19 October 2021


AlexCone, Andrew, anuvrat, AtsushiShimono, AvneeshSingh, BenjaminDeKosnik, Bert_Bos, BertBos, bkardell_, briankardell, ChrisNeedham, cwilso, DanAppelquist, DietrichAyala, dom, EricMeyer, ivanherman, JamesGraham, JamesNurthen, jeff, jeffJaffe, JohnRiv, JohnRiviello, LouayBassbouss, louisMaher, marie, MichaelChampion, MichaelKleber, PeterRushforth, Pierre-AntoineChamin, RakashiMinamii, RijubrataBhaumik, Sam, stephstimac, TakioYamaoka, tantek, WendySeltzer, Williams, WolfgangSchindler, wseltzer
cwilso, dom

Meeting minutes

chairnick: stephstimac

steph: I'm a PM on the MS Edge Dev Exp team, and this is about the WebWeWant project.
… WWW is a x-browser standards initiative to identify problems people who build with the web have.
… started in 2019
… we're asking "if you could change a something about the web, what would it be?"
… initially started as an event session, where attendees could present lightning talks on their wants to a panel of experts;
… due to COVID, we stopped the live events, and we've worked through the backlog of wants; we still get incoming wants, though.
… Right now we have ~160-170 wants from developers
… [goes to https://webwewant.fyi/]

<dom> The Web We Want Web site

<dom> WebWeWant/webwewant.fyi repo

steph: 3 different kinds of wants - 1) easy feature requests for browser features
… particularly devtools...
… 2) broader platform features - CSS or JS features. More difficult to get fixed, as they require standards.

3) "Moonshots" - broader problem that needs quite a bit of research and design effort, as well as standards work to address.
… with the slowing down of events, we've pivoted to thinking about how we can better use this data
… what we're trying to get out of this session today is what's missing from the initiative and the data we have right now.
… we haven't gained a lot of traction in being a definitive source of developer engagement.

<dom> Wants as Markdown files in github

wseltzer: I see a real synergy between this and the incubation pipeline trying to hear needs that people have for the web.

<dom> W3C Incubation Pipeline

<Zakim> jeff, you wanted to comment on moving topics to CGs and standards

wseltzer: are there places the tools could talk to one another, that the people could talk to one another, ???

jeffjaffe: interesting effort; on how to funnel this into various standards groups...
… I don't really understand how far you've gone on the basic processing.

steph: triage has been done, and they've been categorized... the ones that are on the web site have had some basic research/validation done.
… some of the wants have been ranked.
… some of the wants, there's not a lot of detail. Some of them are almost explainer-status.

jeff: if you have an important issue with a spec, sure, you can raise it in the standards WG.
… all of the groups have Team Contacts you can reach out to.

<Zakim> tantek, you wanted to comment on overlapping efforts

tantek: I'm here to comment and also as a participant. I appreciate the detail Stephanie outlined in the breadth of issues that come through WWW.
… in general, many of the wants we've seen are user-centric, which is an important perspective.
… at W3C, sometimes we try too hard to fit everything into a standards-shaped hole.
… standards are developed in a lot of different places (TC39, W3C, WHATWG, etc) so it's good to have both perspective,.

ChrisW: we have a bunch of different participants from across the overall web platform space, not just browser vendors
… it's not triaging in terms of weeding out the junk, but instead feeding ideas
… one challenge in incubating standards is to go from "I have a problem" to a detailed explainer to fill that gap
… how to prioritize this - each vendor will have a very different prioritization
… getting lots of different perspectives helps

JohnRiviello: is there a way those of us who speak at conferences can help promote this?

steph: you can evangelize this as a community platform, people can submit directly on the Github or the web page, and participate in the discussion
… we are looking for more engagement to prioritize
… there is a prioritized view, I'll share

Dom: thanks for the presentation - great source of inspiration. Are there any success stories you can share?

steph: most of the wants that have become features are devtools features: someone did submit container queries, but I don't want to claim that as our success.
… I like that idea.

tantek: did we talk about the distributed upvoting system that Aaron built?

steph: no. on the website, Aaron hooked up webmentions, so we get broader reach via Twitter et al. We also wanted to hook that into Github to show want-ers.

