<Brent> Eric is showing a presentation of how the repositories are structured and how they technically work together.
<Brent> Eric's presentation is being recorded for view by those not present.
<yatil> https://slides.com/yatil/wai-repos-tpac17/live
<yatil> https://www.w3.org/WAI/EO/wiki/EOWG_F2F_November_2017
<shawn> open issue -- markdown or HTML for new pages in GitHub
<yatil> http://prose.io
AWK: Thinking about both the new SCs and the impact that will have on the existing SCs
Katie: May have to update other EO resources such as Perspective Videos to include new SCs such as single character (only SC in there for voice.)
<shawn> ACTION: shawn (or anyone else) add to perspectives idea the speech access page - Kim has video -- add to GitHub
<trackbot> Created ACTION-382 - (or anyone else) add to perspectives idea the speech access page - kim has video -- add to github [on Shawn Henry - due 2017-11-14].
Kathy: Have list and analysis of existing places where mobile needs to be added. Have someone from EO to partner with the mobile TF and learn intent and help with clarity and rewrite.
... overall structure of Understanding docs - do we want more user focus? how to balance against length? need to introduce benchmarks.
AWK: Vision for Understanding docs is that they would move from TR space and would not therefore require the same update cadence (currently 6 months or so)
... comments could then be immediately addressed.
... we recieve about 10 per cycle at this time.
... expect more when new ARIA techniques are published, etc. Antiipate that when new SCs publish there will be a spike of comments.
... will use GitHub to address issues with pull requests and expedite the process. Joint TF is a possibility, clarification during re-chartering about which WG is on point for the Understanding docs.
Eric: Two things - link to the current style guide for new site. Secondly, Understanding collaboration can help lead work on new SCs and Silver.
David Mc: Yesterday we heard a study that reflected the perception that WCAG is too long (400 pages) despite fact that it is really just 38 pages. EO can help with that perception.
<shawn> ACTION: shawn (or anyone else) add to GitHub idea of thinking about Undestanding docs in new site...
<trackbot> Created ACTION-383 - (or anyone else) add to github idea of thinking about undestanding docs in new site... [on Shawn Henry - due 2017-11-14].
Makoto: In Japan we held a meetup focused on new SCs and got feedback on readability of (and ease of translating) Understanding docs. also indicated a need for inclusion of graphic examples, including videos. Could all Understanding docs have graphic and video to support non-English speakers like me?
AWK: Going forward, yes, and in the short term to the prospect of adding video and graphics seems like adding a lot of work.
... we should terefore be inviting more external contribution.
Judy: Need to clearly define responsibilities
Alex: Not sure about how much shelf life 2.1 will have as Silver comes out.
... before we spend all effort to fix something that is soon to be outdated, let's consider that.
AWK: Maybe some of this work can be repurposed.
Shadi: Ownership - while EO can be great support, given the status and importance of the Understanding docs, I feel like it must continue to be owned by WCAG as the arbiter of the standard.
... we can later revisit how we can port some of the material into Silver and it is a fair question but looking globally as it is converted to non-TR space it could be more easily addressed after WCAG 2.1 is implemented.
Kathy: I agree but think EO can help with consistency of writing style, tone, regardless of English as first or second language.
Bruce: Very excited to hear of the move to GitHub of comments, etc. We have not linked externally to the extent we had planned and it will be exciting for people to be able to go in and suggest links and additional changes.
AWK: May be more errors in what publishes but will have the ability to quickly correct them.
Shawn: Why would there be more errors if there is a good approval process?
AWK: We make mistakes now and so we must recognize that if we update more quickly we are likely to have more errors. Must be sure advice is sound and that is my point.
Wilco: Moving to GitHub gives us as well the opportunity to learn from the techniques of the open source movement, not relying as heavily on immediate polls but have a way for more people to think more in depth about it.
AWK: For 2.1 adding material in the mold that we currently have and thinking about changing the mold later on. So what does EO think it can do?
