ICANN ICANN Email List Archives

[xxx-icm-agreement]


<<< Chronological Index >>>    <<< Thread Index >>>

Re: An Alternative to .XXX: IANA Adult Port Assignments

  • To: xxx-icm-agreement@xxxxxxxxx
  • Subject: Re: An Alternative to .XXX: IANA Adult Port Assignments
  • From: George Kirikos <gkirikos@xxxxxxxxx>
  • Date: Fri, 16 Mar 2007 17:13:34 -0700 (PDT)

Hi again,

As a followup, I just read about

http://www.cp80.org/content/solutions/community-ports

posted elsewhere. Same idea (didn't notice that other one -- derived
independently; I guess great minds think alike, or fools seldom
differ).

Sincerely,

George Kirikos
http://www.kirikos.com/


--- George Kirikos <gkirikos@xxxxxxxxx> wrote:

> An Alternative to .XXX: IANA Adult Port Assignments
> ---------------------------------------------------
> 
> As an alternative to the creation of the .XXX TLD, ICANN/IANA can
> assign special port numbers that can be used to label adult content.
> 
> Preliminaries
> -------------
> 
> IANA assigns port numbers as part of its duties. For example, port 80
> is reserved for the HTTP protocol (i.e. the World Wide Web). Port 443
> is reserved for the HTTPS protocol (SSL-secure version of HTTP). Port
> 23 is for Telnet, port 25 is for SMTP, and so on. One can see the
> full
> list at
> 
> http://www.iana.org/assignments/port-numbers
> 
> One can theoretically run protocols over any port (e.g. you can have
> a
> web server on port 25 or port 18666 -- http://localhost:18666/ could
> access a webserver running locally on port 18666). In a real sense,
> the
> IANA port assignments are just suggestions to the world as to what to
> expect on certain ports, whether it be a mail server, WHOIS, FTP, POP
> email or any other service/protocol.
> 
> .XXX Proposal
> -------------
> Ultimately, the .XXX proposal comes down to the use of a
> top-level-domain (TLD) string as a label mechanism. It creates an
> expectation that if one goes to the domain example.xxx, it will
> probably have adult content.
> 
> The .XXX proposal has been very controversial, as one can see from
> the
> public comments at:
> 
> http://forum.icann.org/lists/xxx-icm-agreement/ .
> 
> One of the major criticisms is the name allocation mechanism. There
> can
> only be one sex.xxx or porn.xxx, for example. Should that domain be
> allocated to the current registrant of sex.com? What about the
> registrant of sex.ca or sex.co.uk or sex.net or sex.de or other
> existing TLDs? Allocation mechanisms such as auctions or RFPs or
> "priority to oldest existing registered domain" and other systems
> have
> been discussed, but none that will make everyone happy.
> 
> In addition, there has been a great concern that trademark holders or
> existing registrants will need to make defensive registrations, in
> order to prevent their comparable domains to be registered in the
> .XXX
> space. For example, IBM might never want to get into the adult
> content
> business, but will likely be compelled to register IBM.XXX.
> Mercedes-Benz owns Mercedes.com, and likely will not want to see
> adult
> content at Mercedes.XXX via a webcam girl whose first name happens to
> be "Mercedes." A non-trademark holder (e.g. an individual John Smith)
> might feel compelled to register JohnSmith.XXX to ensure that someone
> else doesn't register it and use it inappropriately. Some have argued
> that new TLDs are almost guaranteed a profit because of the vast
> number
> of defensive registrations that are made in sunrise periods, usually
> at
> premium prices relative to general registrations, in order to prevent
> cybersquatting.
> 
> There are also concerns that registrants will need to pay $60/year or
> more to a new registry, which is considerably higher than their
> existing domains, and that this represents a "tax" on their business,
> increasing their costs for the benefit of a for-profit registry
> operator.
> 
> Suppose .XXX was rejected. Does an alternative mechanism exist to
> label
> content? There already exist mechanisms such as ICRA labels, for
> instance. They can be used with any TLD, and don't require a new TLD.
> Indeed, the use of .XXX is really a very simple form of a label, in
> that a domain using it is allowing a "binary choice" form of
