“Brad ideas,” this time, means bad idea.

Come on - you actually think you can get this djinn back in the bottle? You actually think you can get people to migrate away from .com? And until you do so, you actually think it’s fair (not to mention legal) to restrain trade and not let others compete, on the generic level, with .com?

Not even close, Brad. I respect the hell out of you, man, but not even close this time.

As I say, we are nowhere near this, but that doesn’t mean it shouldn’t be spelled out how it should have been done, and how it might be in the future.

And no, two wrongs don’t make a right here.  Because one party got ownership of “com” doesn’t mean another should get ownership of “museum” or “mobile” or whatever to make up for it.

Instead, if a new and better system is to be set up, the monopoly owner of .com/.net, Verisign, should pay a licence fee on it, an immense one, and that fee should then subsidize the new branded TLDs, probably making all domains in them free for as many years as people want to keep pouring money into .com fees.

I reject the suggestion that when you made a bad monopoly that you should either balance it by selling other monopolies or can never fix it by undoing the monopoly.

First, let me be sure I understand your concern. It appears that it’s with the use of a generic word like “apple”, leading to a requirement like apple.m as well as apple.m.  You’d like to replace all registries with commercial competitive ones—not a generic .com or .m with, at least hypothetically, a neutral steward, but make it a completely competitive environment.

If I understand you correctly, I have several problems with this approach.  First, there is a legitimate operational reason to have consistent behavior from any TLD.  We can see from the .com and .net controversy that even a registrar that theoretically has some oversight feels free to change behavior for its commercial interest.

I understand that DNS is being used as a general directory system, which it was not designed for. Nevertheless, until directory services (perhaps de facto through appropriate search engines) becomes available, from the operational standpoint, I want to see consistent DNS protocol behavior from all TLDs.  Overtly commercializing it runs counter to this, as the TLD operators will reasonably try to maximize their revenue with new “features”. 

I subscribe strongly to the belief that the Internet infrastructure should exhibit consistency.  The directory problem is soluble by user choice at the edge—not by imposing commercial modifications in the TLD.

Second, what about noncommercial uses and domains?  Should Canada have the .ca domain taken away, and have to go to a commercial TLD operator for registering the Prime Minister’s office?  In less developed countries, this could turn into an additional expense that inhibits citizen use of the Internet.

Similar concerns apply to educational institutions.  We have a certain consensus that there is a social value to having .edu. There are less clear, but probably still real, to recognize the value of .org and .int, and the possible barrier to Internet entry of nonprofits if all TLDs are commercialized.

Actually I’ve been proposing a .here TLD. A TLD reserved for nonunique local use somewhat like RFC1918 IP addresses. So you can go anywhere and do https://whats.here/ or https://whos.here/ and get a list of stuff available at a particular physical area/context. And possibly ping brad.here, browse and control your airconditioner.here and so on.

That way what is “.here” won’t clash with what’s outside or elsewhere.

For more info: http://www.watersprings.org/pub/id/draft-yeoh-tldhere-01.txt

I think the monopoly owner of .com/.net already paid their huge license fee… :)

http://news.com.com/2100-1023-237656.html?legacy=cnet

Nothing is really new with your suggestion, as registries could already get “.yahoo.com”, “.dunn.com”, “.yellowpages.com”, “.google.com”, “.wipo.com”, and so on, and sell domain names within those. Some companies have even used exactly this idea, like MSN parnering with some copanies and creating “company.msn.com”, or some guy getting “israel.net” and selling “company.israel.net” to Israeli companies.

What’s the difference in your suggestion - just the saving of 4 more characters by ommitting the “.com”? What does this “saving” get us?

I can tell you what removing that .com will get us - a big mess. Instead of having millions of “.com” entries, we’ll suddenly have millions of TLDs. Why be “Cocacola.com” when you can be just “cocacola”? And worse, the new mess will be irreversable. Unlike today, when the root of the DNS is still “clean” and it’s easy to add a new TLD (e.g., for some new country, or when some new concept I can’t even think of now happens on the Internet), after your change there will be no room for future improvment.