Brent: We wanted to meet to help the greater EO group understand what is the opportunity and the challenge and to figure out the process. Good to hear about the GitHub tool. Could we learn more about where everything is right now and what are next steps to engage two groups?
AWK: Now in the draft stages of developing Understanding there is a repository.
Sharron: How much update to existing SCs?
Kathy: Would like to see mobile considerations added.
Sharron: Do low vision and cognitive TF have the same intention?
Vivienne
<vivienne> https://www.w3.org/WAI/training/Overview
Vivienne: Sent a link to list, can look at where I am now with edits
<yatil> https://www.w3.org/mid/000601d357e7$3c0c5760$b4250620$@webkeyit.com
<yatil> ^^ Vivienne's email.
Shawn: suggest we address the issue of "digital" vs "web" as a separate issue.
Vivienne: Can do global swap and switch but for now the reference is to "digital"
<shawn> in Vivienne's doc, highlight in yellow are Vivienne's comments
<shawn> MEGA THING - add "last reviewed" to evergreen pages with old dates
Shawn: We should consider phases - 1. now fixes, 2. later revisions and now at least add new material, e.g., Perspectives video
Eric: I wanted to make a point on linking to archive pages
Shadi:Similar to better web browsing: some of the specifics are out date; *however*, the concepts are still relevant and useful.
<shawn> ... some of the same issues with links from the Training Suite
Eric: If content is evergreen, why not remove the dates entirely?
Amanda: won't really work if people wnat to cite dates in a paper
Vivienne: People need a reference point
Brent: I want to be sure that there is a standard way to determine what date content was last changed, when it was last reviewed.
Robert: I am in favor of regular review cycles on everything that we publish, similar to medical guidelines ACP American College of Physicians. Things fall off when are not cycled through the reviews.
Brent: and some date in there must be within the last three (or preferably two) years.
KrisAnne: As part of orientation for newbies, we might have them read several of the things that are up for review.
... the redesign can help make that happen.
Charlotte: We need an editorial calendar - how to track what has been updated when.
Amanda: If we want to go with "last review" we need procedures and more than one person to do the reviews.
... keep the group consistently rotating through all the content.
Sharron: We have the RM cycle, but it was interupted by the redesign effort.
<yatil> https://www.w3.org/WAI/EO/wiki/Resource_Development_Life_Cycle
Brent: Prior to what we have done there have basically been just a few editors. making all participants into editors is how it is now done and we may want to look at that.
Shadi: There have been many external editors in the past. In most WGs there are only a few editors who take input and then implement what the group wants.
Brent: My point is that we are changing roles and increasing the number of people doing the work with new tools and a redesigned site.
... after we get through the push of launching the new site, we can go back to the RM review cycle and implement it.
Shadi: in Chairs breakfast we spoke abut about onboarding new chairs and editors. And actually, if you're asking folks to manage resources, you're asking all of them to handle all of this
Sharron: And soon the ability of the participants to amke and keep their time commitments becomes an issue.
Vivienne: And at my small company, we have to do EO and other WG work on our own time.
Eric: And I am just not seeing that we're getting content movement currently. The question is: With the challenge of limited resources, limited time and a large amount of editing work that must be done, how do we get to where we need to be?
Eric: we must be sure our resources are ready for the new site. Must define MVR - minimum viable resource - and then must work to do just that and set up the editorial calendar.
... prioritize individual resources, get them to minimum acceptable state and then the RM makes only those changes needed for the first round.
... must limit discussion time in EO.
Sharron: How can we apply the MVR to soemthing like the Training update? This would take days. Another example is the complete overhaul of the Business case which I am hoping will receive lots of thought and discussion.
Eric: must face the fact that the kind of grand changes we envision just will not happen soon.
Brent: In some ways, the tools are getting in the way.
<shawn> Several: Tools are getting in the way.
Shadi: Do the small group leaders understand how to lead it?
