<jeanne> Meeting: Silver Conformance

<scribe> scribe: janina

js: Louis, not recalling what you were going to write? ...

Louis: How might we score user's ability to define colors, customization

Updates to Conformance section of ED

<jeanne> https://raw.githack.com/w3c/silver/conformance-js-dec/guidelines/

js: Believe I've made all changes from last week ...

<KimD> § 3.5 Sampling

js: Believe we left off at different sample sizes ...

<jeanne> https://docs.google.com/document/d/1y_HOyuMKltOQoZr0Gk7hMQXi3Jd8Mc5fyr-XkH7kZQY/edit#heading=h.8h5xei6fw2my

js: Don't want us to spend time reading, so maybe we return to it ...

sampling proposal - Different sizes, different sample sizes

<jeanne> https://docs.google.com/document/d/1y_HOyuMKltOQoZr0Gk7hMQXi3Jd8Mc5fyr-XkH7kZQY/edit#heading=h.8h5xei6fw2my

js: All sites under 100 pages (including dynamic) need to test all pages
... These would be auto tests; and 10 of those 100 need to be in depth testing

<Fazio> manually or automated?

Makoto: Would test perhapw 40 pages

js: In the U.S. that would currently cost between $200-$300K
... That's prohibitive
... We can work on appropriate phrasing; concept is to apply all available auto tests

Louis: We probably want to indicate that common elements (footers, navbars, etc) need only be tested once

<Fazio> +1

js: Would people be comfortable referring to EM Secs 2 & 3 for how to select sample sets?

<KimD> +1 to common elements comment

Louis: Yes, no reason to reinvent

df: Concerned about how to auto test plain language
... Temp[late structure can remain the same, but the content changes; so testing language remains important

Louis: Purpose of sampling is not to find every issue, but to identify patterns of problems

js: Concerned about the cost for testing every page, especially for small companies

Louis: Wonders on how this might affect flashing content

df: It's about unique content; the more infrequently it appears, the more sampling we might need to get at any issues

Louis: So different scores depending on what pages are sampled?

df: So, perhaps a certain number from each category of content?

<KimD> Sites or products from 100 - 1000 pages or screens may

js: Let's now talk on the 100-1000 range of pages

<KimD> - In-depth test of 40 samples, 15 selected and 25 random

js: I could see doing the same

<KimD> - Automated test of all pages or screens

js: We could say 10%; and discuss selected vs random
... I've done my fair share of testing for companies that wanted to test everything, and the truth is that one doesn't learn much new after a short time

df: Does EM give guidance on how to identify primary work flows?

<KimD> WCAG-EM: https://www.w3.org/TR/WCAG-EM/

js: We're going to need to give that job to the organizations themselves; we can't possibly generalize sufficiently
... EM does explain how to figure it out, and it's well written

<jeanne> https://docs.google.com/document/d/1y_HOyuMKltOQoZr0Gk7hMQXi3Jd8Mc5fyr-XkH7kZQY/edit#heading=h.bex9z1k5ual2

js: I have a rewritten point system proposal
... Please read and provide feedback
... Key is scoring by guideline, not by method
... supports setting a minimum for each disability category; and normalizable to treat each disability group equally

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/concerns/companies/
Present: jeanne janina Fazio LuisG Makoto KimD
Found Scribe: janina
Inferring ScribeNick: janina

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: 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.)