<cb> scribenick: cb

scribenick cb

introduction of participants

<tobie> +present

<vivien> +present

+present

<foolip> RRSAgent: are you logging?

tobie: I am interested: with the given tooling, what holes are there?
... how can these tools be funded
... this work should be paid
... with more money coming in the more incentive to work in these is

jyasskin: would love to find all uses of some definition in all specs

foolip: would be great to get a list of specs we care about

<foolip> scribenick foolip

<foolip> tobie: do we have a good understanding of all the tooling?

<foolip> tobie: there's Bikeshed and ReSpec (and wattsi...)

<foolip> tobie: then there's shepherd for x-refs

<foolip> tobie: for biblio, that's specref

<foolip> tobie: we have PR Preview for showing previews on GitHub PRs

<foolip> tobie: that relies on a bunch of services:

<foolip> tobie: html-diff maintained by dom

<foolip> tobie: and the Bikeshed web service

<foolip> tobie: and the ReSpec spec generator

<foolip> iclellan1: there's also echidna

<foolip> foolip: what's that?

<foolip> TabAtkins: that's autopublifying

<foolip> denis: there's a system where you send spec for publication

<foolip> denis: if you pass checks you get published to TR

<plinss> test.csswg.org/shepherd

<foolip> MikeSmith: with approval from the WG, it allows you publish living standard at /TR/, wher before it was outdated

<foolip> MikeSmith: in other words /TR/ becomes editor's draft

<foolip> MikeSmith: we should quit publishing old stuff to TR

<foolip> foolip: I distrust TR 100% of the time. How common is this?

<foolip> MikeSmith: it's better than before, but you should use ED for the very latest

<foolip> vivien: in the WebRTC all specs use echidna

<foolip> vivien: you can integrate this with your GitHub workflow

<astearns> You can reduce the 100% somewhat now due to echidna

<foolip> annevk: this is like the W3C's build.sh

<foolip> plinss: CSSWG also has a tool that regenerates spec

<foolip> the last time I tried to use Echidna I had to do some curl command, a GitHub App would be helpful

<vivien> +1

<foolip> tobie: I think we've listed all the tools now

<foolip> tobie: do we have gaps?

<foolip> tobie: what's missing?

<foolip> TabAtkins: the spec list obviously

<foolip> tobie: another thing is a way for spec editors to know where to add specs

<foolip> dsinger: my dream is that we manage the past and future better

<foolip> dsinger: in other words, would love a tool where the spec says "there are issues filed against this spec" or "there are provisional updates to this section"

<foolip> TabAtkins: can you file an issue at https://github.com/tabatkins/bikeshed?

<foolip> TabAtkins: there's a ReSpec session at 4 that an intern/coworker did to expose shepherd's link database

<foolip> TabAtkins: it's only giving you ReSpec syntax now, but could do Bikehsed

<foolip> TabAtkins: look up `bikeshed refs -h` for a local tool

<Zakim> vivien, you wanted to ask about use of tr.rdf (legacy) / W3C API ?

<foolip> vivien: don't know if people depend on tr.rdf?

<foolip> tobie: I use tr.rdf (specref)

<foolip> vivien: there's a new API (JSON) that replaces this

<foolip> tobie: I use RDF, JSON didn't exist when I wrote the tool

<foolip> tobie: does anyone else depend on this?

<foolip> foolip: I depend on it via specref

<Zakim> MikeSmith, you wanted to mention mdn annoys

<foolip> MikeSmith: MDN annotations

<foolip> MikeSmith: I built a thing for HTML which adds annotations in the margin where you define an interface / method

<foolip> MikeSmith: It also pulls in data from https://github.com/mdn/browser-compat-data

<foolip> MikeSmith: that's browser support data

<foolip> MikeSmith: in the HTML spec, you can see link to both MDN and the compat data

<foolip> MikeSmith: Marcos had a contributor add it to ReSpec

<foolip> MikeSmith: for ReSpec, it automatically generates those annotations

<foolip> MikeSmith: want to do this for Bikeshed too, not a huge effort, but need a contributor

<foolip> MikeSmith: so that using Bikeshed you get the same linking

<foolip> MikeSmith: maybe unique for the HTML spec, but we need to determine which features only have 1 or no implementation, for process resasons

<foolip> MikeSmith: having a way to do that for other specs would be nice, automating that would be great

<foolip> jyasskin: "Bikeshed can do that" has been said, and experts don't know yet

<foolip> [all] documentation needed

<foolip> jorydotcom: Bocoup's working on a guide for contributing to the web platform

<Zakim> jyasskin, you wanted to talk about documentation

<foolip> TabAtkins: who did the MDN annotations for wattsi?

<foolip> MikeSmith: me

<foolip> TabAtkins: let's talk about how you get the compat data

<foolip> tobie: funding aspect

<foolip> tobie: It would be interesting to find a way for folks working at companies to fund this work

<foolip> tobie: Would like to set up an opencollective fund

<jyasskin> https://opencollective.com/

<foolip> tobie: they're a startup for funding open source project, they're a legal entity that can take money and distrubute it

<foolip> tobie: know that the AMP team funds some projects that it relies on using opencollective

<foolip> tobie: would also be great with some way to monitor services, finding what needs help

<foolip> tobie: cost of hosting is low, and companies have payed for that, but someone needs a credit card

<foolip> tobie: AWS bill might blow up

<foolip> tobie: would like to see a more sane situation for this

<foolip> tobie: also to fund more people

<foolip> foolip: I have requested budget for this, think it would be great if others would do the same

<foolip> tobie: the structure is tied to my LLC

<foolip> tobie: for others to be able to do work, we'd need a bigger fund

<foolip> annevk: how does distribution work?

