What do you prefer: ipfs://ipfs/QmXf8 or ipfs:QmFx8?

[ghost]

Hello everyone! Let's assume that future web browsers will support IPFS natively. I have a thought about the way the hash gets referenced. Somewhere I have read that the links will be defined as follows:

ipfs://ipfs/QmXf8

Now that's a lot of redundancy, isn't it? And why is everybody sticking to //? The RFC speaks of [scheme]:[scheme-specific part] only. I propose following simple way:

ipfs:QmXf8...
ipfs:QmXf8.../about
ipns:QmZ7w...

Advantages: Humans like simplifications. Is it http://www.example.com./ or example.com on advertisements? Let's make it de-jure before it becomes de-facto.

Disadvantages: ipfs:// is recognized by people as "Link"/"Internet". → The whole discussion is futile if humans won't get in touch with hashes anyway since we will have to abstract that away.

Advantage: When we have a simple naming system, e.g. "andrew" => "Qm7Ei9...", "ipfs:andrew" or "ipns:andrew" is still much better than "ipfs://ipfs/andrew".
(We won't have it anyway: who's the registrar? A CENTRAL one? Never ever ^^ First come, first served? → What if the user loses his private key?)

Advantage: the broad public can learn: ".com" means that you have to open the browser, bw-pixels means that you have to open "QR-Code-Scanner", "#" means that you have to open a special service in order to read the #hashtag, "ipns:" means you have to open the browser. Simple, isn't it?

What do you think?

[whyrusleeping]

see ipfs/kubo#1678

[lidel]

and ipfs/specs#152

[ghost]

Thank you!
I close this issue now.