Also, how do you suggest the root DNS servers work after this change? Will they have to track thousands (if not millions) of TLDs? Will that mean less redundancy for the root servers, and because of the higher costs of maintaining them, commercialization of the root servers in the verisign style?

I don’t like the idea of TLDs being proprietary rather than generic… like “.yahoo”... if I wanted somebody else’s trademark to be within my e-mail and Web addresses, I’d use addresses within my ISP’s or hosting provider’s domains.  It’s to get away from this that I registered my own domain.

I agree that the .mobile idea is not well thought out and I definitely agree that we need to reinstate the concept of a generic word into the DNS. Clearly this cannot be done wholesale but it should be possible to set up a trial run at this using the .ALT top level domain. Basically, the .ALT registrar would operate all second level domains in .ALT and Internet users would register third level domains similar to demon.co.uk.

Since anyone can register something.word.alt, this means that noone owns word.alt. This makes word.alt into a generic domain. A lot of people are aware of the alt.* hierarchy in USENET where anyone can create a newsgroup. This idea mirrors that. In the alt.* hierarchy, as in the real world, anyone can make a word and anyone can use a word in any particular context. Therefore, the .ALT TLD should, as much as possible, not limit the creation of second level domains in .ALT and not constrain the third level domains that people come up with.

Inevitable this will lead to domains like drink.coke.alt and ihate.coke.alt and piss-on.coke.alt which the Coca-cola company may not like. But just as our world has accepted the concept of common carrier in transport and communications, I believe we can also see that the registry for .ALT is not damaging Coca-Cola’s reputation in any way by registering such names. They will only form a small percent of all .ALT names and a miniscule percent of all web accesses. But they will provide an important avenue for people to let off steam when they need to, and a democratic system needs to have such avenues or it will become brittle and crumble.

By the way, the .ALT idea has been discussed before. Here are some references that you may find interesting.

http://www.gtld-mou.org/gtld-discuss/mail-archive/01100.html

http://www.gtld-mou.org/gtld-discuss/mail-archive/01815.html

http://www.merit.edu/mail.archives/nanog/1996-04/msg00351.html

http://www.cctec.com/maillists/nanog/historical/9604/msg00367.html

You write near the beginning:
“Nobody should be given ownership of generic terms. Nobody should have ownership rights in a generic word like “apple”—not Apple Computer, not Apple Records, not the Washington State Apple Growers, not a man named John Apple.”

And then later on you write:
I maintain that the only way to get a competitive innovative space is to slowly get rid of the generics and allow a competitive space of branded TLDs for resale. .yahoo, .dunn, .yellowpages, .google, .wipo, and a hundred other branded resellers competing on even footing to create value in their brand and win customers with innovative designs, better service, lower prices and all the usual things.

So on the one hand you are saying TLDs are bad because they limit choice…and then later on you are proposing to eliminate TLDs which will for all practical purposes eliminate all choice!

I for one do NOT want .yahoo, .google, and definitely not .dunn in my domain name anywhere…and why should some companies be special and get their own TLD leaving most others relegated to a domain name “backwater” so to speak.

The snipit about them providing better service, etc…don’t count on it…

And since you reference the EFF in your article, I assume you care very much about freedom of speech and expression…

In your corporate TLDs universe, folks owning domains relating to controversal topics such as abortion, drug reform, etc would likely find themselves with a long, convoluted domain name or very possibly with NO domain name at all…

For example, do you really think Google would allow a domain such as googlesucks.google to be registered? Didn’t think so.

And what about DNS technical issues…how would email function, web browsers, etc. Eliminating TLDs at this juncture, even if everyone agreed, would be a nightmare; heck consider all the problems Verisign’s “wildcarding” caused the other month…you are aware of that aren’t you?

