W3C logo Web Accessibility Initiative (WAI) logo > EOWG Home > EOWG Minutes

EOWG Minutes 16 July 2003 Meeting

on this page: Building the Case for Web Accessibility - User Centered Design of WAI Site


meeting page: http://www.w3.org/WAI/EO/2003/07-eowg-agenda.html



JB: Introduce hosts HP (NL) with support from Wells Fargo (DS).

[introductions around the table]

JB: This is Education Outreach working group. We talk about and work through docs. WAI done through W3C. Four domains. Activities with chartered working groups or interest groups. Chartered to address certain types of work. EOWG awareness, training materials. Active in promotions and training. Exchange of info. Wish list – good idea but not now. Deliverables list. Resources Suites – several page resource.

JB: Agenda today – Two resource suites: Web Accessibility Training – old resources suite; Building a Case for Web Accessibility. Then user-centered design of WAI site. Worked through in past, stop for a while. A year ago, revisited the topic of user-centered design. Building a Case for Web Accessibility - I suggest we start with that.

JB: Agenda review for 17 July. Additional people coming --- 26 with two more cancellations. Training Demos.

HBj: Where in the agenda have you put Euro Accessibility?

PG: Give some slides on Euro Accessibility to you.

HBj: Discussed Wednesday and Thursday because not I am going to be here Friday.

JB: O.K. During lunch let's discuss this.

HBj: Thought these more applicable for today.

PG: Yes. I can show you the slides.

JB: About how long take to explain.

PG: About 5-6 minutes to show slides.

JB: Project that encompasses 24 organizations, include some technical assistance development on access and evaluation approach. Is there some interest here?

SLH: Not losing anyone, some more tomorrow.

JB: Let's discuss at lunch. How many people have an Ethernet port and want to be connected?

[Half for setting up Internet connections.]

Building the Case for Web Accessibility Resource Suite

JB: Go to http://www.w3.org/WAI/EO/Drafts/bcase/soc-new.html. Switch and display from off-line and walk us through it. Start working on some substance. Orientation to the suite. There are 5 pages in all with suite. The first is an overview page. Purpose is to look at a range of arguments for doing accessibility and be able to select from those rationales something to help them in there organization. Often use several rationale. May want to take a social responsibility position but have to take the strict monetary business case. We give a number of factors to look at. Trouble putting into a framework that works. Recently used these. William did a rewriting of Andrew's writing. Principle of tersification that William invented.

JB: changes:

JB: Hope that we can look at the orientation/attitude of this page to see if it feels right. Apologize that this revised draft was not available prior to the meeting. Let’s see if this material makes sense.

JB: Orientation section. That is new. It is the whole social factors case in one chunk. Provide why we are calling out social factors. Is this approach that different folks would understand in different circumstances? Tried to smooth out the language – be careful about claims we are making.

JB: Took out from earlier edits “older people suffering from a wider range of impairments”.

JB: One of the things we were saying was a problem was that we were not telling people how to use the suite. So we tell them how to use it.

JB: Sections ...Previous drafts a lot of detail. Too technical just have the explanation for...

WL: Nutshell be a [?]shell rather than a coconut shell

JB: O.K. Bullets and skip verbiage.

DS: Overall page: add “access to work”.

JB: Charmane’s strong point to have employees. I demoted it but maybe should have its own heading.

DS: Accessible wide range. Use IM (instant messaging).

JB: Separate out access to Web and access to employees.

LK: I like what has been said. The method that I see in this paragraph presents what is going to be coming. Social responsibility - importance to access to the Web. Putting out as if people understand it. Maybe we should put out why is this important. Normalization for people for people with disabilities. Using the concept of usability design.

JB: General reaction? Problem we always have in this working group. Our documents are part of a network. Train ourselves to say when we want to broaden a definition we look at where we have already done that and link to it. Bill did provide links to how people with disabilities use the Web. Link to it from here.

LK: Point to the definition. Otherwise, come to the document and haven't a relationship understanding.

JB: This is the only place we talk about social responsibility.

