Questions and Answers
Q1 - Isn't ICANN (the Internet Corporation for Assigned Names and
Numbers) responsible for domain names?
A1 - The U.S. Department of Commerce discussion
policy documents from 1998 present ICANN's background. Read
them and you realize that ICANN was conceived primarily to solve Department of
Commerce problems, not Internet domain name problems.
Since ICANN was created to implement and administer the preconceived notion
that additional generic top level domains (gTLDs) should be made available,
those who subsequently identified themselves as ICANN's 'multistakeholders' were
groups and individuals with a stake (financial interest) in adding new gTLDs or
protecting their existing interests when new gTLDs were introduced.
In ICANN-speak, 'innovation' and 'competition' have always meant creating new
combinations of letters for TLDs, and registries competing to run these
domains. User choice was initially considered important, but when users chose
one legacy TLD (see Question 2 below) ICANN replaced that objective with
Oh, by the way, did we mention that ICANN received 1930 applications, at a price
of $183,000 each, to open new generic TLDs? That amounts to income for ICANN of
over $350 million for applications to sell something users don't want or need.
The proposal we've outlined is innovative and allows competition between
registries, but we think it meets market needs better than continuing to add more
Q2 - More than one hundred thirty million .com names are
registered. Obviously it hasn't been hard to create new names.
What's the problem?
A2 - The problem is just that: there are over 133
.com names. Each name must be unique while in
reality most names, and most trademarks, are not unique.
There are two separate US trademark owners of the name Google; the oldest US
trademark on the name 'google' was registered in 1939. That trademark owner does
not own the corresponding .com domain name.
Beside the lack of equitable access to names under the preferred or default
TLD (either .com or a country code TLD), many domain names are held for
speculation, which keeps legitimate users' content and commerce off the web.
The legacy system allows one person to 'own' a generic term. Example:
worldwide there can be only one plumber.com on
As early as 2004, well over a decade ago, only 3.7% of corporations around the
world had identical corporate and dotcom domain names.
Imagine the telephone system following Internet rules which allowed anyone
in the world to register your name and thereby prevent you from getting
phone service under your own name.
Or just try to register a domain name that is meaningful, short, and easy to
ICANN's new batch of top level domains doesn't alleviate the problems since
names under many of the new generic TLDs are reserved (right of first refusal)
for owners of the same name under existing TLDs. An example is
the new .uk TLD which grants the owners of existing .co.uk names a full 5 years
to register the same second level name under the 'new' TLD.
That doesn't increase the name space, that sounds like a protection racket.
Q3 - Does this require a new naming system?
A3 - No, we're suggesting an evolution of the existing
system, the fundamentals aren't changed at all.
Here's how it would work:
Domain names are registered in a set format, then
a little new technology is used to introduce a
keyboard character that hasn't been available in domain names
previously. This character is restricted for use as an
'addressing token' in much the same way the @-character is used
in e-mail addresses. This character plus a number allows you to register
names that are the same as - but at the same time different from
- existing domain names. This simple device would bring the Internet into
better alignment with the real world, where different people and
companies often share 'the same' name.
If we look at the alternatives:
We know that 'all the good names are gone' and have been for
years. If you're creating a new venture you may be able to
register a short, catchy domain name and name your company after your
domain, but if your company already enjoys name recognition and
goodwill, even having a registered trademark won't help in most
ICANN received over 1000 applications to register new generic Top Level Domains
- at a cost of $185,000 per application. This may be good for ICANN, but
'fragment and confuse' a not a good policy for Internet users! New top levels
have been introduced before, but have never been very successful.
The most successful 'legacy' new generic TLD, .info, was introduced in 2001 and
has decreased from over 8.3 million to just 6 million registered names,
compared with .com's 129 million.
Another gTLD, .biz, was opened in June of 2002 to complement (or compete with)
.com. It has fewer than 2.2 million registered names.
There are other expansion TLDs such as .aero (June, 2002), .coop (June, 2002)
and .pro (June, 2004). Do you even know they exist?
Users tend to treat new top level domains with suspicion and businesses rightly
view them as second rate addresses.
Wait list (back-order) systems are available for ordering names that
expire, but how many people do you think are in line ahead of you to
buy the 'good names'? How long can your business wait for a name,
and do you want one that the previous owner may have run into
There is a secondary market in domain names, fueled by
speculation. If it normally costs only a few dollars to register
a name for a year, are you willing to pay thousands, tens of
thousands or hundreds of thousands of dollars for the same name?
