If I'm allowed to innovate here: Howabout a DNS server only for Protocols?
since the list is short atm the broser or OS can periodically download the index and register each known protocol to a dummy handler that simply displays what one can do with it.
Monetization (if needed) can be done by auctioning commercial slots in the list. (Marking clearly which are free and open source and which are commercial efforts to support that protocol.) Each entry can have an explainatory link to a world wide web document, pfd, text file, DOCX, XLSX, PPTX, etc (go nuts) explaining these new and exiting times.
Ideally the software to work with the protocol can also be installed from or by the dummy at whim. If multiple [say] video:// handlers are installed the dummy can be configured to pick a default or present a menu every time thereafter.
Then we can end the days of circular finger pointing where not explaining a protocol is always someone else's fault and we silently agree to have error pages (or worse search results) if one tries to open a link like ipfs:// gopher:// news:// nntp:// etc
If all members of the collective of IT nerds know exactly what the user is trying to do or what desired behavior looks like we cant be hiding behind "I dunno!?" type error pages. Its just to embarrassing. It is our responsibility to teach grandma how to use magnet uri's if she appertains it.
What platform would the protocol handlers be built for? Would they have full access to the system like most native programs, or run through a sandbox?
I think it would be ideal if the handlers were able to run on a platform that was defined by open standards and well-sandboxed by default... the web is a good example of this.
I have to admit I was being facetious; my point was that anything solving this problem is going to have the same scope of web browsers and is more directly solved by them. If you imagine your setup was in-place, and pretty much everyone defined their own protocol with their own app, then that's roughly equivalent to the web as it is now, except also that the web has already done the hard work of establishing the open standards and being cross-platform. It's not clear to me what benefits introducing the things in your post would add to the web.
Web browsers have an ideological position that isn't compatible. Mozilla doesn't want to promote anything.
While we can create a system that supports n protocols we really only need a hand full of new things at a time.
The benefit is fetching stuff over ipfs, onion, gopher, freenet, zeronet, dat, blockstack, news and even irc
Until we can visit ipfs://example.com after a clean install the whole project borders a pipe dream.
I argue that if you cant see the benefits we've done a terrible job explaining them to you. It is sort of a chicken and egg problem. Why would I put a news:// link on my website if you cant do anything with it?
The expected behavior is a prompt asking if you want to install a news reader and register with a news server.
since the list is short atm the broser or OS can periodically download the index and register each known protocol to a dummy handler that simply displays what one can do with it.
Monetization (if needed) can be done by auctioning commercial slots in the list. (Marking clearly which are free and open source and which are commercial efforts to support that protocol.) Each entry can have an explainatory link to a world wide web document, pfd, text file, DOCX, XLSX, PPTX, etc (go nuts) explaining these new and exiting times.
Ideally the software to work with the protocol can also be installed from or by the dummy at whim. If multiple [say] video:// handlers are installed the dummy can be configured to pick a default or present a menu every time thereafter.
Then we can end the days of circular finger pointing where not explaining a protocol is always someone else's fault and we silently agree to have error pages (or worse search results) if one tries to open a link like ipfs:// gopher:// news:// nntp:// etc
If all members of the collective of IT nerds know exactly what the user is trying to do or what desired behavior looks like we cant be hiding behind "I dunno!?" type error pages. Its just to embarrassing. It is our responsibility to teach grandma how to use magnet uri's if she appertains it.