<foolip> tobie: would probably imply creating some sort of board

<foolip> annevk: would it be easier to make it part of the W3C?

<Zakim> MikeSmith, you wanted to ask about Outreachy and mention GSOC experiences

<foolip> MikeSmith: want to mention for history: I've tried in the past GSOC (Google Summer of Code)

<foolip> MikeSmith: proposals were rejected

<foolip> MikeSmith: that way of getting contributors didn't...

<foolip> MikeSmith: I know annevk has done work with Outreachy

<foolip> https://www.outreachy.org/

<foolip> MikeSmith: that's been more successful I think

<foolip> jgraham: I've used Outreachy for other test infrastructure stuff, think it's a good problem that I'd recommend

<foolip> jgraham: there are some practical things. you get a lot of upfront work, and you probably don't have 40 trivial bugs to work on

<foolip> jgraham: but if you get a good contributor they can do excellent work

<foolip> foolip: wonder about the W3C funding this, most members aren't super close to this?

<foolip> fantasai: W3C funding this would make sense, spec are core

<foolip> fantasai: but financials seem to be a mess

<foolip> fantasai: staff contacts sometimes haven't been able to attend meetings

<foolip> fantasai: they might be able to be an intermediary, but not give the money

<foolip> fantasai: it's the host orgs that have the money

<foolip> fantasai: hopefully this improves when the W3C becomes a legal entity

<foolip> MikeSmith: we're not a legal entity, and it is difficult for us to fund things

<foolip> MikeSmith: we have difficulties getting funds from hosts

<foolip> MikeSmith: we've had more success brokering funds coming in from members

<foolip> annevk: you'd think writing specs is high priority

<foolip> tobie: back to outreacht?

<foolip> annevk: it was pretty good

<foolip> tobie: I'd like to avoid having people working on critical open source being super low payed external consultants

<foolip> tobie: that's not what I'd like to see

<Zakim> MikeSmith, you wanted to comment

<foolip> mnot: is any of this written down?

<foolip> fantasai: people say "I need this tool" they write it and then we depend on it

<foolip> annevk: when Mozilla funded things our main requirement was that it was open source

<Zakim> dsinger, you wanted to talk about ISO

<foolip> ISO and ITU are both looking to modernize their infrastructure, they have no central anything

<foolip> that was dsinger

<foolip> dsinger: they're looking at commercial tools

<foolip> dsinger: doesn't make sense to me

<foolip> foolip: not having to write an elaborate contract is helpful for the funders too

<foolip> foolip: tobie, do you think this proposed model is problematic too in the outsourcing way?

<foolip> tobie: no, not in the same way [scribe abbreviated[

<foolip> jgraham: with Outreachy you're not super well paid, but it's not a way for the funder for get work done, it's more like an intern that could grow the communicty

<foolip> annevk: they're mentor programs

<foolip> tobie: next steps?

<foolip> foolip: should we just try the opencollective thing or does anything feel strongly otherwise?

<foolip> tobie: we can try this quickly, W3C would take a long time, only other option is individual contracting

<foolip> annevk: mnot raised some good points about this is scoped and managed

<foolip> plinss: I like this as an interim step

<foolip> plinss: only negative thing I see is the fees they charge

<foolip> tobie: but W3C overhead is huge

<foolip> fantasai: that might go down when they're a legal entity

<Zakim> jyasskin, you wanted to say we're out of time

<foolip> jorydotcom: with opencollective you can also have a "hosted" open collective, which is a collection of projects

<foolip> jorydotcom: that makes it easier to donate a large chunk at one time

<foolip> jorydotcom: they also make it possible to lower the fees

<foolip> jorydotcom: more organizational nesting and accounting

<foolip> RRSAgent: make minutes

<foolip> jorydotcom: let's talk to Max

<jorydotcom> foolip: do or do not, there is no try :D

<foolip> foolip: I'd be hesitant about adding delay to this, unless it makes others more willing to fund the work

<foolip> plinss: I'd be happy to be on the board

<plinss> https://opencollective.com/specinfra

<foolip> tobie: let's try and iterate

<foolip> fantasai: can someone link useful resources to specprod list?

<foolip> https://lists.w3.org/Archives/Public/spec-prod/

<tobie> RRSAgent: make minutes

<tobie> ScribeNick: foolip

This is scribe.perl Revision: 1.154  of Date: 2018/09/25 16:35:56  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/all use cases in specs/all uses of some definition in all specs/
Succeeded: s/use it/use tr.rdf/
Succeeded: s/build/built/
Present: TabAtkins Rachel Andrew Peter_Linss Invited_Expert Mek odejesush foolip Lan Kangchan dsinger duerst iclellan1 jorydotcom tobie
Found ScribeNick: cb
Found ScribeNick: foolip
WARNING: No scribe lines found matching ScribeNick pattern: <foolip> ...
Inferring Scribes: cb, foolip
Scribes: cb, foolip
ScribeNicks: cb, foolip

WARNING: No "Topic:" lines found.


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report


WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)