Someone is getting rich on scarcity, but the scarcity is artificial and it
promotes neither communication nor commerce. In fact, it impedes both.
Q4 - Don't search engines make this unnecessary?
A4 - If search engines were perfect we wouldn't need domain
names at all, but domain disputes continue year after year - business
owners want control over 'their own' names.
A normal user only reads the first 2 pages of search engine results. This
means that as web content grows, positioning your business becomes harder
instead of easier.
Q5 - Why not just use numbers – why add a new
character like the asterisk to domain names? Won't that confuse
people and increase the risk of name abuse?
A5 - Good reasons for introducing an addressing token
include security and avoiding possible collisions with existing
names. Another is that company, product and trademark names may
already end in a number. Consider the case of a television
station that wants to use channel5.com
as a domain name. There are lots of channel 5s in the world and
the second registrant can't be expected to register as channel51.com
which changes the meaning and identity, and besides, a UHF channel 51 may
already own that domain name.
For novice users, the asterisk is just another character on their
keyboard. If the e-mail @-character isn't confusing, the
asterisk shouldn't be. One new character with a defined
function is easier to assimilate than several (or many) new top level domains.
An asterisk that means several instances of the same name are
registered is a positive security development for domain
The need to create new and original words to register as domain
names generates uncertainty about name ownership and supports the
abuse called typo-squatting. Registering a name that is not your
recognized name (because that one is already taken) generates
confusion, uncertainty and frustration for the consumer.
Opening an unlimited supply of familiar words and
names should help, and if name-based directories are created,
additional information could be made available to users even before
they type in or click on a domain name.
Q6 - Aren't you asking everyone, everywhere, to install a new
A6 - No, that wouldn't be user friendly.
Typing the [name]*[number] format into your address line requires a new browser
function, but we created an Asterisk prototype to show how the
function could be added to any web browser. This could be done as an
add-on or plug-in, or when the browser is updated.
New web browsers are delivered with software to translate Internationalized
Domain Names, even if you don't realize it or ever use it, so similar
development may be seen for multiplexed names.
Or you can always type the mlx--[name]--[number] native registration format
into your address line. It isn't elegant, but it is transparent and it works.
A site registered under a multiplexed name would be on line as quickly as any
other site. A directory page or database with a link to that site could be
updated immediately. Search-bots have to find and index the site before it
can be displayed among search results, which may take several weeks. Linking
the new site to a frequently updated directory page would speed up the process.
As soon as the site is listed in a directory or found by your favorite search
engine, it can be selected like any other site, by anyone, with any browser, by
clicking on a link.
Q7 - Is the name registration format formally
A7 - The name registration system
is fully compliant with existing standards, but adds a new level. That new level
isn't standardized. If a standard is established it may differ
from the one we have tested.
Q8 - Who wants to be a 'number 2'?
A8 - To start with, the second iteration of a name would be
name*1.TLD. That means we could hypothetically add over 100 million .com
domain names or double the number of names under a country code TLD without
making anyone a 'number 2'
But ask yourself: if your name is Jones and jones*21.com is already
registered, would you hesitate to register as jones*22.com? There may be some
perceived advantage to having a lower number, just because it is
Jones registers as a jones.TLD and the rest is 'disambiguation'.
The asterisk and number aren't a value judgment any more than your telephone
The proposed naming evolution returns domain names to the status
they once had as simple addresses rather than property.
The site you are now reading is the first example
of this naming convention in use to provide information.
Q9 - Isn't there a 'chicken and egg' problem here?
Without a lot of names registered in a standardized format there's
no reason to update a browser to translate them, and
without general browser implementation there's no
advantage in registering a name.
A9 - There was no reason to have a web browser
until there was information available on the web, and no rationale for putting
information on the web until people had access to browsers. Yet now we have both.
When people realized that better information access was possible, these
developments became mutually supportive, which is what you might expect when people
discover that Internet naming can be improved.
Since multiplexed domain names can be kept quite short you can
always use the native name registration format – the name*number format
is a convenience, just as domain names are more convenient than typing an IP
- Won't trademark owners object to losing their monopoly on a domain
A10 - Trademark owners are expected, even required,
to vigorously defend their marks. This applies equally to those companies that
can use their trademark as a domain name, and the much greater number of
companies that cannot.