HBj: I like what Bill said in the first and second paragraph. “This document…” either have them in the same paragraph or change the verbiage. In general in social responsibility, you don’t know if social responsibility is the same in general around the world. In my country, we talk a lot about social responsibility but it is in relationship to employment, taking care of the old workers and the retirees. I don’t know if we will be able to make a case for social responsibility around work. Don’t think that is likely.

RL: Big thumbs up to the bullet points. The link to disability should go to a definition - What disability means.

JB: William had that link. I can add it.

WL: In all the work we do, there is a tinge of elitism. Express an audience that is talking from top down. Need to get the attitude that this is a document to use not just to read. There is not a group of people in white coats that will create this. Can accessible be cool? The attitude that pervaded - "Poor unfortunates" rather than major players

JB: Refilter for reading.

WL: Key elements if language is different, this facilitates that. Causing groups to talk to one another on line. Different than broadcast, a conversation. Allow people to talk to each other - where a blind person could talk to person who is deaf. That is not in our social responsibility section.

JB: Using the wrong concept for the whole section. In reframing be careful about creating a charitable frame. Just say social factors not social responsibility. Would it work? How does that fit into a business case?

SLH: A lot of people will have a limited focus. Note that this is not the only factor but it is one of the concepts.

SAZ: Feel audience is a bit missing. In social responsibility, are we talking about the corporate. Maybe we should try to bring out a bit more who the audience is of the Business case.

JB: It could be many: corporate, NGO (non-governmental organization), university...

SAZ: Why should I implement accessibility for employees, other end users, customers, businesses, and other audience that may need to access this site.

JB: It goes back to Doyle saying we should be splitting public that access the site and employees that have to access to the site.

WL: An organization can be one person, by the way.

MK: Steer away from social responsibility. Promoting interactivity not the passive use of things. How this will connect people so they can share and communicate. Industry thinks social responsibility already but, if we talk about additional people who can be part of their conversation or people in another organization.

JB: Been talking with organizations about parts that give them motivation to do Web accessibility. That is what drives it already. Have to position self. Already have Web access.

MK: Larger companies have funds to do this but smaller companies need to meet budget. Customizable kit part of what we put on this page or just part of it. Company said we don’t need. We understand that. Maybe we need to find a place to say both. Some believe universal design is not possible so they hear those words they shut down.

JB: We cannot make a list of every single item in one page.

NL: Use social responsibility lingo. Have a link to the social responsibility. Have a definition and link to that.We are trying to eliminate the digital divide then access to employees. i.e. connecting everyone to the power of technology and building a foundation for universal access to basic technology. Access for employees.

LK: Clarify where I was going with that include phrases like social responsibility not a problem but I do think the point that Mimi was making and others…maybe social responsibility… you don’t know who you are making this for so why not make it so everyone can use it? Making the case that there are people who access things in different ways, there is a way to do that.

JB: What is your main suggestion?

LK: Between demo social responsibility and then enumerating in different ways.

JB: Maybe a paragraph or two to do this. This is a dart board. Don’t take it as anything other than that.

DS: Agree that social responsibility will be a subset of this. Section 508 and large corporations, the ADA are not strictly about social responsibility, rather policy and law. Nice to have it in there but it helps to establish the larger framework.

MK: Universal access and put social responsibility as a subset of that. Have a concern that if you put social responsibility people won’t click on that. Some aversion to this. Federal law if they want to sell with the Government then the follow Section 508. Goes to the states. Do not think that most companies in this country think about social responsibility. They are just trying to make payroll.

JB: I talk with people from Fortune 500 companies that say social responsibility is what has got them over the hump. One thing might be interested in and then they look at the legal and tech. The are amazed at the federal entities. They have found that Section 508 is not enough but they have found that social responsibility helped move them forward. Let's take the people in queue and then look at one or two or at least the first section.

ML: Large corporation social responsibility… sponsorship establish metrics for the efficiency of human to human interaction. ROI (return on investment). Socially that is where we are lacking HR (human resources). Web stuff what is that? Web people building HR related things but there is no input coming. What kind of efficiencies can you have when your site is translatable. Have some efficiencies. Contact management system. Internally a lot of people.

