You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 21 Next »

The Internet is growing.  From the start of the Internet through 2000 the number of top-level domains was small and the format of domain names was simple. All domains ended with one of a number of common endings like ".com", ".net" and "org.", or a two-letter country code like ".de" or ".uk". Times have changed. Since 2001, new types of domains have been, and are being, introduced:

  • ASCII TLDs of more than 3 letters long (think of ".info" or ".museum"); and
  • since 2010, TLDs comprised of non-Latin characters (such as".中国" and".рф").

While these changes significantly broke the pre-2000 rules of what looked like a legitimate top-level domain the changes happened infrequently enough so that eventually the word got out.  With the ICANN Board's approval of the new gTLD program in 2011 the door has opened for more frequent additions of legitimate top-level domains. From just October 2013 to February 2014 over 150 new top-level domains have been added.  That is just the start, the additions are continuing.

Over time software running Internet services, such as the web, email and many others, has greatly improved.  This sounds like a good thing, but is sometimes backfires in the effort to expand the domain names in use.  With more stable software, updates are farther and farther apart in time.  Software developers concerned with speed, safety and user friendliness have made choices that restrict new names being used through, for example, limiting the top-level domains a user can specify via drop-down box or including a static list of what is considered to be a valid top-level domain.  Given these checks "burned into the code" and the less frequent updates, new names face acceptance problems.

The cooperation of software vendors, open source tool developers, internet services providers, web site developers and others is required so that these new top-level domains are available to all that would use them.  There are three concerns to address:

  • Removing "false positives" in filters preventing the use of new top-level domains
  • The ability to render (see) all names in the native script, not having to resort to using the underlying encoding
  • And, achieve the intended level of safety and convenience in a rapidly changing Domain Name System
Removing "False Positives"

In many cases, there is a sort of "check" imposed whenever you type a name into your browser. The purpose of the check is to screen invalid domain names before a DNS query is sent. Some of these checks still do not allow for all domains such as the newer ones that are four or more characters or are in languages other than those using Latin characters.  Some of the checks consult a static list included with the software version in use.

Those constraints imposed in software on what is allowed as a valid domain name (such as limiting domains to specific endings like ".com") are artificially constraining the growth and utility of the Internet. The TLD Universal Acceptance effort aims to ensure those constraints are lifted:

  • first, by recommending against software performing checks on domain name validity unless it is truly required, and
  • second, that if checks need to be performed, it is done in a way that allows for all domains to function correctly, whether they were registered 20 years ago or yesterday.

Most software should not need to test whether a domain name is valid. The act of connecting to a domain name (in the case of a URL, for example), or sending an email (in the case of an email address) will automatically provide mechanisms of informing if the domain name is valid or not.

If there is a need to check domain validity, it is most desirable that the DNS protocol is used for that purpose. If, however, there is a specific requirement to only check the validity of the top-level domain, a current list of all top-level domains that have been delegated within the authoritative root-server system is available here. This list is maintained by ICANN's IANA function and provides a regularly-updated list of top-level domains that currently exist in the root zone.

Rejection of some top-level domain strings due to outdated length parameters or other erroneous formatting criteria can be avoided by reliance on authoritative information. As described in Support of New Top-Level Domains by Internet Infrastructure Operators and Application Providers (2003), and Evaluation of New gTLDs (2004), several technical acceptance issues were associated with the gTLDs introduced in 2000-2001.

Cooperation among registry operators, Internet service providers, software developers, vendors, and others who deal with domain names on a regular basis is critical to ensuring the continued realization of the Internet's potential for commerce and communications. In October of 2004, ICANN opened a discussion forum on top-level domain acceptance issues to foster cooperation among the necessary parties. The archive of this discussion forum is available here.

One recent hypothesis that may influence how the removal of "false positives" is undertaken is that besides "validity" the issue is determining the "reputation" of what a domain name represents.  Domain names are ingredients in URLs and in email addresses (which themselves can be written as a URL) which represent objects and people.  One place to examine and/or establish current practices in reputation determination.

