12 February 2001 ER WG Telecon Minutes

Summary of action items and resolutions




LK: I would like to create nice modular design for EARL. Not hacked up.

HB: Could this be XML module that gets attached to XML apps?

LK: Like XML but not RDF structure?

HB: No strong opinion.

LK: RDF can do logical operations but has more of learning curve.

LK: EARL could be oriented towards checkpoints (AERT) or could be lower level. Example: Checkpoint 5.3 is... Or attribute 'X' is... What is best?

[lower level EARL now called 'atomic' EARL]

HB: Checkpoints change so EARL must specify checkpoint level.

WC: ATAG checkpoints are used too.

WL: Are we looking at docs or programs?

WC: At FTF we decided to look at docs. Can claim conformance of tools to ATAG.

HB: I noticed that about 1/3 web sites say how they were created. Large amount were PowerPoint.

LK: If we look at authoring tools, they can say checkpoint 'x' pass/fail. Does it make sense for them to look at attributes? AU [reads 1.1] "authoring tool must be able to create accessible markup".

LK: What aspects of authoring can be automatically tested?

HB: Can check if tool can create something like longdesc.

LK: Judging authoring tool requires human. Generates list of RDF like "Len says pass/fail on checkpoint 1.1".

HB: Does this happen once per authoring tool or once per person?

LK: Once we can rate docs then we allow different tools to create docs and compare results.

HB: You create docs with specific tests in mind.

WC: Charles and Jan (from AU) have been working on testing tools. We were going to focus on docs.

HB: Are we integrated into Anotaya[sp?] (feature of Amaya)

WC: Some work but not yet.

LK: How attached to doc?

WC: Can point to whole doc or point in doc. Uses x-path and RDF.

HB: It's separate from document.

LK: Does it contain freeform text or what?

WC: Can contain any RDF statement or text. Anotaya stored in public server.

LK: Storage is not like EARL but use of x-path and RDF is like EARL.

[Review of Anotaya site at: http://www.w3.org/2000/10/annotation-ns]

LK: We want EARL to do computations on statements in Anotaya. Is Anotaya anything more than the yellow stickies?

WC: Just yellow stickies for now. Will you look at schema for use by us?

LK: Will start a thread. Will talk with Amaya group with regard to our use.

WC: This is like semantic stuff. We should contact Anotaya and invite to our FTF.

LK: Back to checkpoints vs. atomic checking. If WCAG was machine readable then could be helpful.

CR: Could we consider WCAG to cover all accessibility problems? EG. 508 has some checks outside of WCAG.

WC: SSB has checks outside WCAG as well.

LK: Machine check to convert to WCAG. Which group would create rule set?

WC: Ideally WCAG would have default ruleset. Would be great if ER could create rule set. Would be test to see if WCAG is specific enough.

LK: WCAG is normative. Would ruleset be informative?

WC: Both might be normative.

LK: Would be good if WCAG is machine readable.

WC: Yes. ER could create rules to check.

LK: Our example takes RDF and converts to human readable.

WC: Could we use our RDF and Anotaya, I wonder?

LK: Question for resolution (one of three options): 1) Forget about 'atomic' expressions. 2) Explore further. 3) Go for 'atomic'.

WC: Good to explore. Need action item that details what to explore. Issue with Jaws - alt/title only reads title. Look at modular rule sets and Jaws.

LK: Clarification of alt/title. Like 'hello world' exercise. Resolved to create at least 2 statements that do not deal with checkpoints and then convert to checkpoints. (Wendy's alt/title is one and Len will create another one.)

WC: Reading order is interesting too.

LK: Also tied to stylesheets [doc readable with stylesheets turned off]. Like with tool that uses stylesheets for layout.

WC: I would like to express in English statement (natural lang, pseudo code etc.) then later in RDF.

LK: Yes.

LK: We've been asked to provide monthly goals for group. Any suggestions?

WC: More work on rulesets for EARL.

CR: Open issues resolved.

WC: Many open issues are in WCAG 1 and may be resolved or moot with WCAG 2.

WC: Should we coordinate with WCAG 2 as we progress?

ALL: yes. Would be great if WCAG2 came with ruleset.

WC: WCAG2 is at least a year away so there is time for ruleset.

HB: Will WCAG2 align with 508?

WL: We are looking at 508 but not incorporating.

WC: We want to make WCAG clear enough that others are not needed. We like some aspects of 508.

CR: Get others to agree and create plan?

WC: We need to get EARL into existing tools (faroff goal) as part of implementation plan. What is immediate goal?

LK: Our goal: By end of March we will get 2 or more examples of atomic EARL. We will answer questions: What is capability of logic machines for N3 and RDF.

LK: Another goal: Answers to our fact finding. Anotaya - can it go beyond stickey notes? What can RDF machines do for us?

WC: Goal is: What tools will we need?

LK: Pointer to N3: http://www.w3.org/2000/10/swap/Primer

Next meeting

March 5.

Minutes by C.R.