1432084 - HTML `rev` attribute is no longer obsolete (HTML 5.2)

Status: RESOLVED INVALID

(Developer Documentation Graveyard :: HTML, enhancement, P3)

Description

[chharvey]

:: Developer Documentation Request

      Request Type: Correction
     Gecko Version: unspecified
 Technical Contact: 

:: Details

Since the release of HTML 5.2, a W3C Recommendation, `<a>` and `<link>` elements (including their `HTMLAnchorElement` and `HTMLLinkElement` counterparts) with the `rev` attribute are valid and conforming. This attribute became obsolete in HTML 5, but was reintroduced in HTML 5.2.

Pages to update:

- https://developer.mozilla.org/en-US/docs/Web/HTML/Element/a
- https://developer.mozilla.org/en-US/docs/Web/HTML/Element/link
- https://developer.mozilla.org/en-US/docs/Web/API/HTMLAnchorElement
- https://developer.mozilla.org/en-US/docs/Web/API/HTMLLinkElement

Please update the 4 pages listed above: remove the obsolete ("trash can") icon from `rev`, and update the Specifications Table to include the newest release of the W3C spec.

Here are some citations:

- `rev` content attribute on `a` and `link`: https://www.w3.org/TR/html52/links.html#element-attrdef-a-rev
- `rev` IDL attribute on `HTMLAnchorElement`: https://www.w3.org/TR/html52/textlevel-semantics.html#dom-htmlanchorelement-rev
- `rev` IDL attribute on `HTMLLinkElement`: https://www.w3.org/TR/html52/document-metadata.html#dom-htmllinkelement-rev
- GitHub Issue: https://github.com/w3c/html/issues/256

Comment 1

[Masatoshi Kimura [:emk]]

https://html.spec.whatwg.org/multipage/semantics.html#the-link-element:concept-element-dom
https://html.spec.whatwg.org/multipage/obsolete.html#attr-a-rev

W3C should really stop tampering their outdated snapshots.

Comment 2

[chharvey]

Which version(s) of the HTML specifications does Mozilla Develop Network follow? I was under the impression that MDN documents *all* versions of the specs, including those by both W3C and WHATWG. This is in the best interest of readers coming to the site to learn. If you don’t want MDN to follow any versions (including HTML 5.2) of the W3C spec, then remove all W3C citations from all documentation pages.

Comment 3

[Chris Mills [:cmills]]

This is an interesting situation, for sure.

We generally cite both the W3C and WHATWG versions of the specs, as we feel it is useful to reference both the snapshots as well as the living standard. We do however generally take the WHATWG spec as the most current/correct version of events, as that is usually where the newest updates to the language are present - we like to get things documented as soon as possible after they are looking like they are on the standards track, stable-ish, and supported by browsers.

Usually, the only difference is that stuff turns up in WHATWG earlier; the features are generally the same, or very similar.

This is the only case I've come across in my 4.5 years working on MDN where the two specs totally contradict one another.

So, I guess the specs need to be updated to agree on a position on this. I don't think it would be wise to make any changes to the documentation until this is resolved.

Comment 4

[chharvey]

Chris Mills---Fair enough. Indeed it is quite an unusual case. Until the Spec writers find a resolution, can we at least make some sort of note or warning on the documentation page, indicating this discrepancy? When I noticed it I was confused and thought MDN was out of date, or simply unaware of the changes in HTML 5.2; other readers might have those same questions.

Comment 5

[Masatoshi Kimura [:emk]]

Reopening for more discussions.

Anne, are you (or whoever in WHATWG) aware of this change?

Comment 6

[Anne (:annevk)]

I'm not, nobody asked for this on whatwg/html. My first impression is that this change does not make sense, as when we removed rev it was pretty clear that any rev could also be expressed as a rel and nobody really understood the distinction that clearly. (Also, implementations developers mostly code against only use rel.)

Comment 7

[Chris Mills [:cmills]]

(In reply to chharvey from comment #4)
> Chris Mills---Fair enough. Indeed it is quite an unusual case. Until the
> Spec writers find a resolution, can we at least make some sort of note or
> warning on the documentation page, indicating this discrepancy? When I
> noticed it I was confused and thought MDN was out of date, or simply unaware
> of the changes in HTML 5.2; other readers might have those same questions.

Good idea. I've added a suitable note to each page cited above.

Comment 8

[chharvey]

I think this can go ahead and be closed. Since W3C and WHATWG have decided to collaborate on a unified standard, the rev attribute has been marked obsolete once and for all. And the MDN pages have sufficient documentation to indicate that. Thanks everyone for contributing.