- From: Sharon Laskowski <sharon.laskowski@nist.gov>
- Date: Mon, 24 Sep 2001 11:47:05 -0400
- To: Wendy A Chisholm <wendy@w3.org>
- CC: w3c-wai-er-ig@w3.org, asgilman@iamdigex.net, emile.morse@nist.gov
The draft of the cif available on the web site has most of the same content as the version submitted to NCITS for standardization; some of the formating and wording has changed. If anyone wants to see the current NCITS version, let me know and it can be arranged. Sharon Wendy A Chisholm wrote: > > Summary of action items and resolutions > - Action everyone: review the CIF > http://zing.ncsl.nist.gov/iusr/documents/cifv1.1b.htm > we'll discuss more next week. > - Action WC: check w/AG about assumptions and connections between CIF and EARL. > - Action SBP: send summary of CR and SBP discussions to the list. > - Action WC: incorporate short descriptions of implementations and look at > incorporating prose from Schema into the primer. Incorporate the summary > SBP sends to the list. > - Action WC: create RDF file of WCAG 1.0 checkpoints. > - Resolved 15 October - next joint meeting with AU. > - Action everyone: nominate items for the F2F agenda next week. Joint > meeting with PF. > > Present > - Chris > - Wendy > - Katie > - Sean > - Sharon > - Charles > > *** CIF > > SL free version (earlier, basically the same) is at: > http://zing.ncsl.nist.gov/iusr/documents/cifv1.1b.htm > I'm not sure of the connection between these two that EARL. > > WC Marking up the data. CIF looks like a schema to me. > > SL Based on best practices. Could view as fields. > > WC Assume Al's connection is that if EARL could be expanded to be the > markup language for CIF. > > SL Don't have large numbers of reports and info that need to be > automated. What kind of machine processing do on a report? Perhaps, > compare info from two users. > > WC Perhaps if consumed by procurement people, could compare tools. CMN - > value of EARL for evaluating AU tools? > > CMN Value is to query results. > > SBP EARL is basically a database format, it is non-proprietary. If putting > reports into a database format, EARL is something to consider. > > SL No incentive to put into database format since not enough of > them. Usability test is not looking at conformance. > > SBP Data analysis can only be done by a human. > > SL Right, it is context driven. > > CMN What you are storing is prose descriptions. > > SL Tables of time on task, or results of questionnaires. > > CMN Can use EARL for anything you can regularize. If you have a numeric > scale, options, that's a useful place to have EARL. The other side of > stacking up data, makes it easy to do a piece of evaluation, have someone > else finish it up next week. > > SL Not how a user test works. Could translate fields into EARL or another > language to store as database. If there were usability tests, how would > one use them? What automated things would you do with them? Some newer > approaches - where collect data from 100's of users - you might want to do > something with EARL in that respect. > > WC Other telecon that KH is organizing? > > SL GSA looking for knowledge management tools. Using CIF to get usability > info out in the open. Possibly looking at using EARL to store comparisons > of tools. But, not sure if enough data to use EARL. > > Action everyone: review the CIF > http://zing.ncsl.nist.gov/iusr/documents/cifv1.1b.htm > we'll discuss more next week. > Action WC: check w/AG about assumptions and connections between CIF and EARL. > > *** Test files and expanded WART > > WC At WCAG F2F decided going through test files and "voting" on what > accessible or not would be useful exercise to determine if everyone agrees > on interpretations of WCAG. Expanding WART to go through test > files. On-line form. > > CR Had associated with WCAG 1.0, now what about 2.0? > > CMN Most are useful tests for both versions. > > WC Status of test files? > > CR Going through second pass. Will there be an AERT 2? > > WC No, there will be an HTML Techniques document with "rules." Those rules > taken from AERT. AERT primarily HTML with a little CSS and scripting. We > will be expanding CSS techniques, and generating SMIL, SVG, etc. with > AERT-like rules. Please keep in touch with Matt May. > > *** tagged PDF repair process > > WC When generating files in PDF, will also generate tags that are stored > w/in the format that are basically HTML. However, don't always get the > best tags. e.g., table header shows up as TD instead of TH. Use outline > view to change. A last resort, not a way to generate content, but to > edit. Particularly since when regenerate the file, your edits are lost. > Not a way to send that info back to the generating application. We don't > deal a lot with PDF, but it is an interesting repair process that I don't > think we've encountered or thought much about. > > KHS Heather sent more info to AU list about Word creating PDF files. > http://lists.w3.org/Archives/Public/w3c-wai-au/2001JulSep/0113.html > > *** EARL Primer to EO > > WC Thoughts? Stuff to change? > > Action SBP: send summary of CR and SBP discussions to the list. > Action WC: incorporate short descriptions of implementations and look at > incorporating prose from Schema into the primer. Incorporate the summary > SBP sends to the list. > > *** EARL to XHTML > > CR Could we have a summary at end: 50 passed, 20 failed. instead of lists > of pass and fails. > > SBP No statistical functions in CWM. > > CR Can it format into a table? or graph? > > SBP Table, yes. graph - SVG? > > CMN If in xml send to graphviz...although probably not the time of graph we > want. Passing 15 out of 27 is not particularly useful info. These are the > ones you passed, these are the ones you failed is an artificial measure. > > CR Useful to know if passes/fails are P1s or P2. > > SBP Good point. Have to put that in the EARL itself. List of test files > and if P1 or P2...then could merge that. > > CMN Say that X, Y, Z meets checkpoint 3.1. > > CR The RDF refers to a test file, test file refers to AERT, AERT refers to > WCAG. Therefore, could get that info that way. Is it possible to go > through all the info using XSLT. > > CMN If AERT is regularly formatted, can strip it out. Could do, but > quicker to do by hand. > > CR Should the info go in the test file or RDF output. > > SBP We'll have an RDF file on the W3C web site so that we can merge that > info in. > > CR Someone needs to create that file. > > Action WC: create RDF file of WCAG 1.0 checkpoints. > > *** New structure of home page > > WC changed structure. Will be keeping up to date. Let me know what you > want to see on here, what is useful or not, etc. > > CMN Racing cars. We need racing cars on the home page. > > *** Joint meetings with AU. > > WC Used to have them the 1st of the month. Haven't had for a while. Next > week is the 1st but we'll be at the f2f. What would we talk about in a mtg > with them? > > CMN Test suite. 1st of november is out as well due to AC mtg. > > Resolved 15 October - next joint meeting with AU. > > Action everyone: nominate items for the F2F agenda next week. Joint meeting > with PF. > -- > wendy a chisholm > world wide web consortium > web accessibility initiative > seattle, wa usa > /--
Received on Monday, 24 September 2001 11:48:37 UTC