clarify the exact MIME for .mjs

[bmeck]

No description provided.

[bmeck]

requested by jshttp/mime-db#88

[Trott]

LGTM

[Trott]

Arguably out of scope for this document, but if people are looking to us for the answer on MIME type, this is certainly the answer I'd expect us to give. So 👍

[ChALkeR]

The first part looks good:

The MIME used to identify .mjs files should be a web compatible JavaScript MIME Type

But I'm not sure about this

preferably application/javascript

because I don't know what's going on the application/javascript vs text/javascript question and which one is preferred now and by whom.

Perhaps /cc @domenic for that.

Note that I'm not implying that I prefer to see text/javascript there — I would have asked the exact same question if this change mentioned text/javascript as the preferrable mime type.

[bmeck]

@ChALkeR we could remove the second sentence?

[ChALkeR]

@bmeck I guess we could, this change would make sense even without mentioning the preferred one of those.

That addition of the preferred mime has its value, so if the decision could be made in a reasonable amount of time to favor one of those for some valid reasons, so I would prefer it to be mentioned, but if that won't be achieved — merging this without an exact preferred mimetype LGTM.

[ChALkeR]

LGTM, but either after elaborating (not in the patch itself, but in GitHub comments and/or commit comment) why should the mentioned preferred exact type (whatever it would be) be preferred, or with removing the «preferred» part if there would be no informed decision on that in a reasonable time.

[domenic]

text/javascript is preferred on the web generally; I believe some RFC tried to obsolete it, but it's the most prevalent on the web, so the HTML spec just made it the default.

[bmeck]

linked to HTML spec and updated

[ChALkeR]

@bmeck I am still not entirely sure if the HTML spec is the correct source for this.
What do other members of TC39 think? Is this opinion by @BrendanEich actual?

[ChALkeR]

• ECMAScript spec doesn't specify a mime type.
• IANA specifies application/javascript, and calls text/javascript «obsolete», but IANA isn't a very actual/up-to-date source.
• HTML spec specifies text/javascript and calls application/javascript «legacy» (upd: not any more), but HTML spec probably shouldn't be the primary source for JavaScript mime type.
• RFC 4329 specifies application/javascript as the primary and calls text/javascript «obsolete», but that's not even a spec (and it's 11 years old).
• @BrendanEich from TC39 wrote 6 years ago that the correct one is application/javascript (referring to RFC 4329) and that text/ isn't even the correct category and should not be used.
• @domenic from TC39 says that text/javascript is correct one today and that application/javascript is not, and refers to the HTML spec.

[domenic]

HTML spec probably shouldn't be the primary source for JavaScript mime type.

I dunno. HTML is the spec for web browsers, and web browsers are the major client that (a) cares about MIME types and (b) executes JavaScript. So I think it's a pretty good place for it.

[bmeck]

I would agree with @domenic here

[bmeck]

@ChALkeR are you ok with text/javascript since it is a preference, and it is not a mandatory thing? Is it ok to merge this PR?

[bmeck]

landed in 6eef91d

[ChALkeR]

@bmeck Sorry for a very delayed response on this 😞, I had some personal availability issues, hopefully resolved now.

The PR as landed in 6eef91d looks good to me.