Web and Software: October 31

Miscellaneous

• Minutes approved from October 24, 2007.
• New draft available
• Judi will not be here next Wednesday due to W3C technical plenary.
• No other W3C attendees, so we can go ahead and meet next week.

Economic impact assessment discussion

• Draft added on each provision
• Start with 3-V Accessibility Services
  • Any thoughts?
  • Peter Korn agrees this will have a large impact and mentioned there may also be an issue with the text of this provision. In 3-U you can use accessibility services of the platform, but we do not have the same here. Action item: Peter will work up a draft. The intermediate layer should have the same services as any other platforms has.
  • These issues should be raised at the committee level.
  • This will apply to some of the newer platforms such as mobile platforms. So we are talking about new things that came out since last time that didn't provide the services.
• 3-W Multiple Ways (New Provision)
  • Any thoughts?
  • Gregg said that benefits would be to users that have difficulty using particular methods. Agencies benefit in a positive usability increase. Significant benefit.
  • Peter Korn agrees it's a notable benefit. Most sights have search so it may not result in a large change.
  • Andrew stated that a lot of sites pass with a search function, but what about Ajax and Flash where there is content that does not exist because its gathered from another resource... what other ways of accomplishing this other then a search feature?
  • Those would qualify for the exception.
  • No other comments.
• 3-X Labels or Instructions (New Provision)
  • Peter Korn described a Dialog box that has required fields and the typical asterisk that the field is required and a label that says required. How do we convey that to a screen reader user, and what is the scenario this is trying to address?
  • Andrew's take on the scenario is it would be taken care of under 3-O. Not the label, but the crucial information to help the user understand the action required.
  • No other thoughts on the economic impact.
• 3-Y On Focus (New Provision)
  • Any thoughts?
  • Gregg felt this was minimal.
  • Katie also stated minimal.
  • Peter Korn is happy with minimal, but concerned about test tools, that it may be difficult to test for this. Any thoughts on how this can be tested for? Since this came from Web and there are a lot of Web tools that may have a difficult time operating.
  • WCAG has test techniques, but we haven't talked a lot about automated tests.
  • Peter Wallack said that most tools today will fail when they see Java script and it could be significant on the tool since they will have to look at the code.
  • Gregg stated that this will not be that big of an issue since most will be adapted to WCAG 2 anyway. This will make it easier on the test tools for harmonization. It should be a net savings...gave example of Java script and Web pages needing to be done twice because of the old restrictions.
  • Andi is not sure if the impact is minimal or significant. Do we want to go with significant?
  • Gregg stated that seems funny to talk about these provisions as individuals. We know its massively expensive when we talk about them all.
  • We were asked to give an economic assessment of each one as individuals.
  • Judy also stated that we need to decide on some of these collectively. Authoring tools and evaluation tools are in a lot of these but doing it once would take care of it. Maybe a blanket statement in a higher level in the document and a reference comment.
  • If there is a block of provisions that go together and you can say something Gregg felt its better then giving lots of bitty pieces.. Overall they are interested in the economic impact of the whole thing. The summary is what they are looking for.
  • Andi stated that the TEITAC has to do the summary at the end. Suggested stating that the impact is minimal, but at the current time the testing is manual and can be significant.
  • Gregg thinks the tools will be there long before this comes out. We should talk about what we think the steady state is.
  • Peter is in favor of giving them the descriptive information Andi outlined.
  • No objections
  • Katie felt it was important to put Greg's point in about time frame on tools creation.
• 3-Z On Input (New Provision):
  • Any thoughts on this one... minimal?
  • Gregg concurs, this is standard practice except where we have Web applications and it would just be carried on through. Fairly straight forward. He is not sure we are talking about the cost to do these things or that he understands the testing cost is involved. Are we just talking about legacy testing? He thinks this will be applied and that this is true of pages going forward.
  • When you develop something you have to test it and its usually not done by the same person. There is a test technique and a manual process to test. On new stuff not just legacy.
  • The first time around the standards were for anything from this point on it was originally the intent and then they went backward and applied the Web components to things that had been previously developed. Used the excuse at times that something had changed.
  • Judy agrees that testing needs to be considered but she would also like to include the blanket statement that says it would be a higher effort until the tools are available and they are expected to be available by the time the provision comes out. If there is a particular note that needs to be added on one then we add it.
  • Andi will look at that further. She is not sure and will need to see how that plays out.
  • Michele stated that any input from the editorial working group, the original game plan was individually, but we can group them and then cross reference.
  • For sake of consistency, can we go with significant similar to 3-Y?
  • Greg suggested we should say the cost for implementation is low the cost for testing may be significant so we show what the issue is which is what Andi proposed in 3-Y and she'll do that on this one too.
  • Gregg said that sounds good.
  • Judy is looking for clarification as she thinks about testing more. 508 requires that the government is accessible to certain standards. The obligation is on us as a committee that this is testable. Looking for guidance from the government people... picture many more agencies going through the motion of implementing but not testing. Would like clarification on what the 508 obligations are.
  • Peter mentioned that we haven't really talked about economic impact for testing in prior provisions. He brought it up because it seemed to be a challenge here and qualitatively different.