JB: Talking about the technical factors are in the other parts of this resource suite.

ML: There is an efficiency that can be gained here, and here, and the social responsibility. How do we justify, retrain several.

JB: Are you saying this document is going to be helpful?

ML: The social part is not yet distilled enough. A cloudy vision won’t help us. Sharper vision will help us.

JB: Let’s keep trying. The translatability needs to go into the technical factors. make the distinction between internal vs. external - split the public and the employee.

HBj: This is not a US (United States) paper. This is international where the social factors are very important. Don’t build on the legal claims but we do have a strong case no matter the size of the organization. They have a responsibility to the community. The reason we came up with the social factors is because people said the economic factors were not strong enough. Therefore, what is there beside the bottom line? You have a social responsibility and the social factors with the community. We have forgotten about the environment. We should not damage the environment. And we have other social impacts on how we have to be able to use this in Europe, US and Australia.

WL: Song and Dance. If everyone put in bullets to the document, I will volunteer to read and comment on them. Most important. Societal as well as social. Social is just one aspect of responsibly – blame.

JB: Should we redraft, so factors more neutral?

WL: Yes.

PG: I will translate social factor. Would use more citizen factors: How Web accessibility can help people be full citizens, full civic participation? If you don’t do it, be the bad guy so they will do what we want. It is regretful that we have to be careful to say.

SLH: The whole document a tool that people we can pick and choose. The document is not intended to be given as the business case as is, but take the points needed and then flush it out for my organization. This will facilitate that. If they want social responsibility, they will pick it out.

JB: Summarize 5 things and leave a detailed change log that Shawn is doing.

  1. Good straw man for discussion about separating employee and the public.
  2. More emphasis on interaction and not treating Web as a passive, read-only medium.
  3. Describe social factors before getting [?]
  4. More conversation about social, societal factors, social responsibilities or what exactly.
  5. [?]

ML: Make it tangible

WL: Michael you now have an assignment– make it tangible


JB: Agenda - stay in this discussion and do the User Centered design to after lunch, Slide some items off the discussion. Do two things: Briefly look at the framework of the suite again, feedback on the intro to the social page.

WL: Move the Gallery into limbo land.

JB: Reactions? comments? Alright. so let’s do that then.

JB. One of the other things not done much is the overview page. Change from original framework, start talking about main subpages. Then would give an idea of how to combine these for a particular business case. Had material that gives idea of how to emphasize some sections. Mix and match. Orientation of how to build business case. These copies don't link properly. Still, can do updates to this copy of cover page. Economic factors, question and answer approach. ROI (return on investment) approach. Technical factors document. Worked over but needs more smoothing and editing.

JB. Back to social factors document, which we discussed earlier. We had made comments about how to recast introduction. Describes area by area. Access for people with different disabilities. Look at previous online version, lot of detail. If we like the detail we can bring some of it in to new draft. What is the level we want to aim at here? Level of detail in this section by section.

WL. Policy document doesn't reiterate all policy bits, just links. The bit about 10% of population can link. Framework with bullets and links to statistics. Instead of being in this. This is very different to others. Need to get them all consistent.

JB. We've already chosen very different formats.

WL. A way of looking at it a glance.

SAZ. Worried about contents; links to such detailed level of information. May raise transition costs. Like to see links to relevant information. May be better to include list of references at end.

WL. About potential audience, thing is that when user is looking for way to present business case, needs a portal to detailed information. For users who already know a lot about it. Where can I find detail?

SJ. Agree with you. In paragraph on people with disabilities; very detailed. Should be list of bullets.

JB. One thing is access to services and information (for public sites) but not why it's important. Access to Web through adoption of guidelines. So general, neutral. Some people may not need this orientation first.

MK. Not necessary.

JB. What links would you add?

WL. Link to guidelines

HBj. Take care not to make it too much hypertext; people may get lost in documents they don't understand.

