Abolish Domain Names
Soon, domain names like “hubrist.com” and all other will be nothing but a quaint memory, anachronistic flotsam from a by-gone Internet era (i.e.: this one). Domain names are stupid. Here’s something better.
Everyone (including companies, cities and the like) gets their own personal IP address in the IPv6 address space. Hell, they can have their own block of addresses, probably a block of four billion. (According to Wikipedia, there are enough addresses in the IPv6 range that every atom in the body of every living human could have 7 addresses apiece. That’s just mental, which is how it should be.) But you don’t get to choose your address block; it is assigned according to an impartial algorithm based on your exact location in space at the time of your assignment. Your exact location is already known. Don’t sweat it.
All Internet resources are addressed directly by IPv6 address. We can dump the stupid socket model. Addresses of accessible services are stored in a hierarchical, distributed database, much like the DNS, but there’s no mapping of specific, hierarchical names like the dotted domain name model. The database includes complete owner name and service entries. To find a service, you do not have to enter in a domain name or an IP address. You simply enter in the names of the entity (person, company, family, city and the specific service into a query application which returns the closest matches, and which you can refine through more detailed information, or simply browse.
So if you entered “Hubrist Philosophical Enterprises, Ltd., Olympus Mons, Asimov Street, Mars; Help”, you’d be directed to the starting point of the customer service division of Hubrist Philosophical Enterprises, Ltd., as well as a list of more specific services: sub-services from which you could select if desired. Entities like Hubrist Philosophical could either leave the default listing service module in place, or provide their own override.
Of course, to access such a system we’d need something much more sophisticated than a mere browser. We’d need a proper Internet Operating System, which would automatically re-arrange its interface to support the sort of streaming data provided by entity service modules, fully extensible. It would not simply download files and open associated applications. It would be much more suave. It would open data streams and tie them into on-demand-available stream-appropriate user interfaces which would interact directly with the entity service. The local file system would be irrelevant, because file services would be a basic and mostly transparent Internet feature. So basic, in fact, you would not even realize there was such an abstraction as a “file” any longer. Service interfaces would seamless transfer persistent data to and from the user’s own file service, regardless its actual physical location.