Shawn: Take the example of Perrspectives Video TF - talked a lot on phone and Brent was going to lead, but didn't have time. Shadi ended up leading and did a lot up front preparation before each meeting
Brent: But even when you asked me to lead the video TF, it ended up that you actually lead it. It was all I could do to join the meeting and be prepared.
... I could not, you are the one who had the time to prepare the meeting, create an agenda, etc. I could not do that.
Shadi: Maybe my mistake was that i did not train you to do it, but took over on my side.
... but now that you have seen how it worked, could you do it?
Brent: Yes I probably could.
Amanda: I have another role as a chair for another group I am involved in and I have a template and understand what is expected and how to conduct the meeting.
Brent: I tried to codify it in the PM meetings I had with the editors
Shawn: It is a combination of time available, skills people bring to it, but mostly lack of progress is because people are not meeting.
... selecting tools was an example where we could have gotten in sync if we had been on the phone earlier.
... eowg has said we want to work on just a few things at a time and you all have changed that process without giving enough consideration to the fact that people do not have enough time to be a lead editor, a part of a review team, and participate in the approval process within the same time frame.
Vivienne: I need definition about how to priortize those roles, can someone help me since my co-editor is not available? What we will actually do with this (update on existing site? wait for new site?) Teams are better, I need someone to bounce ideas off.
... is there someone who wants to work on this with me?
... concern on AG work (Understanding docs) how will we be able to do this when we cannot get our core work done? So what do I do?
Sharron: I think we agreed no need to wait to do the updates to the Training Materials until the other resources are finalized.
... Vivienne it may help if you put your recommendations on the GitHub (Shawn will show you how)
Brent: We still are stuggling with how to follow W3C consensus process while allowing RMs more actual agency. I see where we send things out for review and no one does it and other times where after what I think is a thorough review and then find other people have found things that I missed. I may or may not agree but the fact of a way to register and consider the varied perspectives is really important. We do not want to diminsih the quality or the broad relevance of what we post. We need to distinguish how to determine when issues are important enough to stop or delay publication.
Shawn: how often do you find error states in things you read in the public space and not report it? My point is that if we put things out there, we may find that there could be errors that we do not find that no one ever reports to us.
Amanda: If there is a review cycle as proposed, there would be fresh eyes on it anyway.
Brent: I am trying to say I am not sure which direction we should go - we *have* to get stuff done and out there but I see the other point that the only way we catch some of these is to do the full cycle of review.
... some times it takes three months - and we have more than 80 resources.
Shawn: It does not have to, it may be a problem with the fact that we have so many things going on at once.
James: I would like to see some data on engagement, can we find out from everyone without putting them on the spot. Do an honest survey about is it worth your time? are your constibutions valued? Those kind of questions.
Shawn: We should know how it has changed? When was engagement higher and why?
James: I put a lot of work intro entirely rewriting the Intro document and I have no idea what happened there - why to prioritize when the value is not clear?
Shawn: You are right and I feel bad. It is on my plate and I often work all night long. We have too many things going on at once, work that is not getting followed up on by the planning team and by the reviewers, and by all of us.
Brent: In a survey we need open questions to learn what is bothering everyone. It may be different for everyone.
James: We will see trends and in engagement surveys you find patterns. Connecting the work we do to our mission.
Amanda: I finished a listening survey where volunteers said they would help and then did not follow through. The survey had lots of complaints - not helpful and so asked how to improve and got lots more useful suggestions.
Brent: Yes, Vivienne and I did that somewhat.
Eric: When you comment on stuff and don't get the feedback recognized or no steps moved it forward, interaction falls off, then it is frustrating and you lose motivation. We need to make progress by having prioritization, assigned timelines, and more actual management of production.
Shawn: I agree with both these points and one of the big issues is to invest an enormous amount of time and it sits there. So why not crank them out in a linear way, dedicate the time, and make people feel a sense of a accomplishment.
