Closed foolip closed 4 months ago
And here are the raw responses: https://coda.io/d/Pain-Points-Missing-Features_d7UVQ_fZW0V/Pain-Points_suclo#_lu3Cv (Working on manually classifying the Interop ones)
Thank you @LeaVerou! If you see something in the data that doesn't come through in the summarized form, or quotes that especially telling of the web developer need, please feel free to share here or on the relevant proposals.
I've commented on all proposals I could find now. These responses don't have a proposal:
backdrop-filter
(edit: now filed as https://github.com/web-platform-tests/interop/issues/564)After manually reading all responses to the interop question and classifying them manually (including a fair bit of MDN research to identify alternative names, older names, map method names to APIs etc) these are the top 25 items:
# | Normalized content | Count | % of all issues listed |
---|---|---|---|
1 | Popover API | 263 | 9.74% |
2 | View Transitions API | 176 | 6.52% |
3 | <dialog> |
162 | 6.00% |
4 | Temporal inputs (<input type="date"> etc) |
130 | 4.81% |
5 | :has() |
110 | 4.07% |
6 | <selectlist> |
89 | 3.30% |
7 | PWA APIs | 75 | 2.78% |
8 | FileSystem API | 72 | 2.67% |
9 | Notifications API | 48 | 1.78% |
10 | Subgrid | 40 | 1.48% |
11 | Customized built ins | 39 | 1.44% |
12 | <datalist> |
35 | 1.30% |
13 | Web Components (generic) | 34 | 1.26% |
14 | WebGPU | 34 | 1.26% |
15 | Declarative Shadow DOM | 33 | 1.22% |
16 | CSS Anchor Positioning | 27 | 1.00% |
17 | Container Queries | 26 | 0.96% |
18 | CSS Nesting | 24 | 0.89% |
19 | Intl |
23 | 0.85% |
20 | Color picker () | 22 | 0.81% |
21 | inert attribute |
21 | 0.78% |
22 | Web Share API | 21 | 0.78% |
23 | @scroll-timeline | 20 | 0.74% |
24 | Scrollbar styling | 20 | 0.74% |
25 | Web Bluetooth API | 19 | 0.70% |
Huge thank you @LeaVerou! I know that categorizing freeform text responses is a lot of work, but it's also very valuable! I'll take a look over this and see if any make sense as proposals.
Taking the top 16 items (>1% of responses) from that table and mapping it to proposals or existing focus areas:
Some of the N/A items are worth looking into and considering a proposal, but overall I'm very encouraged by the alignment between these survey results and current+proposed focus areas!
Customized built ins does not have a consensus-based web standard. Apple has repeated stated our objection. Which is why it wasn't part of Interop 2023.
@jensimmons (or anybody else that's interested): let me know if you'd like to get early access to the data. We have a Discord group we use to coordinate and share early previews of the dataset and survey results.
Preliminary results in slide form from today’s presentation: https://docs.google.com/presentation/d/1AVsalCAZVITkYZCLxux5D2LbwLukjjuH9cm-vZRYy-w/edit#slide=id.p
More notably:
Thanks @LeaVerou.
What does "data table" mean? Under "missing elements (predefined)".
Thanks @LeaVerou.
What does "data table" mean? Under "missing elements (predefined)".
This was the full question:
(You can see all of them by going to https://survey.devographics.com/en-US/survey/state-of-html/2023 (though note that the response UI is now disabled, since the survey has closed)
So "data table" means an HTML table as we have now, but with built in functionality for sorting and filtering? Same semantics, new behavior?
So "data table" means an HTML table as we have now, but with built in functionality for sorting and filtering? Same semantics, new behavior?
Yup. I'm currently drafting a new Open UI issue on this.
So "data table" means an HTML table as we have now, but with built in functionality for sorting and filtering? Same semantics, new behavior?
I'd also hope for/expect built in support for pagination (or at least graceful hooks of some kind to help manage pagination (client and server)).
Update: Look at @LeaVerou's comment instead, as that categorization has been done manually and can be trusted to a higher degree.
State of HTML 2023 is still running, but @SachaG has shared some preliminary results for this question:
This is a freeform text question, and the normalized responses so far look like this:
Note that
not_selector
andhas_selector
are almost certainly overcounted, as "not" and "has" are common words, and the categorization hasn't been manually verified yet. For words that are not common in English, like "popover", I think the results are reliable.I will comment on the proposals for features that are listed here.