Point system/Levels & Conformance document

Lauriat: We had left off on Tuesday...we had talked through Bronze, Silver, Gold...how points would add up for each and then the points didn't matter for overall because the level would be determined the lowest of bronze, silver, or gold
... if you scores silver and gold in some categories, if you got bronze in one category, your overall score would be bronze since that's the lowest medal achieved in a category
... we were about to explore what kind of points scoring would we create to end up with bronze/silver/gold. for example, 30/50/100 for the medals, respectively
... generally higher points means higher level
... we had talked about having more difficult or methods more beneficial to user be worth more points
... if we have, for example, 30/50/100 for the levels. When we add different methods for a guideline, how would we maintain the levels and how do we determine.

Jeanne: One of the things JF said at the end of the meeting was setting a minimum test in each category that you must have and then have the level be your overall points
... so you had to do the minimum...it seems simpler and might be simpler to maintain

Charles: I think it has merit, but still faces challenge that a large enterprise could earn a lot more points than a small private site with a couple of page

Jeanne: only if you're scoring individual things. you wouldn't want to give points for each image that has alt text

Lauriat: If you have just a page of text. Like a restaurant that just has a single page with their menu. A lot of tests won't apply.
... we had talked about the applicability of user needs or some tests

Charles: will still have scenarios where some guidelines or methods or heuristic evaluations don't apply
... what happens if I have 9 gold and 1 bronze?

Lauriat: We talked about the concept of beyond bronze/silver/gold if the user needs doesn't have anything that would ever come up. It wouldn't count against you.
... As a side note, in the conformance super drafty draft document, I made an addendum called "sketching things out" and I'm listing open questions
... we also wanted to award more points for organizations that do more extensive usability testing; for example with people with disabilities
... but the points system right now is around methods, how do we translate that?
... for instance, if you did usability testing with users with just one category of user need. then you could essentially get a boost in that user need category

scottcooley: If they add heading structure, that would give them extra credit for applying heading structure method to make it more accessible

Lauriat: what we're trying to do is center around the impact to user. Maybe on the restaurant page, it has heading structure, it's all text, but the bottom of the page there's a one pixel image for logging
... and that's missing alt text
... as far as the user experience goes, it has no impact on the user for figuring out what's on the menu/navigating the menu
... it doesn't prevent the user from getting information. If the user stumbles on the image they don't get the information but it doesn't prevent them from getting info from the page
... but if you have a similar case with one image with no alt text, but there was content, like the address of the restaurant in the image that's much different
... we need to come up the lists of use-cases and tasks that people would want to do and information the user would want to get
... there should be some examples in the education and outreach documentation that we come up with
... If we go through the success criteria we have today and build a list of those that apply to one particular category of user need
... and then without rewriting them yet, look at them for a category and put them in a list, how would we build up a point rating system for each of them

<scottcooley> what is the URL of the wiki?

<Lauriat> Wiki link: https://www.w3.org/WAI/GL/task-forces/silver/wiki/Main_Page

<scottcooley> thank you

Lauriat: If you need some things to get to the minimum and you don't have content to get you there, how does that work?
... for coming up with points. that will probably come from the exploration you're going to do. The "how does that work with future maintenance" we can talk about now
... we could have a method today that is very popular, but if something better comes out a year from now. Would we want to change the number of points a given method that already existed and score the new method in relation to what exists?
... would we want to change the scores needed to get certain levels?
... this kind of brings us closer to how conformance applies to physical space
... if a buildling is old, the regulations changes, and in the future you need to renovate based on new regulations at some point

Jennison: What if there is a shopping flow. A site upgrades just the checkout flow. They make the checkout conform, but don't have budget to fix everything else.
... could they claim the other flows conform to WCAG AA 2.0 and the checkout flow conforms to Silver

Lauriat: I don't think so. We're moving away from full-page/full-site structure of website and moving to use-case and overall use-case for conformance
... the checkout flow would just be one piece of a user story
... this kind of gets to LuisG's point of how do we guide organizations to defining the tasks

Jennison: And some flows not be the high priority flows. I guess it's the same with WCAG. we're leaving it up to judgment whether they're claiming conformance
... and which flows are they prioritizing and how?

Lauriat: A company could say "these top interactions flows" work for users. Similar to having "meets with exceptions" in VPATs

johnkirkwood: I have some concerns. What happens when it comes to legal information. It doesn't have a lot of traffic, but it is important. It wouldn't meet a bar by traffic or user flow, but still important.

Lauriat: As another example, if your web application has a setting that says "enable braille display" or "magnifier" the end user that would want to use that likely only hits that once

<Lauriat> trackbot, end meeting

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)

Default Present: LuisG, Cyborg, KimD, Makoto, jeanne, Shawn, Jennison, Angela, MikeCrabb, Charles, AngelaAccessForAll, kirkwood, Lauriat, JF, shari, johnkirkwood, Scott, Cooley, scottcooley
Present: LuisG Cyborg KimD Makoto jeanne Shawn Jennison Angela MikeCrabb Charles AngelaAccessForAll kirkwood Lauriat JF shari johnkirkwood Scott Cooley scottcooley
Regrets: JohnFoliot
No ScribeNick specified.  Guessing ScribeNick: LuisG
Inferring Scribes: LuisG
Found Date: 14 Dec 2018
People with action items: 

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.)