9 July 2001 - WAI ER WG Telecon

Minutes taken by Wendy

Agenda:

Sean's EARL Primer: http://webns.net/roughterms/

Meeting Summary:

  1. We went through some of the issues on the AERT issues list to determine which were closed, which were for WCAG and which were for AU. WC will update the AERT issues list and send a note to both groups to let them know that we have these issues to add to their issues lists. However, since WC maintains the WCAG issues list, she'll update it to incorporate the AERT issues.
  2. We briefly talked about Annotea and EARL and the possible tools that might generate EARL.
  3. We talked about what WART will produce, a checklist, linearized EARL, EARL. WC will continue to work on WART.
  4. WL and SP agreed to continue working on a primer.
  5. SP feels the EARL schema is stable. There is some concern that recent discussion in RDF-core might change aspects of RDF, but that those changes wouldn't affect EARL much or at all.
  6. We're considering having our next F2F the first week of October in Brighton and we would like to try to coordinate with PF. CMN will take idea to PF. SP will send list of possible meeting places to CMN.

In Attendance:

Action Items and Resolutions:

WL and SP: Agreed to continue working on a primer.

Tools Techniques Issues List

#9

CMN: AU issue, AU point to EARL. say it's closed.

WL: EARL is the solution, we're addressing the solution not the issue.

#30

WC a WCAG issue

CMN: Argue for AU. in WCAG the test is simple, if you don't have an accessible page, you can use an alternative. If not, it's part of the assessment of accessibility.

WL: Can say that to both.

CMN: In WCAG, prefer not to do that, say "make an accessible page."

WC: Part of this is manual, but WCAG needs to make checks clear.

CMN: Run it against WCAG.

CR: There are some ways to do checking, "is this object described in the text page."

CMN: It's a possible negative test.

WC For ATAG Techniques.

HB: Do we assume that each version has a date that we can compare.

WC: Big lump or smaller messages?

CMN: smaller is better - easier to deal with.

WL: Separate subject lines.

WC: Easiest to say to WCAG, "issues 1-30 in our issues list is yours, add to your list."

#30 - WCAG, w/CC to AU

CR: Always disliked text-only page solution.

/* Chris gives an example */

CMN: A good technique.

CR: Then never come back to ATAG.

CMN: Real-world examples where can't.

#45

WC, CR, CMN: WCAG

WL: detecting accessibility issues...

WC: Waiting for checkpoint solutions for scripts.

#12 = ATAG

#14 = ATAG

#15 = ATAG

#16 = WCAG

#17 = ATAG

#18 = WCAG

Annotea and EARL

CMN: EARL is a language to describe problems in content. The problem that EARL doesn't address is how do you find those when you are doing an evaluation. Where do you put those results. Annotea provides one possibility, to associate results with content.

SP: Wouldn't have one inside the other. Since you have URI of page inside EARL.

CMN: Annotea associates 2 pages, it might get it from extracting from EARL. The annotation scheme needs the info. Annotation at: URI, it might be in annotation or referenced in annotation. An implementation detail of associating the two in a look-up exercise. It gives us an implementation of attaching the things and querying them. If we use that mechanism we will repeat the URI twice.

SP: Take it out of EARL.

CMN: Do that if we tie ourselves to Annotea. Don't think we should. Use it as A method.

WC: could store locally.

CMN: Or inside the page you are working on.

WC: Talked w/SSB lately?

CMN: No, have talked with Hiawatha Island (Frontpage plug-in - ACCRepair) and Usablenet (plug-in for Dreamweaver). They are both interested.

WC: What determines if they will implement them. SP and WL, where do you think we are?

WL: Saw thing for linear-EARL, is there an authoring tool for EARL?

SP: There is the bookmarklet.

WL: There is also linear version, could form basis of people wanting to write EARL who don't know.

SP: When handling RDF, know model. If you handle a subset, should be easy to output.

WL: WART should generate EARL w/out people realizing it is generating EARL. And the linearized version could be how EARL gets generated from plain text. Anything like UWIMP?

SP: I see what you're getting at, WART should be structured enough ....

WC: What about a checklist?

SP: If WART could generate EARL, then transform into XHTMl version, "here's what we transformed it from" if you care.

WL: Then generating data points. e.g., what problems scripts make, then say "what portion are the result of script stuff." My original point is, the DI WG could use a more technical explanation of what EARL could do.

SP: Started writing new primer. Quickly got technical. I drew a circle labeled animal circle inside "human" to show subclass example. when i loaded on IE, it only had labels. Want some way to linearize it, an alternative version. Transform gracefully.

WC: State of schema? Open issues?

SP: We should be fine despite some of the RDF-core discussions to change things.

WL: Anyone from WAI following?

SP: Nothing too controversial. It's stable. Need some more general documentation.

Face 2 Face

First week of October in Brighton?
Try to meet with PF?
CMN will take idea to PF.
SP will send list of possible meeting places to CMN.

Next Meeting:

Monday, July 16th, 2001 @ the usual time, on the Longfellow bridge +1 617-258-7910.

Telecon Details: Regularly scheduled ER WG calls are Mondays, 10:00 am to 11:30 am, Eastern USA Time
(GMT -05:00) on the MIT bridge (+1 617-258-7910), except when there's a joint ER WG/AU WG meeting that day.

Joint Meetings: Joint meetings with the AUWG are held the first Monday of each month,
except where noted, 12:00 noon to 1:00 pm, Eastern USA Time (GMT -05:00) on the MIT bridge (+1-617-258-7910)

Last Updated: $Date: 2001/07/14 16:24:00 $
by: Wendy Chisholm or Katie Haritos-Shea