Do Over Dot Com Years on Mobile

Suppose you could travel back in time. There are many ways you could imagine making money. Ken Grimwood explored this theme in his wonderful Replay. And of course many other authors have done time travel. A favorite idea is to bet on something. A baseball game, horse race etc. Those have the allure of quick money betting on a simple event. So long as you remember that horse race and its winner of course.

What if you are into tech, like so many these days. But you are not one of those hyper geeks like the founders of Google, who were PhD students in computing at Stanford. Or Jeff Bezos, who was a computing undergrad major. You might be an arts major, or marketing or sociology. In tech, the Web dominates. And so do mobile phones.

Figure 1

Today is 2020. If you had a time machine, when would you go? How about 1993. (Was it really 27 years ago? Crikey.) That was the first year anyone could buy a web domain. Prior, you had to be a university (, or a tech firm doing research about and on the Internet (,, Now 1993 let any unwashed dweeb (you?) sidle up to a Registrar like GoDaddy or VeriSign, plonk down some money ($300!) and get a domain that none had already snarfed up.

You could buy a domain like, which someone actually did, and hold it for years until a good offer came about. That owner sold it for $5 million around 99, at the height of the dot com. So you go back to 93 and make sure you get it. Ok, you know that $300 price to get it in 93? That was the annual fee the Registrars charged back then. (Nowadays it’s around $12 a year.) So sure, you’d have to keep paying $300 a year till 99. But you’d know the big payoff would come.

To be sure, $5 million to sell is nice though not Google or Facebook level moola. You are modest. A few $million suffices. You don’t aspire to be a billionaire. And starting and running something like Google takes tech smarts and a full time commitment. So if you could get early, why not other domains.,, Whoa. Getting was legal in 93 if you weren’t Toyota. But Congress would make a cybersquatting law a few years later, so that trademark owners could get ownership of their domain retroactively. Ditto for or Kodak and Xerox were deliberately chosen to be made up non-dictionary words to ease getting trademark protection.

Still, and use dictionary words. You could get those. How about if Apple did not have it in 93? Be careful. Apple had the trademark on their name for use with computers. You have 2 choices here. One is just own and do not put a website on it. You are speculating. Which is fine. By deliberate intent, when ICANN, who regulates domains, delegated authority to the Registrars to sell domains, owners could hold for speculation. This made liquidity in a secondary market when domain ownership could change. Which in turn likely raised the value of some domains. The other choice is to put a website at your, about making recipes using apples. Or how to own an apple orchid. All these are outside the remit of Apple’s trademark.

What about getting or in 93? These are made up words, which would later be trademarked by those firms. When that happens with non-dictionary words, the corporate owner might be able to get a broad scope for the trademark. But in 93 those firms did not exist. Yes, you could own and

The problem is Shakespeare. A rose by any other name would smell as sweet. All those English poems about roses mention the loveliness. But the latter is about the flower itself and not the sound or writing of the word. If you owned, Pierre Omidyar in 95 would have come up with another name and been as successful with it. Brin and Page would have made a different name for their search engine.

The rule is if you are going to cybersquat a domain, pick a dictionary word.

Why dot com domains? Today if you look, the biggest Top Level Domain is dot com, with over 110 million domains. Of course, you don’t have to restrict yourself to getting these. In practice, however, these domains have the most value today. You want to go where the demand after 93 will be the greatest.

All this is just a pleasant daydream. Yes? You imagine the money you could have made on yesterday’s horse races. A harmless and useless gedanken, as a physicist might say. There are no time machines outside Hollywood. 93 is gone forever.

If you now look for a useful dot com domain, it’s hard, with 110 million existing domains. Most of the good ones are already taken.

Could you replay the 90s in another field?

The dot com happened on the desktop, on the PCs of the time. While mobile phones appeared in the mass market by the late 90s, most of those phones had no cameras, no Internet, no GPS. Essentially everything we now take for granted about phones was impossible in 99. The vast mobile market of the last 5 years is like the desktop market of the 90s, only bigger. It made Apple into a trillion dollar firm. And it gave an already large Google a vast Android market. In the developing world, the cellphone is the only personal computer most people have ever had.

We suggest this equivalence. 2020 on the cellphone is akin to 1989 on the PC. 1989 was the year the Web was born.

Figure 2 — Equivalence between 1989 and 2020

So what? In 1989, software on the PC consisted largely of isolated programs. Most ran standalone. A few connected via dial up modems to servers. But most of those servers, in turn, were isolated. In hindsight, we know what happened in the 90s. The Web was built out on the desktop.

Figure 3 is the global phone market in 2020. Globally, 80% of the time spent on the phone is in apps. Only 20% in the browser. Schematically that 20% is shown as the bottom row of the phone. It depicts the most common mobile browsers on Android and iOS. Everything you know about the value of a good domain name is for the browser. A domain is used for webpages that appear in a browser. The on the right hand side is a proxy for all domains.