  • The onus currently is the government has to say to a vendor what sections of 508 they are expected to meet. Then the vendor has to produce this thing. Some agencies test themselves, some have the vendor do it... there is a number of different ways. In general the government does not want to be running test labs and expects the vendor to provide testing.
  • Judy said that complicates it a bit, her sense of what might be useful for the Access Board would be to mention the impact separately.
  • Katie stated the testing implementation is on the government. They want to know the economic impact on the government.
  • Andi mentioned that ultimately on the government, but it does play out in different ways. Products for a market, the cost goes out to everyone as the cost of producing the software.
  • The economic impact have not been made only on the government so far.
  • Andi asked if what we agreed to for Y and Z, can we go with that and move on?
  • Katie is concerned that we need to measure these all in the same hand. That has been her understanding (government impact) all along and is not sure that is the case now.
  • Andi asked if we are objecting to saying this is a significant impact.
  • Katie says yes
  • Is that because tools will be available?
  • Katie said yes, because it will not come out until 2009 or 10.
  • We are at an impasse here because its important to say what the current state is. We think this is a significant impact with current tool and we do think tools will be available... the issue should be mitigated.
  • Michele suggested to add that the sub-committee could not come to an agreement and what some people feel regarding significant and minimal.
  • Judy suggested this be dealt with as a blanket statement... and refer to it. Has that been tossed out?
  • Andi said no, but we do not have language for that. Potentially yes, we can make the blanket statement, but we need to capture the thoughts and concerns that the people in the group have had.
  • Peter Wallack doesn't think there will be a testing tool that can solve this and that is why some of us think its a significant impact. Its an impact on everything not just testing. It also involves the creation of the product.
  • Andi said that was a good point, it was felt that it was minimal from an implementation stand point.
  • Gregg felt that if we assume people did not design good this would have an impact. Not sure that is an assumption we should start with. That the government is buying something that is a bad design.
  • Peter Wallack said there is this hope that tools will make it automatic, but this is a provision where the tool can not make it automatic. Today, potentially ever entry can change the page.
  • Andi will document that some feel its minimal, some significant, and document the concerns.
  • Gregg agrees.... also mentioned that much of what Peter Wallack is saying be taken into account.
• 3-AA Error Identification (New Provision):
  • Any thougths?
  • Katie said minimal
  • Andi agrees for the most part, everyone ok with minimal?
  • Gregg thinks so too.
• Authoring Tools - 8.2-A Accessible output:
  • Any thoughts?
  • Andrew said this is going to be a significant impact.
  • Andi mentioned that the benefit is huge.
  • Judy stated that this is one more place we need a blanket discussion because it will effect each provision related to authoring tools. The burden of acquiring a new tool that does this, but the benefit is it should make a lot of things easier. It seems that each have a different degree of benefit once new software is acquired.
  • Peter Wallack said this goes back to the economic assessment on the agency. For this one the impact would be low. On the vendor the impact is high.
  • Gregg stated that net cost to an agency is a positive savings, the cost of the software is nothing compared to what they will save in time.
  • Andi asked if we can say... going back to a blanket... the cost to a vendor to provide it is going to be significant, but the benefit to the agency is going to be large. Thinking this is a change of the economic impact. We can say this will be a cost savings...
  • Andrew to Peter Wallacks point, agrees the greatest cost is on the vendor and wants to be sure this is part of the picture we are creating here. Does not want to suggest this is minimal on the agencies. There will be additional burden on the agencies that these tools meet the testing. Agrees a large positive impact but does not want to be too naive that this is a magical bullet without additional resources on the agency.
  • Peter Korn does not thinks that most of the economic impacts were written based on the impact to the agency. Its important we clarify when we talk about agency verses industry. He thinks tools are one of the best ways to improve accessibility and believes the impact is large and positive and more then off sets, verses acquiring using and training of the tool. Thinks the next version on many authoring tools will have the requirements in them.
  • Judy stated the primary impact is the government impact. Assuming that comes back yes from the Access Board, she agrees we should go back and look at the other provisions and thinks it is important to note the long term cost savings. If people are looking at authoring tools as she hopes it may cause movement from product to product that provide more accessibility.
  • Agencies are allowed to use the VPAT as verification... and are encouraged to use it. The agency makes the statement and if it doesn't meet it, it is a cost to the vendor. The vendor that comes up with the tool will get the sale and improve their business. Significant cost to vendor is relative because it will also increase sales.
  • Peter Korn stated the question is of our interpretation of economic impact, we should review the assumption of the net positive impact on the government and the population at large.
  • Alex said that no one is going to sell more. In the end the total revenue is the same but the cost goes up.
  • Andi said that for authoring tools we want a blanket statement of the economic impact that it is a significant cost to vendors and stress the benefit and cost savings to the agencies. Then she also heard that we need to go back and look at the economic assessments if they were for industry and government. She will take an Action item: to look at everything we have done and go back and discuss that next week.
  • Peter Korn Agrees.
  • Two things to consider when doing the cost, we should talk about harmonization savings and about net cost. In the end it is what the net impact is of these rules.
  • Andi agrees that harmonization is an important point.