WL. Use definition list.

JB. People may not understand the information.

RT. Where else is the statistics information?

JB. Extraordinarily difficult to do internationally. So we decided not to do page of demographics. Is presently unsubstantiated.

RT. Georgia tech has this information. Is very important here. We shouldn't leave it out.

JB. Getting many nods all around.

HBj. Do you mean link to the information?

RT. I wasn't meaning that specific, just the concept.

HBj. We don't want to put up statistics, because you see figures are not that high. In Denmark the blind say it's not just about the number of people. It also has wider benefits.

JB. I think Helle says we don't need to leave it out, but that it has wider implications.

PG. Thinking about how I can translate that into French. We say people with impairments, which means a big difference as it may be a temporary disability. Affects the scope. Not only the percentage of blind people, but many more people. Helps a lot in justifying accessibility.

JB. You're not saying miss out disability.

LK. We have done work on statistics. To make a point about these areas, you do need to emphasize statistics. Don't want to avoid this issue. Internal and external references. Should reference papers about it?

JB. We have decided not to include links to off-W3C sites in a page like this, but use a separate page that we can update more frequently. Thought of doing a demographics section that way.

LK. I'm not sure how strong the case is without statistics. If you have to present to a board of directors, for example.

DS. In Wells Fargo we don't actually track disability statistics. Is a problem. Statistics help correct problems. Research at MIT on attention deficit user interfaces. Without statistics it's hard to discuss seriously.

JB. We've discussed at length; statistics tracking. Difficult, so we decided not to include demographics. Adopt a minimalist approach here. Emphasize the idea behind it. Based on what I'm hearing, seems nobody's opposed to leaving it out.

JB. Social factors. General reaction I hear to first paragraph is that it doesn't work that well. For comparison here's what we had in March. Miss out "ubiquitous"; doesn't translate well. Try to find William's draft. This section has long bullets and uses Andrew's intro. Trying to do it shorter, but hearing good reactions about how it looks. Is not right approach. Not what people want. Not right yet but we know why.

JB. Employees. Do we need a table here? Audience, access, what might help?

HBj. Last time we tried a matrix it was difficult (auxiliary benefits).

JB. [Shows auxiliary benefits tables on projector.] Social responsibility included in table. Detail is above in page. Interesting way of organizing information. Some people liked it some didn't.

CC. I think trouble is the seven subheadings (social factors document, new draft). Formatting is the problem. There's a heading missing. Before additional resources. Would read better if there was one heading for employees.

JB. My impression when we were reviewing it together is that people don't grab it at all.

CC. Going all the way to a table seems excessive.

JB. First two paragraphs need to be split. Access for public and for employees.

CC. Access for people, and split it up with subheadings.

JB. Social factors heading is missing. So what do we say within each section?

CC. Do the social factors change with category? If not you can use a general overarching paragraph.

RT. You could write it to include all factors

CC. Social responsibility can apply to all and so it can be pulled out into a common paragraph.

JB. So one approach is umbrella statement to say ensure that site meets needs of all audiences. I started by assuming the umbrella statement. William is that new in your draft?

WL. No, I took it from Andrew's version.

LK. I think that in each of your categories you've got the initial statement, and the population groups, and then each audience's needs. Actually have some of that, access to government info, entertainment. Could be enough.

JB. Can we presume the needs of each group?

SLH. Second sentence goes up a level.

JB. Biggest task is to build the umbrella statement. Then something about why universal design benefits them.

CC. Explain why each group is included. As opposed to some other group. We're not saying this is all the groups.

JB. Say why each group is relevant.

HBj. Include employees.

SLH. See employees, but disability groups not in the list.

JB. Introduction is brutally shortened. Bullets. Social responsibility as a hook. ... Who is and how it relates to accessibility. Comments? Discussion?

HBj. Easier to continue discussion after someone has drafted the document the way you describe. Would be much easier.

ML. Outline we're talking about user groups and stakeholders. Distinction between user groups and what actual impairments may be or accessibility barriers. Distinction between internal and external audience. Specific impairments. If cognitive or physical, some are applicable and some aren't. Distill everything into a core message.