James: I think no one feels like it is a realistic expectation to churn it out because it will not get published.
... the videos were different, we paid a team to do them
Shadi: Or look at the Policies, or the other resources that got made. I liked the point you made about connecting it to the mission of the organization. WAI is part of a standards organziations and though we are a bit different than the other groups. Other examples that have worked fairly well besides Videos -- QuickRef, Getting Started Tips, etc.
Shadi: each change is surveyed and they spend so much time on minituea and a level of scrutiny that we do not come close to. We are trying to fit in something that has to be considered, as a WG we have a commitment to consensus, otherwise would be a Community or Interest Groups.
... there is also the question of values, members value consensus and the broad points of view reflected in the work. It is unique that it is done in a multi-stakeholder format and the tradeoff is that you have to take feedback and raise issues and so yes your stuff gets "chopped up" as James said and you must be prepared for that.
Eric: When we put editors who are volunteers through that scrutiny it is discouraging. Can we get people to change the way they report problems, become easier for the editor to respond.
KrisAnne: Have not been an resource editor but as the end user who needs this stuff. When it comes to all the things that people are required to do around accessiiblity and the speed in shich they change, I think somehow we must facilitate the process and rework what we mean by consensus, becasue people are not responding.
... what we do is really really important and I want to see resources out and developed and published since people need them and will continue to need them badly.
... let's remember the improtance of what we produce, connect to the mission of education and outreach.
... power through the times of disengagement and low energy to recall why we do this and how important it is.
James: I totally agree and thank you for that perspective. These types of documents that you are talking about. If we subject the document to the same level of review, we cannot scale. At some part you have to accept when things are not done in the same way or not the way you would do it, or you cannot scale.
... my point is that if we have documents that people who have interest and skill can write why would we put it through a grinder and take it apart - it does not scale.
Shadi: To make sure that we maintain the quality and broad perspective. You must have a way to review the products of a group of volunteers who have highly distributed writers - how to do the quality assurance?
... if as KrisAnne says these are valued resources,it is because of the qulaity.
James: So you are saying that the quality is not sufficient on its own for us to be able to scale?
... we must be careful giving groups autonomy because different levels of skill make them unable to be trusted?
Shawn: It is more of the broader perspective - example of the Biz case that was heading down a road that is narrow (US-centric) and only from other input did we see that we needed a broader view.
Shadi: in terms of autonomy, there is a charter, minimum required participation, mechanisms that assure that despite group differences there is a consistency in quality output.
... we *do* have a scalability issue. We want the broad perspective, stakeholder input, etc while introducing more rapid ability to update and change.
James: And so maybe we just cannot scale. If we were a community group would we have the same responsibility for the resources?
Sharron: no, and no staff support
Shawn: yes
James: Then we should accept that we cannot scale and remain a WG
Shawn: We should learn from what Sharron and Brent have tried to do and keep working to make it better.
Robert: First to the point that "nothing has shipped." We have shipped the Policies resource. Our group is unique, may be ways that other groups operate that we can learn from. One of the things Eric said before he left was "perfect is the enemy of good" Policies was held up by one tiny thing after another.
... we do not ship, and we too often operate out of fear. We need to ship, we need to deliver, we must embrace the culture of shipping things, publishing. We live in 2017, we can change things that are less than perfect, we are capable of embracing a more agile process. We should have consensus and maybe make a lighter process that looks at overall quality but allows for more trust and confidence in one another. Figure out a way to ship things without getting bogged down in process.
Shawn: I am very aware of what has shipped, we shipped a lot more before we started this distributed process. We have not yet gotten one thing through the small review team to the larger EO. I actually do belive in that process of small team review but it has not gotten there. We do not know yet how the process will work because we have not done it. I do really think that it will work better as we learn to manage it.
Vivienne: As someone who has woked on many WGs I have seen situations where nothing at all got published. This group in fact has one of the most coorperative and with the strongest sense of belonging and that we are all about wanting to help others use what we produce.