Another recent hypothesis is that there are natural language analysis tools that try to turn URLS into "clickable links."   These tools take otherwise plain script and try to make the links stand out.  These tools need to be smarter in the sense that they recognize all legitimate URLs and not just those that end in a fixed set of TLDs.

Rendering Only In Native Script

This is newer goal of TLD Universal Acceptance, one beginning with the first IDN top-level domains.  There has been some reference to this in the existing slide decks and consultations but the critical mass of disappointment has yet to be reached to see much progress on this.  The critical mass is building though, there will soon be more on this topic.  More information on this will be forthcoming.

Rapid Emergence of Top-level Domains

This is a more recent goal of TLD Universal Acceptance, emerging with the first of the new top-level domains added in October, 2013.  As the project progresses, more information will be forthcoming.

Where We Need to Be

The project can be considered a completed success when anyone can use a domain name from any TLD in any service.  That's a wide ranging goal and written as such, may open the door to this project never claiming complete success.  So, to use a less complete but sufficient criteria, TLD Universal Acceptance will be considered complete when any person can register and use a domain name in any top-level domain in widely distributed web browsers, email clients, in setting up accounts for Internet services and other services (when desired by the service provider).  In short, in any exchange in which the client and the service provider wish to accept any validly delegated domain name, they are able to do so.

How to Start Contributing or Ask Questions

To submit questions or contribute additional material that may be helpful in further work on this, please email to: tld-acceptance@icann.org

Past Presentations, Public Comment, and other Resources

ICANN 49: TLD Universal Acceptance session archive - 24 March 2014

ICANN Blog Entry - 21 March 2014

W3C Mail thread on TLD Universal Acceptance (with APTLD draft of barriers) - 3 March 2014

TLD Universal Acceptance as part of Asia Pacific Regional Discussions at APRICOT 2014 - 24 February 2014

CircleID Post on TLD operator "use case" - 7 February 2014

Joint ccNSO-gNSO IDN Working Group draft final report announcement - 25 June 2013

Internationalization and Unicode Conference 36 (slides: ICANN's Work On Internationalizing the Domain Name System) - 24 October 2012

Universal Acceptance of All TLDs - ICANN 45 (Toronto) (Session Details) - 17 October 2012

Universal Acceptance of All TLDs – ICANN 44 (Prague) (Meeting Announcement) - 27 June 2012

Presentation to Joint ccNSO-GNSO Internationalized Domain Names Working Group - 12 June 2012

Registry/Registrar Meeting (Los Angeles) -18 May 2012

Universal Acceptance of All TLDs – ICANN 43 (San Jose, Costa Rica) (Session Details) - 14 March 2012

Public Comment: Initial Report on Universal Acceptance of IDN TLDs - 6 January 2012

ICANN Releases Beta-3 Version of TLD Verification Code - 22 March 2007

Github repository (2012)

ICANN has issued today a new version of code intended to assist software developers and application providers whose work assists others in using the DNS.

ICANN Releases Beta TLD Verification Tool - 3 December 2006

Today ICANN released a beta version Top-Level Domain (TLD) Verification Tool. This verification tool has been developed in response to problems reported by gTLD registries and end-users of the non-acceptance of some existing TLDs. These problems occur in some current applications because: 1) they do not recognize any TLD of more than three characters; or, 2) they rely on legacy information where only com/net/org and a handful of ccTLDs are recognized as valid.

ICANN Creates Area on Universal Acceptance of TLDs - 20 March 2006

In order for the full resources of the Internet to be available to all users globally, service and application providers must make use of the complete range of top-level domains (TLDs). To promote accessibility of up-to-date TLD information, ICANN is making available a new page of resources related to this topic.

Universal Acceptance of all gTLD Names - 18 October 2004

Other Resources

Recent space activity

Space contributors

{"mode":"list","scope":"descendants","limit":"5","showLastTime":"true","order":"update","contextEntityId":48341315}

 

  • No labels