Plans for Additional Backup Material

• • Color, 3-B, and 3-C we can refer to WCAG.
  • 3D we may need to do some work there, as well as 3E based on the old 508 provision.
  • 3-F, 3-G, and 3-H refer to WCAG.
  • 3-Q we'll need to provide additional information on this one.
  • 3-R and 3-S refer to WCAG and 3S. We have also have additional notes WCAG doesn't have.
  • 3-R WCAG has a 20 hour condition provision. We harmonized with the May draft and haven't looked at the new ones. Recommendation to Access Board is where we harmonize with WCAG they keep it harmonized.
  • 3-T still information we can leverage.
  • 3-U we need to do significant work on, any volunteers? This is AT interoperability. We need people to collect the back-up information that goes with this.
  • Action item:Peter Korn would volunteered to provide techniques, and explanatory etc.
  • Information is needed by early December. There is an additional wiki page with the kind of information we want to collect and then it will need to be put together as one document. This is supplemental material to the standards that is modeled after WCAG's sufficient techniques. Has the editorial work group talked about a delivery mechanism?
  • Michele stated, they haven't talked about it, but she will take it up in there meeting on Friday.
  • Action item: 3-V, Peter Korn volunteered to do this one too.
  • Judi thinks we can leverage ATAG information for some of these. There could be a lot of good information, but it needs to be plugged in with care. Harmonization of intent, but not to specific provisions.
  • Action item: Judi will work on Authoring tools.
  • Peter Wallack had a reminder that all the text in WCAG may refer to Web page and we need to be looking at all software. Hopefully we can cover that with a blanket statement. Where you see Web page come up with something broader. Action item: Peter Wallack agreed to come up with a generic statement.

New Action Items

1. Peter Korn will work up a draft for 3-V Accessibility Services.
2. Andi will go back and look at everything we have done for economic impact related to government and industry to discuss at next weeks meeting.
3. Peter Korn volunteered to provide techniques and explanatory information for 3-U and 3-V additional back-up material.
4. Judi will work on Authoring tools additional back-up material.
5. Peter Wallack agreed to come up with a generic statement because WCAG may refer to Web page and we need to be looking at all software.

Attendees:

1. Tom Brett
2. Katie Haritos-Shea
3. sean hayes (Microsoft)
4. Sharon Snider - IBM
5. Peter Korn, Sun Microsystems, Inc.
6. Eric Damery, Freedom Scientific
7. Andrew Kirkpatrick (Adobe)
8. Michele Budris, Sun Microsystems
9. Peter Wallack (Oracle)
10. Andi Snow-Weaver - IBM
11. Samuel Choy - IBM
12. Alex Li (SAP)
13. Jim Elees (Access Board)
14. GreggVan
15. Angela Hooker
16. Laura Ruby - Microsoft
17. Amy Chen (Oracle)
18. ken kipnes (Oracle)
19. Judy Brewer (W3C/WAI)