Figure 3–2020 Phone Use

Now look at the bulk of the screen, showing popular apps. The difference between an app and a webpage is that the app is standalone, whereas a webpage links to other webpages. I’m not talking about the difference in coding languages but at the user level. Most apps are standalone. When was the last time you saw a link in an app to another app?

Those exist. For example, Facebook owns several apps. A Facebook app could have a link to a (Facebook) Messenger app. So if you click the link, and Messenger is not on your phone, it is installed and run. But this is a special case where Facebook has a family of apps and it codes them to allow this.

What about an app Alpha made by a firm and an app Phi made by another firm? Can Alpha link to Phi? Examples of these exist. For example, suppose Alpha is a travel app from National Geographic. You are using it to look at photos of the Grand Canyon. Maybe you’d like to travel there. The app has a link or button next to the photos. If you click it, an app made by a travel agency will be installed and run. It opens up not in a generic landing page, but in a page about the Grand Canyon. You don’t have to do a manual search of the app store for the travel app. And you don’t have to do a manual search within the app once it boots up, to find the Grand Canyon. In this second app page, you can book a flight or motel room for the Grand Canyon.

This use case is ecommerce, where the intent is for you to buy an item. All this sits on top of the Internet. Apps talk to their servers. But at the user level, it is as though each app lives largely in its own world. Just like the desktop apps of 1989.

While important, there is a bigger picture in how to go between apps. This is shown by [Jill Gamer] in Figure 3. It is a linket. The square brackets are its boundaries. It is clickable. Suppose Jill plays computer games and others watch her remotely. This is esports. Or they want to play her. A fan, Tom, might search his mobile browser for “top gamers”. He gets a page with URLs to various webpages of gamers. But one result is Jill’s linket. It is clickable. If he clicks it, he does not go to a webpage. Instead, Jill has preset her linket to point to a specific game app in the app store. Her linket maps to that app and to an Internet address.

On Tom’s phone, if the app is already on his phone it is run. Else the app is downloaded (with his permission) from the app store and run. In either case, the app connects to the address. At that address is another program that interacts. His app might be a read-only version of Jill’s game. He watches her play. The game monetises him by showing ads or subscriptions to remove the ads.

The game is its own game watching platform. No Twitch or YouTube acting as a middleman. Those sites skim off the cream. They do not bear the cost of making or sustaining a successful game.

There can be non-gaming uses. One is remote tutoring. Bob is a high schooler in Chicago. He needs tutoring in math. The problem is that local tutors who do live in-person tutoring can charge $50/hour. When he searches for tutors, he finds [Learn Math]. Clicking it installs a tutoring app picked by tutor Jane in India. Her tutoring is via the app and she charges only $10/hour. Bob benefits by finding affordable advice. She arbitrages the difference in cost of living between India and US.

If Jane finds another math tutoring app better than her current choice, she changes her linket to point to the new app. Her expertise is in math, not the app. The app needs her to get Bob. He pays her thru the app, with it taking a commission.

Figure 4 — Remote Tutor

I gave 2 use cases of linkets, for gaming and tutoring. But apps already are for many different topics. Just like domains. Linkets act similarly for mobile apps, not the browser. This is the equivalence between the 2020 phone and the 1989 PC.

How do you make money? Right now our firm (Linket Corp) gave away 20,000 linkets. Soon we will charge for new linkets. That 20,000 can be compared to the 100 million domains. The linket namespace is essentially empty. Here is the time travel aspect. It is as though you could go back to 93 and buy whatever domains you wanted.

YOU rerun the 90s on mobile.

Plus the linket namespace is far more expressive than domains, which are restricted to Roman letters and are case insensitive.

One advantage. White space. This is the point of those square brackets. There can be several words in a linket, with white space separating them. Like the earlier example [Jill Gamer]. Notice also the upper and lower case. You can easily read a linket just you are are reading this sentence. We have forced ourselves to put up with domains like for far too long. There is good reason why many languages have white space.

Another advantage. Full expression in any human language. Imagine a linket in Russian, [Ваня Петров] (Vanya Petrov.) Or Chinese, [方志成] (Fong Chee Sing). Or Hindi, [राजन सिंह] (Rajan Singh). You can also mix languages in a linket.

We studied the success of the domain Registrars. You can own a linket without attaching an app to it.

ICANN regulates domain Registrars. We believe we are not under the aegis of ICANN. We are our own regulator. This makes us the equivalent of adding up all those domain Regulators.

Interested? Get a free linket. At some point we will charge an annual maintenance fee, just like GoDaddy. But the first year is on us.

Dr Wes Boudville,

Inventor. 20 US patents on cellphones. Founded for mobile brands for users. Linket competes against Twitch and YouTube. PhD physics.