Remove mentions of historic Accept-Charset header

[ddbeck]

What was wrong/why is this fix needed? (quick summary only)

Over in BCD, we're inclined to drop Accept-Charset. Browsers haven't sent this header for many, many years (IE since 2008, Chrome since 2013). After giving it some thought, I think mentioning in this decade confuses more than informs. Some text I removed makes a pretty good case for me:

With UTF-8 now being well-supported, being the preferred way of encoding characters, and to guarantee better privacy through less configuration-based entropy, browsers omit the Accept-Charset header: Internet Explorer 8, Safari 5, Opera 11, Firefox 10 and Chrome 27 have abandoned this header.

Nobody ought to send this header, servers ought to ignore it, and it's interesting mostly as a historic artifact. This PR scrubs most mentions of the header (except as one of the forbidden header names) and deletes the reference page.

MDN URL of main page changed

https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Accept-Charset

[ddbeck]

@wbamberg Perhaps of interest to you, since you and I chatted about it previously.

[github-actions]

Preview URLs

• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/MDN/Structures/Compatibility_tables
• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Web/HTTP/Headers
• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Web/HTTP/Headers/Accept
• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Web/HTTP/Headers/Accept-Encoding
• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Web/HTTP/Headers/Content-Type
• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Web/HTTP/Headers/Accept-Language
• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Web/HTTP/Status/406
• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Web/HTTP/Content_negotiation
• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Web/API/Headers/Headers
• https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Glossary/Quality_values

Flaws

URL: /en-US/docs/MDN/Structures/Compatibility_tables
Title: Compatibility tables and the browser compatibility data repository (BCD)
on GitHub

No flaws! 🎉

---

URL: /en-US/docs/Web/HTTP/Headers
Title: HTTP headers
on GitHub
Flaw count: 2

• heading_links:
  • h2 heading contains an <a> tag
  • h3 heading contains an <a> tag

---

URL: /en-US/docs/Web/HTTP/Headers/Accept
Title: Accept
on GitHub

No flaws! 🎉

---

URL: /en-US/docs/Web/HTTP/Headers/Accept-Encoding
Title: Accept-Encoding
on GitHub

No flaws! 🎉

---

URL: /en-US/docs/Web/HTTP/Headers/Content-Type
Title: Content-Type
on GitHub

No flaws! 🎉

---

URL: /en-US/docs/Web/HTTP/Headers/Accept-Language
Title: Accept-Language
on GitHub

No flaws! 🎉

---

URL: /en-US/docs/Web/HTTP/Status/406
Title: 406 Not Acceptable
on GitHub

No flaws! 🎉

---

URL: /en-US/docs/Web/HTTP/Content_negotiation
Title: Content negotiation
on GitHub

No flaws! 🎉

---

URL: /en-US/docs/Web/API/Headers/Headers
Title: Headers()
on GitHub

No flaws! 🎉

---

URL: /en-US/docs/Glossary/Quality_values
Title: Quality values
on GitHub

No flaws! 🎉

External URLs

URL: /en-US/docs/MDN/Structures/Compatibility_tables
Title: Compatibility tables and the browser compatibility data repository (BCD)
on GitHub

No external URLs

URL: /en-US/docs/Web/HTTP/Headers
Title: HTTP headers
on GitHub

• http://www.iana.org/assignments/message-headers/perm-headers.html (1 time)
• http://www.iana.org/assignments/message-headers/prov-headers.html (1 time)
• https://datatracker.ietf.org/doc/html/draft-ruellan-http-accept-push-policy-00 (2 times)
• https://datatracker.ietf.org/doc/html/rfc4229 (1 time)
• https://datatracker.ietf.org/doc/html/rfc5988 (1 time)
• https://datatracker.ietf.org/doc/html/rfc6454 (1 time)
• https://datatracker.ietf.org/doc/html/rfc6648 (1 time)
• https://datatracker.ietf.org/doc/html/rfc7230 (2 times)
• https://datatracker.ietf.org/doc/html/rfc8942 (1 time)
• https://developer.microsoft.com/en-us/microsoft-edge/platform/issues/18488178/ (1 time)
• https://developers.google.com/search/reference/robots_meta_tag (1 time)
• https://docs.microsoft.com/en-us/previous-versions/windows/internet-explorer/ie-developer/compatibility/jj542450(v=vs.85)? (1 time)
• https://en.wikipedia.org/wiki/List_of_HTTP_header_fields (1 time)
• https://html.spec.whatwg.org/multipage/semantics.html (1 time)
• https://httpwg.org/specs/ (1 time)
• https://w3c.github.io/ServiceWorker/ (2 times)
• https://wicg.github.io/webpackage/draft-yasskin-http-origin-signed-responses.html (3 times)
• https://www.adobe.com/devnet/articles/crossdomain_policy_file_spec.html (1 time)
• https://www.iana.org/assignments/message-headers/perm-headers.html (1 time)

