2006-114 Meeting Minutes

Return to Meetings

PDF-UA 2006-114
May 3, 2006

PDF/UA Working Group Meeting Minutes

Teleconference

PDF/UA (Universal Accessibility) Working Group Draft Meeting Agenda
Wednesday, May 3, 2006,  3:00 P.M. - 5:00 P.M. EST
Conference Call Information
International Dial-in: +1-303-928-2600
U.S. Dial-in: 1-888-742-8686
Conference ID: 7083945

Duff Johnson called the meeting to order at 3:05 pm EST.

Roll Call - Attendees

Duff Johnson DJ - Chair - Document Solutions
Angy Brooks AB - IAC
Joe Clark JC - IAC
Betsy Fanning BF - AIIM
Mark Gavin MG - Appligent
Loretta Guarino-Reid LGR
Richard Herring RH - IAC
Karen McCall KM - Karlen Communications
Greg Pisocky GP - Adobe Systems
Neil Soiffer NS - Design Science

Special guest: Professor Jacques Distler 

Action items indicated with an AI:

Approval of PDF/UA 2006-113, May 3, 2006 2006-113 DRAFT Agenda

JC: Jacques Distler University of Texas Professor who is knowledgeable about MathML will join us at 4:00. It was agreed to modify the agenda to accomodate Professor Distler.

Guarino Reid / Soiffer

Approval of PDF/UA 2006-112, April 11, 2006 MINUTES

Herring / Soiffer

Report on / update and discuss PDF/UA - Action Items

RH: Action item from March 14 now done.

DJ: Updating Wiki Contacts page. There is a roster page which has to be completed in order to remain on the Wiki and be recognized as a voting member.

DJ: Wiki development. The Wiki will only be accessible to registered users. Certain pages of the Wiki are public. The homepage with a link allowing people to join, and the meetings page with links to agendas and meetings. Off the AIIM website, links to the Wiki and promoting the Wiki for that purpose.

AB: Moved some items of existing content.

DJ: Added note to not add numbers to section headings

RH: Begin syntax. With MG's assistance entered items into the syntax portion. Implement breadcrumbs.

DJ: Email from Colleen. Seeking to reassign keyboard section. Change and mark complete.

GP: Fonts no, Actions yes.

DJ: Continuing to urge others to contribute.

JC: Netherlands re: font issues. Done as we speak.  Everson at Evertype.com.

DJ: Make note under organization and use of fonts we can refer to Unicode without fear of violating copyright issues.

DJ: Actually spoke to Max, who will be unable to contribute regarding this issue. JavaScript needs to be assisgned at least for outlining.

DJ: Finding people and getting more people to contribute is a standing task and a vital process.

DJ: Whittle down subjects that don't belong in the draft. This has been completed by LGR and MG.

DJ: Annual report is complete and will be loaded to the Wiki

Select a topic from recent additions to the Wiki and discuss

Review deletions

MG: What we attempted to do (MG and LGR). Looking at the document working on the grammar of the language, you do not change words or characters, but how you put the words together. Objects and items that make up the underlying foundations of PDF should not be changed.

 Items not marked as delete merit further discussion. Where parents are marked for deletion, the children should be considered included in that excision.

NS: Are their restrictions on required elements like objects that we should be aware of to keep them from being used in an inaccessible manner?

MG: Users never see this stuff, this is the stuff that makes up the underlying file format.

Objects are deleted.

Filters: Cypt filter left primarily for discussion. The reason it has been left open for discussion. XML meta data can be applied to a PDF 1.6 file and you can use crypt filters to do that.  Can be applied as encrypted or unencrypted and we need to discuss whether or not to allow encrypted meta data.

For Encryption, the parent is being preserved but the children are being deleted. The working group needs to determine whether or not to allow encryption in PDF/UA.

Professor Jacques Distler joined the call at 4:03 pm

DJ: We have a few questions, perhaps we can provide an overview of what this Working Group is up to.

We are tasked with developing a document that defines the terms under which a PDF file is optimally accessible. One of a number of concerns involves the correct manner in which Math can be displayed in a PDF file. Indeed Neil Soiffer from Design Science is also a member of the working group and on the call. We need your help in identifying those situations where it is appropriate to use MathML in PDF files.

NS: Good to speak to you Jacques. Joe had 3 issues that got raised. 1) Simplicity. At what level do you make the determination that it has gone beyond text and become Math.