In short, eliminating the TLD portion of domain names is a bad idea, will cause serious problems, be more expensive, and all and all limit choice to individuals and companies alike who rely on domain names everyday.

Ron Bennett
.(JavaScript must be enabled to view this email address)

I need to explain this better because I do commonly get people thinking that .yahoo would be for use by yahoo itself.  It would not, it would only be for resale.  One could be hard and require that they get the domains for their corporate use, everything not the admin of domain resale, from another domain company, or softer and just require 99% resale so as a major reseller they could put their own domains like search.yahoo within their own TLD.

The reason this is not the same as sub-selling yahoo.com relates to the problem with .com.  Since .com is generic for commercial, ownership of generic terms like drugstore.com has been handed out to the first to ask, or in some cases the highest bidder.  Thus we fight over them.  Under my proposal, once we wean ourselves from .com, nobody owns any generics.  They own only brand-based names, which is the way trademark law figured out it should be.

Daniel Tobias says he doesn’t want the TLD’s brand in his domain.  Of course you don’t.  Your domain, dan.info, is a perfect example of the problem.  By being first to register, you own the term “dan” in the info space.  Weren’t you annoyed when Dan Cohn got dan.com before you?  There are tons of Dans out there and the name should not belong to any of you in a generic sense.

Having a small collection of low-meaning generics (com,net,org,info,biz and even alt or mobile) helps the problem, but it’s really just a patch.  So now 5 people can own the term “dan” and that’s better than 1, but how much better?  And even so, everybody still wants .com the most.

The web links pointed to provide much more explanation over the confusion that this would just create a vanity domain for Yahoo.  I use .yahoo as an example because they have a global established brand in internet directories, and that is what I am suggesting is what would be needed to own a branded TLD.

Only branding allows a free competitive market between TLDs with chance for innovation.  With the current attempt at competition (many registrars to one registry of .com) we get a monopoly for Verisign and ICANN on policy, and the registrars can only compete on price and a very limited subset of services.

But there is no natural monopoly to serve on branded names, the way there is on generic names.  No need to regulate.  The space is infinite, as it should be.

Still getting some questions so let me make this clearer.  Imagine the system where it was required that TLDs be entirely made-up names with no generic meaning.  Drive down any silicon valley street and you’ll see scores of companies named this way.  “Agilent”, “Flooz”, “Napster” and such.

Made-up names qualify very well as potential brands, they are inherently not generic, and nobody else is using them.

In particular, none are more valuable than any other until the owner puts effort into making them valuable as brands.  Thus they are all on an even footing.  And thus they can be given a total hands-off ownership of their namespace because it’s very clear they created it out of thin air.

I use examples like .dunn and .yahoo because they have already established excellent brands in naming and directories based on non-generic terms.  They represent today what another TLD company might like to build.

For example, I notice that Google finds no reference to “wurkul” so it would make a fine new brand.  You could registere to operate the .wurkul TLD.  Then you must convince people to become your customers.  How you do that is up to you.  Price, service, policy, terms, quality, who knows?  Or perhaps you spend a lot to advertise to make the term cool, which is what many people do to build a brand.

Unlike .museum, all value in your brand you built yourself.  With .museum, there is an inherent reason that the Natural History Museum would want to register there. It’s inherently better than all other TLDs, and has a monopoly on that attribute.  And while that monopoly might be convenient and make the domain naturalhistory.museum very memorable and attractive, which of the many such museums should win it?

Now, as it turns out, I don’t think we need to be so restrictive on TLDs to insist they are made-up terms that never existed before.  It’s enough, as trademark law already allows, to require that the term not have a generic meaning in directories.  But to undersetand the pure concept, consider the made-up names.

The PW domain will address some of these issues.