... without Shadi's and Shawn's staff participation nothing would go anywhere. It is part of the W3C process and we must work within it. It has a distinct value, meaning and distinction that when we tell people we are members it is meaningful with a purpose different than other types of organizations.
... it is hard when people are volunteering on their own time and that causes constraints. Many of us are providing staff within smaller budgets. We have the potential to do a lot of positive stuff.
KrisAnne: Would people feel differently if we did develop standards. Do people look at us differently?
Shawn: There are people who think only those who develop standards should be WGs. There is also a movement within W3C for there to be more "explainers" developed for W3C specs.
KrisAnne: I feel like everyone is in this room for what we believe in and not because we have an ulterior motive around getting standards that help our own products.
... we take the jargon and make it available for everyday people.
... people are here because they want to be here doing this work to help others.
<shawn> +1 for little victories --- maybe going back to focusing on fewer resources at a time will help?
Sharron: A big part of the reason for not shipping, which I think is often overlooked is the redesign and the related big decision to not put old stuff in the new wrapper.
Brent: I appreciate Shawn how you have said what hasn't worked and pointed to ways it could work better. What we have to do now we must better understand and accept what are the constraints of the nature of the group and then figure out how to scale as much as possible within those constraints.
Amanda: Is the consensus 100%?
Shawn: Yes but people can abstain. Everyone must have the opportunity to review.
Brent: There are things brought up today - like the question James asked about why it is not working for them - that we need to answer. When Vivienne explained the things that caused her to struggle it is something that once I understand, I can address.
... seems like maybe to meet with editors more often or thinking of other ways to come to consensus, follow the process and may have to accept the fact that we may not be able to scale to the extent we want to.
Shadi: I think you have closer contact with the editors but they may not actually really know what they need to do, how best to do it, and may need more guidance about how to meet by phone when Friday meetings are cancelled, etc.
Sharron: Maybe we need a meeting at least once a month for people on the other side of the world.
James: We have had this conversation a lot, can we make some final decisions? Perhaps what we can do is 1. find out what is problematic for people 2.look at ways to address their problems 3. shepard them more closely 4. public shaming.
... if we set expectations and everyone agrees we report in what everyone has accomplished every week.
Brent: I want us to come out of this meeting with an answer to Eric's question about how we get things done to ship before Christmas 2020
... maybe we do the two week rotation, and start shipping.
Eric: I am not sure the timing will work, if we are on two week rotation it will depend on objections raised only for the big issues, etc. We would end up considering two to four at a time instead of 20.
Brent: That process sounds very much like how we did when Kevin and you and Shadi were working on those resources and everyone had to touch each one several times.
<shawn> +1 to James about setting up expectations and sticking to it ... with a little cautions
<krisannekinney> +1
James: If we set the cycles, stick to the timing, introduce a steady cadence. If you don't come to the meeting you don't get to play. And six weeks from now, we have a resource to ship, the next week another and so on.
Eric: If we make it work, it will be like a train where a lot of energy needed to get it going and then it will flow
... we would have to not have big resources all the time and make sure they are rotated in a way that makes sense.
Brent: And people understand that if they miss the review cycle you can always enter an issue in GitHub for next review.
Shawn: if we make it really clear and stick with it, it will facilitate the process. I still think we do want to take advantage of small group work as was done in Policies and videos.
Amanda: I think that's right, can speed things along without all people having to touch it at every stage.
... maybe on THursday we can come to the table with some ideas that are a hybrid of things.
Brent: And we heard from people who wanted more focus on fewer resources. I also notice that we have this discussion every meeting. We seem to be making progress however and are figuring out how to engage people in the work, ensure that people feel valued and we can improve the process within the constraints of the WG expectations.
Eric: With the smaller review groups, the work seems distributed but there are both the editing work, review groups, and then another review with all of EO.
... more steps and so more work in some ways.
Shawn: I thought it would help but maybe have not seen any results.