| 08:22 | <sideshowbarker> | annevk: After looking some at https://github.com/mdn/content/pull/40085 I realized that the MDN docs (and the spec) seem to completely lacking any high-level explanation or use-cases description for web developers of what they’d want to use the Encodings API for in the JavaScript code for a web application (especially not TextEncoder). |
| 08:22 | <sideshowbarker> | And I realized I also personally don’t have any idea myself of what developers would want/need to use it for, and what motivated adding that JS API to the Encodings spec to begin with. |
| 08:23 | <sideshowbarker> | I’m not questioning whether there are plenty of good use cases for all the parts of that API. I assume there must be. But I’m just saying that I don’t know what those use cases might be — and I don’t know how to find out what they might be, since the MDN docs and the spec seem to be lacking any details at at all about any use cases. |
| 08:25 | <sideshowbarker> | And lacking any understanding about the use cases, it’s pretty hard for me at least to really understand what the MDN docs rightly ought to saying about the parts of the API, and how developers should use them. |
| 08:27 | <keithamus> | As an aside: I don't have the permissions to edit that issue title but given the attribute has gone through a rename, would it be worth someone (@zcorpan?) editing the issue title to reflect the new attribute name of headingoffset? |
| 08:28 | <keithamus> | zcorpan: ^ not sure why your handle didn't turn into a mention. |
| 08:28 | <zcorpan> | Done |
| 09:05 | <annevk> | sideshowbarker: e.g., you have an application that deals with a custom file format that uses one or more of these encodings |
| 09:22 | <sideshowbarker> | Do you recall if there was a GitHub issue or something where people put forward some specific such applications? Or anyway, some specific use cases? I don’t remember, but was there, like, at time when then spec didn’t yet have the JS API but then enough people gave reasons for having something that you ended up adding it? Or instead was is that just already knew of use cases, and you based the API design on what you knew about the relevant use cases? |
| 09:30 | <annevk> | sideshowbarker: Joshua Bell proposed it. I don't recall the specific circumstances, but I think to most it seemed somewhat self-evident we needed such an API. |
| 09:31 | <annevk> | I think it makes sense to have APIs for most of the building blocks of the web. We still need one for MIME types. |
| 09:31 | <sideshowbarker> | yeah |
| 22:01 | <sideshowbarker> | @hsivonen:mozilla.org: if you've not seen it yet, https://www.ryanliptak.com/blog/better-named-character-reference-tokenization/ is a great read |