> filtering,
> either something is in .XXX, or not in .XXX (i.e. "on/off"). Others
> have counterproposed .KIDS, as a white-listed TLD, instead.
> 
> Use of Port Numbers As a Label
> ------------------------------
> Another alternative would be for ICANN/IANA to assign, reserve and
> register port numbers specifically for adult-related content. As an
> example, port 18666 is currently unreserved/unassigned, and can thus
> be
> used as a label to the world to expect adult content to be on that
> port. Ports 18001 through 18180 are also unreserved/unassigned at
> present. [Age 18 is typically the age at which one is considered an
> "adult", thus motivating those particular numeric choices; the "666"
> is
> there for those who recall the April Fool's Joke about the "Evil
> Bit", 
> i.e. RFC 3514, see 
> 
> http://slashdot.org/articles/03/04/01/133217.shtml
> 
> so, one can in a way consider this the "Evil Port", if one has a
> sense
> of humour, although there is no technical reason why any random port
> can handle the job of being the label] Ideally, two ports would be
> reserved, to be able to have counterparts to secure (HTTPS) and
> non-secure (HTTP) protocols.
> 
> Adult companies that wish to label their content could thus do so by
> serving their content on the chosen port (I'll use 18666 in the
> following examples, but it can be anything that ICANN/IANA decides
> upon).
> 
> Instead of having a nude image at http://www.example.com/nude.jpg the
> webmaster could instead "label" it by having the nude image come
> instead from http://www.example.com:18666/nude.jpg . Browsers like
> Internet Explorer, FireFox and Opera could eventually even shorten
> the
> above using "adult" as a replacement for the combination of "http"
> and
> "18666", so that one could use "adult://www.example.com/nude.jpg" as
> a
> URL.
> 
> On a technical level, this is very easy to implement, as Apache and
> other webservers can be configured to serve up content on any port.
> At
> a first approximation, the cost is $0. It is obviously much cheaper
> and
> less disruptive to implement for adult webmasters than registering
> .XXX
> domains.
> 
> Also, the contention over the allocation mechanisms of a .XXX domain
> would disappear under this alternative. Sex.com, sex.ca, sex.net,
> sex.de, sex.co.uk, and all other TLDs can co-exist, all serving up
> their adult content on ports 18666 instead of port 80, if they wanted
> to use a port-based label mechanism. Companies like Playboy,
> Penthouse,
> etc. need not register any new domains, they just would change a
> webserver setting instead if they wished to use this alternative form
> of a content label.
> 
> There would be no need for defensive registrations, as folks could
> continue along with their existing domains. Mercedes-Benz, Gucci,
> Yahoo, Chanel or other brandholders would not need to worry that
> cybersquatters have another playground in which to infringe upon
> their
> trademarks.
> 
> Some supporters of .XXX only support it if governments make it
> mandatory. Their theory is that it would be a lot easier to filter
> adult content if it was all located in the .XXX space. However, if
> the
> government instead made it mandatory that all adult content was
> served
> from port 18666, it could be filtered just as easily (it's a very
> trivial firewall rule to permit/deny access to a single port). ISPs
> or
> parental filtering software could filter a single port just as easily
> as they can filter a single TLD. 
> 
> Some supporters of .XXX want to make a lot of money (i.e. through
> operation of the registry, being a registrar, or speculating in
> domain
> names)! The use of port 18666 would not make these people happy,
> though, as there'd be no pot of gold at the end of the rainbow, to
> implement use of a port label such as port 18666 to identify content.
> However, if it is the goal of ICANN to find efficient and low-cost
> solutions to "problems", the use of ports as a label mechanism is
> offered as an alternative to solve the same problems that .XXX
> purportedly solves.
> 
> Sincerely,
> 
> George Kirikos
> http://www.kirikos.com/
> 



<<< Chronological Index >>>    <<< Thread Index >>>

Privacy Policy | Terms of Service | Cookies Policy