"DOMTimeStamp": renamed to EpochTimeStamp

[simonvarey]

MDN URL: https://developer.mozilla.org/en-US/docs/Web/API/DOMTimeStamp

What information was incorrect, unhelpful, or incomplete?

This page has a link to specification of DOMTimeStamp, but the linked page https://webidl.spec.whatwg.org no longer has a specification of DOMTimeStamp (as of 21 Mar 2022). I believe EpochTimeStamp has replaced DOMTimeStamp.

Specific section or headline?

Specification

What did you expect to see?

A page specifying DOMTimeStamp

Did you test this? If so, how?

Followed the link and searched the page.

MDN Content page report details

• Folder: en-us/web/api/domtimestamp
• MDN URL: https://developer.mozilla.org/en-US/docs/Web/API/DOMTimeStamp
• GitHub URL: https://github.com/mdn/content/blob/main/files/en-us/web/api/domtimestamp/index.md
• Last commit: 2279e5a
• Document last modified: 2021-09-14T18:08:09.000Z

[sideshowbarker]

Thanks for catching this and reporting it.

Given whatwg/webidl#1021: To whoever fixes this, what needs to happen is:

• Create a new EpochTimeStamp article, based on https://w3c.github.io/hr-time/#the-epochtimestamp-typedef, or else rename the DOMTimeStamp article
• Add the Deprecated_Header macro to the EpochTimeStamp article
• Add EpochTimeStamp to https://github.com/mdn/browser-compat-data, marked as deprecated
• Make sure the current MDN article DOMTimeStamp ends up resolving to the EpochTimeStamp article instead (either due to renaming or else because you set up a redirect)

[hamishwillee]

@sideshowbarker @simonvarey It isn't this simple, as you will see from this discussion: #11675 - and in particular #11675 (comment) :

DOMTimeStamp originally referred to itself as a timestamp definition where specific details were covered in other specs (I think this was in WebIDL spec). It was removed from that spec. It was separately redefined in the high resolution time spec to be this epoctimestamp, with precise meaning. So my concern was that the redefinition might only work for this specific spec definition and not the definition in other specs.

In other words it is good to create EpochTimeStamp, which I did in https://developer.mozilla.org/en-US/docs/Web/API/EpochTimeStamp , but we were unsure if the DOMTimeStamp should simply redirect because it is mentioned in other specs which have their own definitions that do NOT map to EpochTimeStamp (at least yet).

Thoughts?

[sideshowbarker]

In other words it is good to create EpochTimeStamp, which I did in developer.mozilla.org/en-US/docs/Web/API/EpochTimeStamp , but we were unsure if the DOMTimeStamp should simply redirect because it is mentioned in other specs which have their own definitions that do NOT map to EpochTimeStamp (at least yet).

Thanks — OK I now see whatwg/webidl#1021 (comment)

I don’t really have a good suggestion about what to do here. I can see now that this problem was created for us due to DOMTimeStamp having been removed prematurely from the WebIDL spec — despite the fact that other specs are still referencing it. That was a spec mistake that should not have been made. But we can’t fix that.

But given all that, I don’t think we should remove or change the spec URL in our DOMTimeStamp article, because it’s actually the correct spec link: It points to where DOMTimeStamp should still exist in the WebIDL spec — until such time as there are no other specs referencing DOMTimeStamp, at which point we can be confident that we don’t need it in MDN any longer, and somebody can do the steps listed in #14667 (comment)

[hamishwillee]

I don't think we should change the URL, but we might add some context in https://developer.mozilla.org/en-US/docs/Web/API/DOMTimeStamp ?

For example a note

DOMTimeStamp has been removed from the IDL in the linked specification, but is referenced in other specifications. In the High performance timer specification DOMTimeStamp has been redefined as EpochTimeStamp.

[sideshowbarker]

I don't think we should change the URL, but we might add some context in developer.mozilla.org/en-US/docs/Web/API/DOMTimeStamp ?

For example a note

DOMTimeStamp has been removed from the IDL in the linked specification, but is referenced in other specifications. In the High performance timer specification DOMTimeStamp has been redefined as EpochTimeStamp.

Yes, adding that note would mitigate the problem and be sufficient for resolving this issue, I think

[hamishwillee]

Done in #14675