The PW will reserve all second level domains.  There are no distinctions between made-up words, generics or trademarked terms.  They are all reserved.  Anyone who wants to can form a community at a second level term.  So, google lovers can start one at goggle.pw.  But they do not have exclusive use.  So google haters can form one there too.  Or, they choose a different string, such as googlerocks.pw.  Google shareholders could also form one there, and so on.

The second level becomes a portal for all things “google”.  And this is true for names, i.e. “smith.pw”, places, “seattle.pw” and groups: “redsoxfans.pw”.  The ideal is to allow communities to form and be directly accessible via the second level.

Individuals get to own the third level and establish a permanent presence associated with their community.

This is not too different than what the .mobile folks hope to do.  But they will be too hobbled by various policies (udrp, whois, sunrise) to make it work.

Tom Barrett

You can do that, but it seems very web oriented.  When I try to summarize DNS goals in a short form, I say this:

“I want to be able to meet you at a party, and tell you my e-mail address.  We both want it to be easy to remember and type, and when you use it, the mail should come to me with no chance of it going to somebody else.”

As noted in my goals essay (linked above) we have other goals, some similar to this (put a URL in a radio ad or on a billboard) and others different.  With e-mail there is no chance to intersperse a search or sitefinder, so it’s very important to consider that as well.  So how does PW deal with that?

Oh boy, where to begin…

You write the following:
” They own only brand-based names, which is the way trademark law figured out it should be.”

Trademark law doesn’t differentiate to the degree you suggest. While there is a “Priciple” and “Supplemental” register for U.S. registered/recognized trademarks, there are numerous “generic” word marks registered on the “Principle” register. So one can’t even rely on those…then there are the issues of foreign marks, “common law” marks, etc…

In short your statement about how trademark law has it “figured out” is way off! If anything, trademark law more convoluted than ever and no where near as clearcut as you make it out to be.

Later on you write this:
“But there is no natural monopoly to serve on branded names, the way there is on generic names. No need to regulate. The space is infinite, as it should be.”

Really? Many brand names are generic words. The domain name space is vast, but who wants a domain name that’s like 15 characters long and difficult to remember. Your suggestion of sharing a namespace sounds good in theory, but it’s never taken off for it doesn’t work in the real world…look at what happened to .name ... they tried the “sharing” approach and it fell-flat with few people using it; bet even you didn’t register a .name domain!? ... did you?

Then you contradict yourself:
“I use examples like .dunn and .yahoo because they have already established excellent brands in naming and directories based on non-generic terms. They represent today what another TLD company might like to build.”

Fact check:
yahoo is a *generic* word!

dunn is the base of *generic* some tenses of the *generic* word dun

yellowpages is a *generic phrase* ... yes, really! Check any decent dictionary.

WIPO is an acronym - that raises some tricky issues when it comes to branding.

So in short, all of your examples, other than perhaps “dunn” are clearly of a *generic* nature…if you’re going to use examples, pick some better ones! But as you’re likely find, it’s not so easy.

Your later write:
“The second level becomes a portal for all things ‘google’. And this is true for names, i.e. ‘smith.pw’, places, ‘seattle.pw’ and groups: ‘redsoxfans.pw’. The ideal is to allow communities to form and be directly accessible via the second level.”

Your examples are looking more like ordinary domain names - you’re basically re-inventing what already exists. Google and many other websites now have specific URLs for localized content, etc.

Your idea is .NAME revisted…
“Individuals get to own the third level and establish a permanent presence associated with their community.”

.US had basically that structure…and look at what happened…most folks chose to stick with their ISP/freebie hosts and/or registered their own domain name.

People want control and portibility…neither of which a third-level offers. For example, say I had a domain like bennett.wyomissing.us and then later I moved…I’d have to change domain name, email address, etc. Control and portibility are very important to many people.

Lastly, you are a bit hypocritical suggesting all of this when you own a generic domain in .COM of all places “TEMPLETONS.COM” yourself! The pot call itself black LOL!

