Re: [dispatch] RFC 3896 and 3987 vs WHATWG URL Living Standard

[Alwin Blok <[e-mail]>]

Hello, 

I have recently subscribed to this mailing list after seeing Larry Masinter's message on the issue tracker of the WHATWG URL Standard <https://github.com/whatwg/url/issues/479#issuecomment-855303022>. 

I have done a lot of research on the WHATWG URL standard. Some of the results can be seen on my personal GitHub page here <https://github.com/alwinb/url-specification> and here <https://github.com/alwinb/spec-url>. I'm doing that work as an individual, I'm not part of a WHATWG steering group. I mention that just in case, to prevent confusion. 

I believe that an addendum to RFC3986 (URI) and RFC3987 (IRI) that specifies the WHATWG parsing and resolution behaviour, would be _very_ valuable. Another, maybe preferable option would be to combine RFC3986 (URI) and RFC3987 (IRI) into a new document that specifies elementary operations that can be easily combined to accurately and exactly reproduce the WHATWG behaviour. 

Personally I think that the WHATWG standard would improve greatly if the 'basic-url-parser' is refactored and split up into spearate parsing, resolving, normalising and encoding operations. I am doubtful that this will happen at all, though I do estimate that over time support for relative URLs will have to be added. 

I have thought a lot about how to resolve the, difficult situation and I've come to the conclusion that the WHATWG and the IETF can actually (I am serious) complement each other well here. The style of the IETF is more suitable for specifying the elementary operations and the deeper structure, whereas the WHATWG is limited in that area and can instead focus on the step-by-step instructions in pseudocode and on the specification of the web APIs.

The most important thing will be, that:

1) Any new IETF effort at an updated RFC3986/ RFC3987 specifies elementary operations that can be used by the WHATWG to implement the WHATWG behaviour, accurately and exactly, and

2) That the WHATWG acknowledges the new IETF effort, and explicitly points out that the behaviour is equivalent to theirs. This is similar to what has been done in the WHATWG Encoding Standard <https://encoding.spec.whatwg.org/#utf-8> for UTF8. I will add that personally I think that it is good for readers, but also as a gesture to the IETF, to make such a statement a bit more prominent.

All of this, is just an outline, of course, and it does require a lot of effort, still. 
But I believe that it can be done and that it is worthwhile. 

Regards,
Alwin Blok