MK. In our project we identify groups and the need for them in the proposal.

ML. Groups with a disability. How does it translate into a proposal.

JB. Are you suggesting trying to create a segmented business case? It doesn't work, but I've heard it said. Unified business case tied to the guidelines.

AC. Is that explained in the document?

JB. Could be interesting.

ML. Does that help?

LK. Maybe I can clarify. I have list of categories. Older people can be a significant part of disabilities.

ML. Mixing social and economic influences.

JB. In papers on digital divide, it's frightening the way they ignore disability. Huge overlap with disability categories. Don't think we can arrive at clear-cut groups.

MK. Michael expresses things I was trying to describe. Economic effect of disability.

JB. People often not thinking about disability issues. Umbrella should say that disabilities interact.

JB. Glad we went into more depth on this. One thing I take away is to expand umbrella, expand categories and interactions, and see how it works out when we do individual sections.

CC. Will have a go at a section.

JB. When we come back, we should go right into user centered design. During afternoon break talk about training presentations tomorrow. Work on training resource suite later this afternoon.

[ lunch ]

User Centered Design (UCD) of WAI Site

JB. Two years ago we did a site redesign, but it wasn't systematic. Wanted customized access for different audiences. Last summer, July 29 and 30, we discussed in Toronto how to do a UCD approach. February in Boston, we discussed what to do, need for people to do different parts of the process.

SLH. Want to give overview of UCD, then specifically talk about project proposal for usability testing of WAI site. Documents from last March F2F. UCD also called Human Centered Design in Europe, and also customer-centered design. Multidisciplinary area. Many resources available. Lots of resources online. Process with very specific techniques. Ensure that the process includes audience, tasks, workflow. Iterative process. Meeting in March got interesting feedback. Have organization interested in doing usability testing of WAI site at no charge, and provide input to process. Any questions?

WL. Overall question from outreach point of view. Has accessibility community reached the usability community?

JB. Tangential issue.

SLH. Short answer: I have been giving presentations for a few years and reception is good.

SLH. Questions about the process?

SAZ. Any policy in W3C?

SLH. Very few.

JB. Site has grown out of contributions by individual members, with little overall coordination, except for few guidelines. Role of WAI EOWG is to lead work in this area. In March had participation by other groups.

PG. I used to work in a Web design company and know importance of usability. Are we going to create guidelines that combine usability and accessibility?

SLH. Couple of answers. Not in scope of WAI EOWG. Mimi has project that integrates. There was a group of people that wanted to start usability WG within W3C. We should keep in contact with the other initiatives.

PG. More and more questions in France from Web design companies asking about accessibility, want method to avoid wasting time, ask for checkpoints during design phase.

SLH. Not priority for this group but others are working on it.

MK. We are going to put materials on the Web for designers. Issue of integrating users with disabilities into testing procedures.

SLH. Two years ago did a workshop on it, I can give you the address.

WL. Are we talking about usability of all W3C materials?

JB. No. This discussion is only about the WAI site.

WL. One way to start is to create personas.

SLH. One problem with doing full-blown UCD, including personas, is the time. I used to do UCD before starting in accessibility, but now there's not enough time, either mine or this group's. Need to minimize time and maximise benefits.

ML. Who heads this effort? You [to SLH]?

SLH. Yes.

ML. What is our scope?

SLH. This group has input into the process itself.

ML. Are we looking for requirements? This is my speciality area.

SLH. In March we talked about setting up a task force, to devise proposals for the group to review.

LK. Have we had comments about why the site is not usable?

SLH. Very few formally, the link to submit comments is at the bottom of the page. There is an archive, but it's very small.

JB. There are many more comments than go to the archive. We've tried a number of experiments for the design over the years.