Ron

Yahoo has some uncommon generic uses, but as I presume you know, trademark law cares on whether the use is generic or not.  Apple is of course a generic word but a perfectly good brand for computers because there is no such thing (until they made it) as an “apple computer.”

In the purest sense, as I write above, consider the proposal only with made-up, totally non-generic names if that makes it easier for you to understand.  I am considering removing Yahoo and the others from the examples because it does confuse people how trademark law allows a term which is generic in some contexts to be a suitable brand in other contexts.

Trademark law is indeed complex because dividing up a namespace is not trivial.  However, trademark law largely works, and has tons of established precedent, which is why I point to it.  Literally trillions of dollars of commerce depend on it.

I don’t understand your comment about my own use of a .com.  Even if that use had not predated these essays, my whole point is that it’s so seductive to get to own ordinary words, free of context.  Of course I’m going to do it and so is everybody else if we’re allowed to.

I hope this addresses all your concerns.  It is best to start from the position of using only made-up terms for TLDs, and then explore how one can fairly include existing but brandable terms, so I think I will revise the essays in that dirction.  I don’t understand your point about 3rd level domains, as I never suggested anything related to this.

Actually, I see that I already modified my main essay long ago to make it clear, so I presume the confusion here is people reading only the blog entry and not the real essay it points at?  I encourage those who might comment to read the real essay on how to break up ICANN for a better understanding.

There are two very important issues being raised here when it comes to online identities.

1. portability
2. navigation

1. portability.  What does this mean?  It needs to be framed in relation to something.  In the context of email, I prefer to think of it as independence from an ISP or email provider.  In other words, “branded” email is not portable.

Therefore, email addresses such as .(JavaScript must be enabled to view this email address), .(JavaScript must be enabled to view this email address), .(JavaScript must be enabled to view this email address) are not portable in the sense they are tied (and branded) to a particular provider.

Examples of portable email addresses are “non-branded”.  For example; .(JavaScript must be enabled to view this email address), .(JavaScript must be enabled to view this email address) I would consider as portable addresses.  Let’s leave aside the issue of whether the top level extension (i.e. .com) is a brand.

So, I think all would agree that the goal should be to have a portable email address.  Early adopters and geeks who have their own domain already have one.  But most of the world does not.  And, under the current gtld’s, probably never will.

Most internet users would like a simple portable email address.  Ideally, something they can relate to.  .name tried to deliver this.  They “failed” only in the sense that adoption of the tld did not happen fast enough to suit their investors or the expectations of the internet community.  However, this does not mean .name was not a good idea.  They just ran out of time and patience.  I think if they had a 5 year horizon instead of 2, they would have succeeded.  But opening up the second level means they want to be just like .com.

2. Navigation.  By this I mean: “how does someone find YOU on the internet?”.  Must they remember your email address?  Or, would it be sufficient if they just remembered your name or something else about you?

In the current scheme, only one individual owns “smith.net” or “smith.com” and uses it for themselves or small number of people.  There are millions of people who might desire to use that particular string in their email address.

Rather than remember their email address, what if you could go to a directory for people called “smith” and look them up?  Or go to a directory for people who live in Boston, attended a certain school or worked for a certain company?  All of these are potential “navigators” to locate someone, regardless of what email address they may have chosen to use.

These are some of the underlying principles behind the PW Registry.  Yes, it incorporates some elements of the original .name.  But it also leverages the reserved second level for directories and online communities.

Tom Barrett

Most folks don’t want to own a domain name to begin with. Costs money and is another thing they have to worry about…

Most folks are happy with an ISP/freebie email address(es) that works…with that said, I tend to agree with you that .name should have held out longer…many people want a simple, easy to remember email address and are willing to pay a few dollars per year for that.

However, websites and email are two very different things. While sharing email addresses in the same name space can work very well, sharing websites in the same name space has historically not worked well.