It's understandable that companies with a registered trademark think
they have a greater right to it use it as a domain name than someone else, but
cybersquatting is based on the uniqueness of domain names,
not the uniqueness of trademarks. Trademarks generally aren't unique.
A trademark owner who also owns the corresponding domain name may
complain if he loses his artificial monopoly, but remember that each domain name
based on the same name string is still unique and as different as Smith and
Smyth. A name-based directory would eliminate confusion about name ownership -
which is the objective of a trademark! Since most trademarks aren't unique
we might even see class action opposition to the artificial and unnecessary name
string monopoly in domain names.
Instead of focusing on contention, consider how additional names
can benefit major trademark owners. Imagine the largest suppliers of goods and
services - think of companies like Wal-Mart, Pizza Hut and McDonald's. These
three companies have thousands or tens of thousands of physical locations, but
they're restricted to three domain names. If they recognize the
need for each location to have a separate telephone number, then why not separate
domain names utilizing the familiar registered trademark?
Q11 - Besides making more names available, are there
other advantages to evolving the naming system?
A11 - You can't hold something for ransom unless it
is unique. If domain names aren't unique you reduce the
rationale for name hoarding, cybersquatting, registration hijacking
and inflated prices in the secondary market.
Registered under an appropriate End User License Agreement, multiplexed
names would solve multiple problems and combat multiple abuses.
Q12 - Shouldn't we look for a completely new system
A12 - Since it has been difficult to gain acceptance for new
top level domains, how can we expect more radical changes to
succeed? You have to respect the experience and expectations of
a world full of Internet users. What would you do with existing
Internet content if a different naming system were introduced?
Name*number identities are optional, not mandatory, and coexist
with legacy names. They comply with all existing DNS standards.
They 'do no damage' since they're backward compatible.
Q13 - I'm an 'informavore' - I use the Internet for
collecting Information but haven't felt the need to register my own domain name and
possibly never will. How do more domain names rock my
A13 - Web usability studies have shown that given a known
company or product name, many people will attempt to guess and type in
the 'right' domain name several times before they turn to a search
Multiplexed names support the creation of name-based directories that
provide another alternative for finding information efficiently,
but the major advantage is that by eliminating the artificial scarcity
of domain names, more information should become available from more sources.
Q14 - Are multiplexed names built on the same
software as Internationalized Domain Names?
A14 - No, IDN software isn't involved. Both are browser
applications that can co-exist.
Modern browsers can
read normal domain names and IDN names with characters such as
the Swedish Å, Ä and Ö. The browser for
multiplexed names adds the alternative of reading names
containing the asterisk as an addressing token. That addition is
transparent and does not interfere with legacy or IDN names.
IDNs can not use the asterisk for creating multiplexed names -
they can't use the asterisk at all - but a different
character such as the Euro symbol (€) could serve the same
function as an addressing token for European Union domain names. If
you want character universality (the asterisk always meaning the
same thing in names, regardless of language) additional browser
software could be written to allow multiplexed IDNs.
Q15 - What business aspects apply to
A15 - In mid-September, 2007, a local Internet
registrar listed 61 names for sale on the secondary market.
Asking prices ranged from $1,500 to $1.8 million, with an average
price of nearly $245,000.
By early October, 2009, auctions for secondary market names had become
established. One major reseller offered 10,000 names with bids starting
at $2,000. The first 2 pages of names listed by another major reseller had
'buy now' prices averaging $6,800.
In late November, 2010, the local registrar named above listed 58 domains at
an average price of nearly $280 thousand, with two listed at $1.8 million.
On July 1, 2013, a well known reseller listed 473 'premium' .com domain names
for sale. The average price was $42,468.
During the 2nd quarter of 2015 the average aftermarket sale price for the top
10 .com domain names was $192,038.
If you could find an appropriate name to buy for $2,000 on the
secondary market (plus customary registration/renewal fees) and compare it
with a multiplexed domain name hypothetically costing $50/year (plus the
customary fees), it would take 40 years to realize pay-back on the
secondary market name.
If we compare a high-priced name at $192k with a multiplexed name costing
$50/year, pay-back would take over 3,800 years.
Multiplexed domain names are covered by US Patent 8,543,732,
parent applications and a grandparent Patent number
Last updated March 10, 2018