---

URL: /en-US/docs/Web/HTTP/Headers/Accept
Title: Accept
on GitHub

• https://tools.ietf.org/html/rfc7231 (1 time)

---

URL: /en-US/docs/Web/HTTP/Headers/Accept-Encoding
Title: Accept-Encoding
on GitHub

• https://en.wikipedia.org/wiki/Brotli (1 time)
• https://en.wikipedia.org/wiki/DEFLATE (1 time)
• https://en.wikipedia.org/wiki/LZ77_and_LZ78 (1 time)
• https://en.wikipedia.org/wiki/LZW (1 time)
• https://en.wikipedia.org/wiki/Zlib (1 time)
• https://tools.ietf.org/html/rfc7231 (1 time)
• https://www.iana.org/assignments/http-parameters/http-parameters.xml (1 time)

---

URL: /en-US/docs/Web/HTTP/Headers/Content-Type
Title: Content-Type
on GitHub

• https://tools.ietf.org/html/rfc7231 (1 time)
• https://tools.ietf.org/html/rfc7233 (1 time)

---

URL: /en-US/docs/Web/HTTP/Headers/Accept-Language
Title: Accept-Language
on GitHub

• https://datatracker.ietf.org/doc/html/bcp47 (1 time)
• https://tools.ietf.org/html/rfc7231 (1 time)

---

URL: /en-US/docs/Web/HTTP/Status/406
Title: 406 Not Acceptable
on GitHub

• https://tools.ietf.org/html/rfc7231 (1 time)

---

URL: /en-US/docs/Web/HTTP/Content_negotiation
Title: Content negotiation
on GitHub

• http://httpd.apache.org/docs/current/en/content-negotiation.html (1 time)
• https://datatracker.ietf.org/doc/html/rfc2295 (1 time)
• https://wiki.whatwg.org/wiki/Why_not_conneg (1 time)
• https://www.eff.org/deeplinks/2010/01/primer-information-theory-and-privacy (1 time)

---

URL: /en-US/docs/Web/API/Headers/Headers
Title: Headers()
on GitHub

• https://fetch.spec.whatwg.org/ (1 time)

---

URL: /en-US/docs/Glossary/Quality_values
Title: Quality values
on GitHub

• https://bugzilla.mozilla.org/show_bug.cgi?id=672448 (1 time)
• https://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html (1 time)

(this comment was updated 2021-04-26 16:58:52.162579)

(this comment was updated 2021-04-27 12:58:16.996664)

(this comment was updated 2021-04-27 13:02:06.244975)

[hamishwillee]

FWIW If it were me I would keep the Accept-Charset page itself and the BCD entry, but replace most of the content with a single note capturing your points above - essentially it should not be used. The rest of the PR is great.

Why? Well, the header is still in live specs, and at some point people will reasonably search for it - resulting in "why haven't you got information on this" questions. I'd rather we had a minimal placeholder that explains that it exists but should not be used and that we have deliberately decided not to include info for it everywhere.

[ddbeck]

I'm more set on removing the BCD than I am removing the page. It meets all our criteria for removal and is the last feature between us and a stronger data quality check.

So if we let the page stand, then we'd have to drop the compat section and, with the progress being made on generated spec tables, the specifications would go someday too. It's a very thin page at that point.

[hamishwillee]

@ddbeck Perhaps I wasn't clear in #4250 (comment) - I am absolutely happy with removing BCD and most of this cleanup. I am also already suggesting a much much thinner version of the doc - so that does not phase me.

Essentially the difference is that in your proposal the header is missing even though it is in the specification - so we look like we're doing a crap job, and there is no way for anyone to know they shouldn't use it. With my proposal people can find the doc which says "don't use this".

Does that make sense?

[chrisdavidmills]

I think I agree with Hamish's proposal here. For something that was supported, but not isn't, it is useful to have a page available with a definitive conclusion on it of "this isn't supported, don't use this". Having nothing would lead to uncertainty.

In the past, the main cases in which we've removed pages are cases in which something was documented early, but then was never actually implemented. That's a bit different.

[ddbeck]

For some reason, the preview comment doesn't mention the page, but I've reinstated it, revised a little, and dropped most of its content:

https://pr4250.content.dev.mdn.mozit.cloud/en-US/docs/Web/HTTP/Headers/Accept-Charset

[hamishwillee]

@ddbeck Works for me (so approved). How about we add a deprecation warning up the top like this:

{{Deprecated_Header}}
<div class="notecard warning">
  <h4>Warning</h4>
  <p>Do not use this header. It has not sent by any major browser for many years and should be ignored by servers.</p>
</div>

[ddbeck]

@hamishwillee I tried your suggestion, but {{Deprecated_Header}} links to the (now non-existent) compat section, which feels pretty broken. (Also, it is so, so wordy.) So I removed it, but I left the warning as suggested.

[hamishwillee]

Awesome. Thanks!