Like it or not, vanity domain names will likely become more scarce and valuable as businesses increasingly acquire “prime” domain names in .COM much like has happened with toll-free prefixes such as 800 and 888.

I’ve personally made well over a million dollars selling domain names (can’t get more specific due to NDAs) - sold $150K worth in the past few months; prices have greatly increased lately as companies recognize that domains are akin to virtual prime real estate. The cow is out of the barn…domain names as we know them are here to stay well into the foreseeable future.

In regards to what Tom wrote:
“Rather than remember their email address, what if you could go to a directory for people called ‘smith’ and look them up? Or go to a directory for people who live in Boston, attended a certain school or worked for a certain company? All of these are potential ‘navigators’ to locate someone, regardless of what email address they may have chosen to use.”

You have just described what Google, Yahoo, and numerous others are increasingly offering…localized searching. No need to reinvent DNS to support this when the private sector has already figured out how to provide such search facilities.

Ron

The error in this thinking is that just as .(JavaScript must be enabled to view this email address) is tied to Microsoft, .(JavaScript must be enabled to view this email address) is tied to Verisign and ICANN.  They are your “provider.”

The goal in my proposal of lots of competing providers for namespace is that companies would have to convince you, in order to get your money, that they would provide the permanence you demand.

Many people have lost their ICANN/Verisign domains and addresses due to the UDRP, or in cases like sex.com, NSI/Verisign’s alleged negligance.  So before I bought a domain, I would want to assure they had good terms, good reliability, and an audit and escrow procedure to assure me if they go out of business that my domain lives on.

A search, by the way, can help with a remembered E-mail, but what about when I give you my E-mail on my business card?  I don’t want you to search or have to search then, but I still want it to be readable and easy to type, and also easy to remember since I don’t really want to put a hard-to-remember name on my card and use a different easy-to-search name when I tell you the address.

That you can make a million speculating on names is the problem.  Did most of the names you sold have generic meanings?  Perhaps I am just a sore loser.  I was around when I could have had any domain for free, and decided not to register more than one because it would be wrong to do otherwise.  Perhaps I was an idiot, but I think it was a mistake to have built a system where people could make landgrabs for ordinary words and then make a million selling them.

Brad writes:
“The error in this thinking is that just as .(JavaScript must be enabled to view this email address) is tied to Microsoft, .(JavaScript must be enabled to view this email address) is tied to Verisign and ICANN. They are your ‘provider.’”

ICANN/Verisign set policy and helps facilitate registration functions…a whole different ball of wax, in my view, than a private company like Microsoft providing email service, websites, etc.

Brad writes:
“Many people have lost their ICANN/Verisign domains and addresses due to the UDRP, or in cases like sex.com, NSI/Verisign’s alleged negligance.”

Yes, the UDRP system has and still is being abused/overused. However with that said, many is a relative word…thousands of registrations out of tens of millions is a very small fraction. And the situation has improved - respondents are increasingly winning URDPs and domain name security has improved.

Brad writes:
“So before I bought a domain, I would want to assure they had good terms, good reliability, and an audit and escrow procedure to assure me if they go out of business that my domain lives on.”

ICANN isn’t totaly inept as some may believe. There is a basic level of service ICANN accredited registrars must maintain. With that said, registrars don’t have to provide much beyond maintaining a local Whois and allowing one to update DNS settings, Contacts, etc. Verisign does the hard work of running the actual zones - and has done an excellent job so far.

Brad writes:
“A search, by the way, can help with a remembered E-mail, but what about when I give you my E-mail on my business card? I don’t want you to search or have to search then, but I still want it to be readable and easy to type, and also easy to remember since I don’t really want to put a hard-to-remember name on my card and use a different easy-to-search name when I tell you the address.”

You’re mixing two different things…individuals and business. Most folks with business cards either own a business and/or work for a business…

Folks who run their own business will buy a vanity domain (just another cost of doing business in the 21st century) and will have that on their business card, while employees who work for a businss will have an email address assigned by their employer on their business card.

