3 ms·
Can't we come up with a simple, good technical solution to identify hosts? Something that is scalable, is easy to use from a technical perspective and also eas
by DeathArrow 7y ago
Can't we come up with a simple, good technical solution to identify hosts?
Something that is scalable, is easy to use from a technical perspective and also easy to use by humans?
The biggest thing to solve is that meaningful dictionary words are few and many people fight over them. But the current solution to that means thousands upon thousands of tlds, which nobody cares of, are hard to remember. Most people still fight over .com domains and prices got to insane levels.
To have the chance to use a decent .com name for your business, you have to first buy another tld, work for a few years, pile up some millions of USD and get it from the domain hoarders.
Having a .com domain with a name relevant to your business is still perceived as a big advantage.
- clarry 7y ago> Can't we come up with a simple, good technical solution to identify hosts? We can, but every single solution is a set of trade-offs and even geeks will never agree on what's the right set of trade-offs. If the people who could give the tech widespread support don't agree, how is there ever any hope of rolling it out? The only way I see to shatter the status quo is if big companies decide to push ahead with whatever they came up with and the rest of the world just has to follow. QUIC is a bit like that, I think. Only problem is I don't there being big incentive for big companies to overthrow the system. It works well enough? Of course, I would love to throw away DNS and start using a system where I can be John Smith and you can be John Smith and I can pass your fingerprint to my friend John Smith so that they won't accidentally connect to John Smith the rich prince from Nigeria. I think one global namespace is a dumb idea.
- organsnyder 7y ago> Something that is scalable, is easy to use from a technical perspective and also easy to use by humans? That's exactly what we have right now. This is a business/culture/government/economics problem—not a technical one. As long as the end goal of the system is "type in a string of text on any computer connected to the internet, and get the same result[1]", we're going to have issues like this. When I got my first domain name, the cost was $100 for two years (and $50/year thereafter). At that price, no one thought of squatting on any but the most lucrative domain names—certainly not mass-registering thousands of domains and holding on to them for years. I don't want to go back to those prices, but those are the sorts of parameters that impact the domain name market. [1] Yes, there non-public zones, geo-routing, etc. But as long as we define "result" as "exactly what the organization controlling that substring wants you to get" I think it works.
- finnthehuman 7y ago>I don't want to go back to those prices, but those are the sorts of parameters that impact the domain name market. The amount of money businesses are willing to spend on branding is large enough that there is no price point that would allow me to both afford renewals on my email server's domain, but doesn't also enable squatting.
- labawi 7y agoThe mere existence of squatting is not as much an issue, as the scale when most domains are speculators and it's hard to get a reasonable name. Also, some TLDs have prior rights and domain dispute processes for speculators. Namespaces can be handled much better than current .com situation.
- throwaway55554 7y ago> Can't we come up with a simple, good technical solution to identify hosts? You mean like DNS? What we need is a simple, good technical solution that identifies hosts but removes money from the equation.