tl;dr the problems are 1) relevancy, 2) integrity, 3) content management/curation.
If you've ever tried to maintain a large corpus of documentation, you realize how incredibly difficult it is to find "information". Even if I know exactly what I want.... where is it? With a directory, if I've "been to" the content before, I can usually remember the path back there... assuming nothing has changed. (The Web changes all the time) Then if you have new content... where does it go in the index? What if it relates to multiple categories of content? An appendix by keyword would get big, fast. And with regular change, indexes become stale quickly.
OTOH, a search engine is often used for documentation. You index it regularly so it's up to date, and to search you put in your terms and it brings up pages. Problem is, it usually works poorly because it's a simple search engine without advanced heuristics or PageRank-like algorithms. So it's often a difficult slog to find documentation (in a large corups), because managing information is hard.
But if what you actually want is just a way to look up domains, you still need to either curate an index, or provide an "app store" of domains (basically a search engine for domain names and network services). You'd still need some curation to weed out spammers/phishers/porn, and it would be difficult to find the "most relevant" result without a PageRank-style ordering based on most linked-to hosts.
What we have today is probably the best technical solution. I think the problem is how it's funded, and who controls it.
If you've ever tried to maintain a large corpus of documentation, you realize how incredibly difficult it is to find "information". Even if I know exactly what I want.... where is it? With a directory, if I've "been to" the content before, I can usually remember the path back there... assuming nothing has changed. (The Web changes all the time) Then if you have new content... where does it go in the index? What if it relates to multiple categories of content? An appendix by keyword would get big, fast. And with regular change, indexes become stale quickly.
OTOH, a search engine is often used for documentation. You index it regularly so it's up to date, and to search you put in your terms and it brings up pages. Problem is, it usually works poorly because it's a simple search engine without advanced heuristics or PageRank-like algorithms. So it's often a difficult slog to find documentation (in a large corups), because managing information is hard.
But if what you actually want is just a way to look up domains, you still need to either curate an index, or provide an "app store" of domains (basically a search engine for domain names and network services). You'd still need some curation to weed out spammers/phishers/porn, and it would be difficult to find the "most relevant" result without a PageRank-style ordering based on most linked-to hosts.
What we have today is probably the best technical solution. I think the problem is how it's funded, and who controls it.