<Zakim> dka, you wanted to ask: do you encourage developers to document the user need when they are registering a want?

tantek: in particular, this was trying to plug into the community, and use the actual web to participate in the Web We want, not just build another silo

dka: one of the things we do in TAG is ask to document user need
… what do you to do inspire developer wants?

steph: this does come up when we do research to validate whether these are validate wants

<dietrich_> i wonder how the WebWeWant list compares to the MDN DNA results of top developer pain points....

dka: when we ask for use need, we could ask them to point to WWW

cwilson: we've looked at tying back to webwewant as one of the signals of developer support in our chrome process

<Zakim> cwilso, you wanted to respond to Dan

<dietrich_> thanks @tantek, i'm in a mutezone!

steph: re: MDN DNA, it's somewhat comparable.

bkardell: they ask the questions in very different ways
… we also have State of CSS, which asks for similar data but in very different ways; I don't think the data ends up being highly correlated.

steph: any other thoughts?

bkardell: I wonder if anyone on the call has submitted a want?
… it would be good to get some feedback

<jgraham> The MDN survey ususally shows compat as a big problem, but I think the web we want is more related to features rather than difficulties with existing features

<jgraham> *new features

peter rushforth: I submitted one
… it's excellent to be asked what you want.
… please keep it going
… it doesn't come up on my radar too often, but I'd like to help promote

<tidoust> [I note that the "If you are already attending one of the events we’ll be at, would you consider presenting your idea in person?" question does not seem to have a "yes" choice]

tantek: you can post a like of a want on your own website, and that will show up

jamesn: when you look through these on Github, if we know there's work to solve this want going on, is it useful to add a pointer?

steph: yes, definitely. any references are very helpful

wseltzer: in terms of browsing the site to add that metadata... is there any way to browse the site to see the metadata ("security", "css" etc)

<bkardell_> this is kind of an interesting idea wseltzer

steph: yes. On the website there isn't a good way to navigate the tags, but you can go into a want that is tagged "accessibility" and click the "accessibility" tag to get all the wants tagged with accessibility.

wseltzer: thanks, the WebAppSec WG is looking for good ways to get developer feedback.

steph: the website is in the process of being redesigned to make it easier to navigate.

bkardell: I think it's a great idea to let WGs know they could participate in their own triage.

kleber: do you have any thought of the freshness of things on the site? Some things seem a little stale (article posted from 2014 on promoting secure sites in Google search results, e.g.)

steph: I think we need to grow capacity and depth of research bandwidth.
… we still want to bring up things that weren't addressed; but we have limited triage capacity, so we're not always experts.

tantek: how many folks here are going to go to WWW to submit a want?

bkardell: this seems like a somewhat biased crowd.
… but pull requests are welcome!

tantek: or file bugs!

tantek: wondering if folks here every run into times that they don't know what group or organization to go to?

Dom: working on web components frequently has this kind of confusion

steph: incubation pipeline is the right way to express our primary intent.

steph: question for the group: is there anything that may be missing from this data that would make it more helpful?

peter rushforth: maybe search? More categories?

Sam: can people submit contact info?

steph: yes

JonRiviello: would be nice to know if there was like a playbook for these to go through the process

the web we have

james: can you vote without sharing on social media?

steph: on github; we want to implement an upvote system on the site that doesn't use social media, but that's tbd

james: github upvotes aren't synced with the site


tantek: are there other places we'd want to post upvotes other than social media?

dom: +1s/thumbs up on GH would be useful.

steph: we have to figure out how to get them properly synced

steph: any other feedback?

dom: we have the collection of data from browsercompatdata/MDN - that might be another entry point?

<tantek> also answered to james: you can upvote from your own website in addition to social media

steph: we talked about caniuse, etc.

cwilso: not sure if it's possible to unify

dom: related to this remotely is browser-specs data

I look forward to steph solving this! :)

browser-specs: A machine-readable list of specifications used to build Web browsers

bkardell: one of the things that makes this hard to track is that although you can track back from shipping, most wants don't progress in a linear fashion to shipping
… it's a squiggly line.

steph: thanks all!
… you can find me on WWW or twitter (seaotta)

<dietrich_> thanks for hosting this!

Minutes manually created (not a transcript), formatted by scribe.perl version 136 (Thu May 27 13:50:24 2021 UTC).


Succeeded: s/web mentions/webmentions

Succeeded: s/plug in to the community/plug into the community, and use the actual web to participate in the Web We want/

Succeeded: s/browsercompat data/browser-specs data

Maybe present: bkardell, chairnick, ChrisW, cwilson, dka, james, jamesn, JonRiviello, kleber, steph