongs at the top of the article.--Elvey (talk) 06:42, 13 December 2009 (UTC)

Kbrose reverted me, writing: "the point raised was that the section is complete original research, unsourced opinion, and not a dismissal of validity of subject matter"

Duh. I'd already responded:The linked articles provide/are reliables sources for this section. RS does not require a specific (i.e. footnote) source format. --Elvey (talk) 00:54, 14 December 2009 (UTC)

Wikipedia articles cannot be used as references for other Wikipedia articles; we are not a "reliable source", per our own definition of same. If the linked articles actually contain solid references for the material you want to add, just add them here and be done with it. --Ckatz^chat_spy 06:35, 14 December 2009 (UTC)

Elvy, search google book and google news archives. This is a good start:

Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL

Also search this: "domain name" abuse. with the quotes only around "domain name". Ikip (talk) 23:16, 15 December 2009 (UTC)

Abuses section

User:Elvey was concerned about the inclusion of this section:

Abuses

==Abuses== As domain names became interesting to marketers because of their advertising and marketing potential, rather than just being used to label Internet resources in a technical fashion, they began to be used in manners that in many cases did not reflect the intended purpose of the label of their top-level domain. As originally planned, the structure of domain names followed a hierarchy in which the TLD indicated the type of organization (commercial, governmental, etc.), and addresses would be nested down to third, fourth, or further levels to express complex structures, where, for instance, branches, departments and subsidiaries of a parent organization would have addresses in subdomains of the parent domain. Also, hostnames were originally intended to correspond to actual physical machines on the network, generally with only one name per machine. As the World Wide Web became popular, site operators frequently wished to have memorable addresses, regardless of whether they fit properly into the structure; thus, because the .com domain was the most popular and therefore most prestigious, even noncommercial sites began to obtain domains directly within that gTLD, and many sites desired second-level domain names in .com, even if they were already part of a larger entity where a subdomain would have been logical (e.g., abcnews.com instead of news.abc.com). Shorter, and therefore more memorable, domain names are thought to have more appeal. As a convenience, methods were implemented to reduce the amount of typing required when entering a web site address into the location field of a web browser. A website found at ''http://www.example.org'' will often be advertised without the http://, since the HTTP protocol is implicitly assumed when referring to web sites. In many cases, web sites can be also be reached by omitting the www prefix, as in this given example. This feature is usually implemented in DNS by the website administrator. In the case of a .com, the website can sometimes be reached by just entering example (depending on browser versions and configuration settings, which vary in how they interpret incomplete addresses). The popularity of domain names also led to uses which were regarded as abusive by established companies with trademark rights; this has become known as cybersquatting, in which a person registers a domain name that resembles a trademark in order to profit from visitors looking for that address. To combat this, various laws and policies were enacted to allow abusive registrations to be forcibly transferred, but these were sometimes themselves abused by overzealous companies committing reverse domain hijacking against domain users who had legitimate grounds to hold their names. Such legitimate uses could include the use of generic words that are contained within a trademark, but used in a particular context within the trademark, or their use in the context of fan or protest sites with free speech rights of their own. Laws that specifically address domain name conflicts include the Anticybersquatting Consumer Protection Act in the United States and the Trademarks Act of 1999 in India. Alternatively, domain registrants are bound by contract under the UDRP to comply with mandatory arbitration proceedings should someone challenge their ownership of a domain name. Often email phishing scams will abuse subdomain names to appear to be a legitimate site. For instance, an email might purport to be from Bank of America, and include a link to a fake login screen hosted on http://www.bankofamerica.com.abc.def.ghi.jkl In this case, the actual domain is ghi.jkl, but appears at first glance to be bankofamerica.com.

Should it be included? Ikip (talk) 23:14, 15 December 2009 (UTC)

Limits

Perhaps someone who knows could add a section about length limits of domanin names. I vaguely remember something about 255 characters, but are there for the complete name or per part (2nd level, 3rd level)? --Klaws (talk) 12:40, 24 January 2010 (UTC)

This information is in the Domain Name System article, but I have copied some material from there into this article since it clearly is relevant here and needed as you pointed out. Kbrose (talk) 18:48, 24 January 2010 (UTC)

There's something not quite right about the size limits. If there can be up to 128 labels, with each label having a minimum of 1 character, then there can be up to 255 characters (128 one-character labels + 127 separator dots). And yet the maximum size of a domain name is 253? —Preceding unsigned comment added by 90.216.15.140 (talk) 21:52, 4 February 2010 (UTC)

The article states that there are only 127 levels in the DNS. Kbrose (talk) 22:33, 4 February 2010 (UTC)

127 subdivisions. And the wording seemed to suggest that it's 127 subdivisions relative to the TLD. Making 128. However, if the 127 subdivisions is relative to the root domain, which is an empty label, then the numbers work out fine. —Preceding unsigned comment added by 90.210.92.152 (talk) 11:49, 5 February 2010 (UTC)

Can someone source why can't non-for-profit organizations like Wikimedia can't regulate domain names and not have people pay $35 for a single entry in a database?

. —Preceding unsigned comment added by 195.74.249.114 (talk) 17:20, 5 March 2010 (UTC)

domain names seized by government

seziure of domain names is important as it speaks to their de -facto treatment under the laws of the united states, specifically that they are just another form of property that can be forfeited under certain regulations based on court order.

but besides all that. there are hundreds of websites and dozens of mainstream news stories, about this nov 2010 seizure, so i believe it is notable. and it needs to be at least mentioned with minimal language, that it happened. all the legal ramifications can perhaps be dsicussed later when mainstream news/legal sites do more legal analysis. but a basic mention i think is in order. Decora (talk) 01:52, 28 November 2010 (UTC)

Business.com resold for $350 million in July 2007?

why was "Business.com resold for $350 million in July 2007" removed? 213.8.90.74 (talk) 18:08, 16 January 2011 (UTC)