Brad writes”
“That you can make a million speculating on names is the problem. Did most of the names you sold have generic meanings?”

Yes, all generic.

Brad writes:
“Perhaps I am just a sore loser. I was around when I could have had any domain for free, and decided not to register more than one because it would be wrong to do otherwise.”

Don’t feel bad…I totally missed the first wave…I grabbed some decent domains on the second wave around 1997 when the “free” registrations expired in masse. The bulk of the domains I sold for large sums were of an adult nature - Verisign didn’t allow “dirty” domains to be registered (some were grandfathered in prior to early 1996), but I wasn’t involved with domains back then.

When I got involved with domains in early 1997 I became curious as to what domains could be registered and which couldn’t…did some research and by mid 1998 I believed that eventually another opening would present itself for “dirty” domain name registrations. By 1999 with the introduction of competition it became clear the opening was right around the corner…the word was CORE would have no restrictions unlike Melbourne IT and Register.com. Most folks the first day CORE opened didn’t understand that CORE was a group of registration companies (that was an alien concept back then) and didn’t know how to register a domain through them. I’d done my research, was prepared, and had some great luck…I happen to be reading a computer news site called NewsLinx that afternoon when I saw an article saying CORE had just opened for registrations. Needless to say, that’s the day I really hit paydirt…

Brad writes:
“Perhaps I was an idiot, but I think it was a mistake to have built a system where people could make landgrabs for ordinary words and then make a million selling them.”

“landgrabs”...yes absolutely - much like the 1849 goldrush, late 19th century landrushes, mineral rights, etc. Such actions can be bad and certainly are unfair. But on the other-hand, such “landgrabs” gives people the extra incentive to actively develop resource, often at much cost and risk, that likely wouldn’t be fully utilized, if at all, otherwise; Capitalism 101.

Ron

You wrote, and I agree, that you should be wary about getting your email from your ISP because if you change ISP you lose the address.  However, if you buy a .com address you are getting it from Verisign (or ICANN if you prefer).  The registrar is mostly an illusion doing the paperwork for a fee. The real record, the real serving up of your records, is from Verisign.  And if you want a domman that says “commercial” you have no choice, they own that space.

The landgrab here isn’t like one from the 19th century to encourage people to develop things.  People pay for generic names because they already have value, not because they have had value added.  Not that they can’t have value added, but the core problem is they have this intrinsic value, only because of this artificial monopoly given over a natural resource—in this case, our human natural languages.

I don’t think land grab characterizes the current situation.  The land grab here was over a long time ago, during the bubble.  Now there is a vibrant market for domains and websites owned by individuals and the property is valued by the market based on both its intrinsic worth as a name and any value-add the current owner may have performed (e.g., developing repeat traffic by publishing useful information). And this part of the market has little to do with Verisign.  Verisign usually doesn’t make a penney more when an Afternic.com member buys or sells a name for thousands of dollars.

I would challenge that.  I think .com has kept its cachet.  Names are owned.  Indeed, it’s pretty common that people won’t choose a name for a new business or product line unless the .com (or at very least .net) is available.  Domains themselves don’t sell for big bucks unless they have generic value.  If you are selling a domain that gets traffic due to its reputation, you are really selling a business, with not many exceptions.

Again I challenge—what merit is there is names from an infinite space being scarce?  What great value to we get that makes it worth granting this artificial monopoly?

I thik that, may be, Brad mistook the “xxxx.mobile”(1) for the “mobile.xxx”(2) or the “xxxxxx.museum”(1) for the “museum.xxx”(2).
The (1) are generic but not the (2)
EJS

There is also the teeny issue that trademarks are specific to a trade and region of operation. How do you get past that problem, even when you allow for namespaces?

Should .ibm belong to Internation business Machines, India Book Mart, or Indiana book mart?