Publish dweb: URI spec draft and documentation

[flyingzumwalt]

Publish URI spec draft and documentation for fs: or dweb: protocol

Prerequisites: #3, #4, #6

[flyingzumwalt]

@jbenet @nicola is it going to be possible to work on this, or any of its prerequisites, this week?

[ghost]

I have the skeleton of a draft here: https://github.com/ipfs/specs/tree/doc/fs-paths/fs-paths

Should I just make that a PR so we can discuss inline?

(Sorry for my super-slow last week.)

[nicola]

I wanted to propose something silly, just to make sure we can also take it into consideration:

• rename (or create a symbolic link) /ipfs to /ipfs:
• so that we can justify ipfs://hash, without loosing the original juan's idea, assuming the dweb

However, while I am writing this, it is aesthetically really ugly!

[flyingzumwalt]

@lgierth please do make a PR.

Also - how can we decide whether to stick with fs: vs dweb: ? I think dweb is more accurate/appropriate, and it also seems to be getting more traction (I hear people mention it more than fs:)

[ghost]

Also - how can we decide whether to stick with fs: vs dweb: ? I think dweb is more accurate/appropriate, and it also seems to be getting more traction (I hear people mention it more than fs:)

I haven't heard dweb: anywhere before this sprint :) I don't have a strong opinion what it should be, dweb: is certainly less generic than fs:, and has less historical baggage.

This is likely a matter of gathering feedback from externals and friends, who are interested in the future of the project but have less of an internal perspective.

[ghost]

dweb: is certainly less generic than fs:, and has less historical baggage.

... and sounds less like ipFS, this potentially being more inviting to others.

[nicola]

Also, when it is about "data", fs might not be the best name!
fs/dweb is basically the scheme for the content addressable web (and actually a bit more)

[ghost]

If we're all confident with dweb: that's cool :) @jbenet @diasdavid what do you think?

[pawal]

I have offered a full spec on ipfs: and ipns: - short and easy to implement for everybody. Where is the spec on dweb?

ipfs/specs#139

[ghost]

@pawal let's get the IANA registration for ipfs:// and ipns:// rolling based on ipfs/specs#152 (comment)

[pawal]

@lgierth Not entirely sure what changes are needed based on the background material. Perhaps change the word multihash to hash?