JD: In HTML for instance elements for subscripts or superscripts. I wouldn't need to invoke MathML to do that. However, if I needed to do something slightly more complicated like indicate the square root, I would have to use MathML.

DJ: What would you consider the testable boundary?

JD: Should we have tagged a particular portion of text in some way?

JC: If you can use Unicode characters without a line break there's a case where you don't need MathML. We can create our own tags to indicate inline math for instance.

NS: It is definitely critical that we be able to mark things up as math.

JC: THe big thing is no line break.

JD: We typically distinquish between inline math and display math.

DJ: Could this be a good argument for a new formal tag?

NS: Some unique tag that indicates this is different from text. It is not a superscript or footnote, it needs to be translated into Math.

JD: Inline MathML, display Unicode characters and so on.... Whether or not you want to tag something inline as being explicitly math.

LGR: THere already exists a formula tag.

That might work for us.

NS: However, that was typically meant to identify an image, but there is no reason why we can't appropriate for our own purposes.

JD: Most Mathematical text appearing as PDF is generated from TeX. Brackets by single dollar signs are inline equations and double dollar signs equate to display equations. Typographically treated very differently whether single or double dollar.

TeX is a presentational language. Need a smart translator to produce useful MathML.

NS: Braille codes are all notational. No semantics. Not important for Braille, for speech it is important to have semantic information.

DJ: In terms of published math either display equiations or block equations.

JD: The same problems occur in TeX. A slash B in one case, in another A over B in an equation.

NS: We need a mark that tells it that it is Math. THe specification has to say it has to be tagged as math. THere could be separate ways, a simplified form and a MathML form. It has to be tagged, otherwise it will never be translated to Braille properly.

JD: There are things that TeX is bad at, for example prescripts. MathML has a facility for that, TeX fakes it.

JC: Proposition, this should be a part of the spec that we have the option of redefining.

NS: This is no different from most things. These are all moving targets.

DJ: So will this be a fixed reference or a floating reference.

JD: Raman's Astor renders TeX to speech.

DJ: What we must do now is summarize.

AI: NS: Could you summarize the discussion in a manner that we can make some use of this. Angy has also been updating the MathML session of the Wiki.

DJ: Collective thoughts will pile up on this page.

AI: JC: forward Jacques contact information to DJ so that he can be invited to contribute to the page.

Planning for the F2F meeting

DJ: Draft agenda page for the meeting and linked to from the minutes page. For the face to face I would prefer if others would contribute to developing the agenda. We need to complete the "whitling" of the draft specification. Please contribute to the agenda, I would like to hear your thoughts.

AB: This could be an opportunity to go through the different pages so far.

DJ: We won't be able to do a work up job on the pages during the meeting. We could have the task of surveying the pages and develop objectives for the committee. Make heading notes as appropriate, bring it up to the level where the pages can be sufficient to be an assignalble item.

LGR: Here's a proposal, we should go through all the sections that have not been marked for deletion and identify what the issues are that should be captured for those items.

DJ: We will run the Wiki live. Someone will run the Wiki live. There will be a projector and web access in the room.

AI: DJ check with Renee.

JC: Suggestion on Loretta's proposal, let's get something down for the known topics.

DJ: Building on Loretta's idea, we can build some content for each heading that we choose to keep in.  Make the Wiki more and more of an instrument that will attract interest.

RH: Let's continue to have assignable responsibility. There's power in that.

DJ: Our Action Items will have an additional column: Who, What and Which Page. Let's provide an easy way to get at what people are doing. We could go with the survey approach or go in depth.

AI: DJ come up with a bit more of a template for the page. Drop some notes on each page and make sure the entire table of contents is linked. That's what we're going to do on Wedneday. 3 hours 9 to noon.  Thursday is the continuation of Wednesday.

LGR: Where people have put in some work already we can also give them feed back.

DJ: That's a solid proposal. What do people think of that? Take 3 hours, pick 3 subjects, and try to work them up. We have a basic idea for Wednesday and a basic idea for Thursday. I would still welcome suggestions from people. I will finish the agenda, circulate it sooner rather than later.

AI: Everyone has to fill out their portion of the roster. Pick a heading and populate it with something for fhe next meeting. That will make Wednesday more efficient. Add a Wiki link to your chosen heading.

Validate PDF/UA - Action Items assignments specific to the conclusions developed

Wrap up and schedule next meeting

Philadelphia at the AIIM Conference, Pennsylvania Conference Center downtown Philadelphia and by conference call. May 17 and 18.

Adjourn: 5:05 pm
Johnson / Soiffer