SLH. There are many people waiting to try. But we need to be more formal, systematic. Create usability tests to help identify and quantify the problems and strengths of the current design, and provide benchmarking of the site. The people volunteering to do the work will do the usability testing quite formally in facilities equipped for the testing. Eight participants. Not enough to allow any quantitatve statistical analysis. In last teleconference there were lots of ideas. Let people in other countries conduct the tests, for example. Lots of wonderful ideas, and we hope to do some of them, but we have to temper the scope because of the resources available. Are there any questions about usability testing in general?

CC. What is the normal number of people for testing?

RT. Some people say that 6 to 8 people is good, but more than that and you get diminishing returns.

SLH. What we have to do today is to talk about that. Some of the things we have to put in the pre-test questionnaire? What tasks do we want the participants to complete? For example... find resource on site. Host test questionnaires, about their site in general. Determine the characteristics of the participants. Who do we want these participants to be? Particpant screener. First topic is: we've got eight slots, who do we want in each?

ML. First we need a Web development person. Within that, someone on the business side, and someone who does coding.

RT. Someone with different disabilities.

DS. Testing tools. JAWS, some screen reader.

SLH. In the test, state what tools they use.

SAZ. Different countries and different languages.

SLH. The company that will do the testing is in Massachussetts. We have people in the UK and Australia interested in this. Save time and energy to be able to do tests in other languages and countries.

MK. Important, but as this company is donating their resources, perhaps a recent immigrant or visitor to this country could satisfy the requirement, or does it have to be someone not related to US culture.

SLH. How much effort do we need to put in to ensure that some test participants are not American or americanised?

SAZ. I have found that in Europe, Web sites tend to be organised differently. If it is too American, it would be a problem.

MK. But if we use someone who is native to other country but lives in US, could be OK.

PG. Vocabulary: when I first saw "authoring tool" or even "education and outreach" it was confusing.

LK. People in other Web development areas. Someone who is responsible for Web accessibility

CC. Either a secretary or executive staff assistant, as they are often called to find out about things for boss. On my campus, for example.

ML. The representative for accessibility in a company, who was initially in charge of Web for. People we have to think who we are designing the site for.

SLH. Absolutely. The site was not developed following formal UCD. All deliverables are done in response to needs of users. List of deliverables is updated every quarter. There was a list but I can't find the link.

SAZ. Management, decision makers. Not developers, but who have some ownership of the site.

PG. We train public site developers who have to justify it to their bosses.

WL. Professional usability specialist.

JT. Beginner in accessibility.

HB. In government sector, people who are setting up contracts with design companies. So they know what to ask about.

SLH. A range of people, with a range of skills with authoring tools. Who we want and the task, we have to ensure that the people can't find the info on the Web site. Choose people and tasks that we think are important.

HB. Different countries have different criteria. More international tasks. For example, find a country's legal requirements.

WL. A rehabilitation counsellor.

LK. A teacher.

PG. A student? They are young and learn about disabilities.

CC. Not sure this was covered in the training user.

WL. A disability rights advocate.

LK. A legislator.

SLH. A press reporter.

CC. Or in public relations.

SLH. We have to look at prioritising and overlap. One of my questions is people with disabilities. Different profiles. Would all the users have disabilities? Material for people with disabilities, advocacy, etc. What are our goals in terms of requirements and range of disabilities. Let's be more specific about the disabilities represented in the eight users.

DS. Visual impairments, motor disabilities. Test: that a visually impaired person can understand things fairly easily and not get lost. That parts of site like tables are OK.

RT. Recoverability if people get lost.

CC. Since we are testing usability and not accessibility, we should perhaps minimize the number of disabilities represented. Agree with DS that visual is important, though. Do accessibility testing later in development.

SLH. There is a difference between usable accessibility and technical accessibility.

HB. What about reading disabilities? Low literacy and dyslexia.

RT. Error recoverability is important.

NL. Are we taking about the old site or the new?

SLH. The old site.

NL. The idea is to give the disabled user the same experience as the others, and until you test you can ever tell.

HB. We should include a braille user.

CC. Someone with braille has a different experience.

DS. A friend of mine has cerebral palsy and is blind.

SLH. Types of organisations

ML. Different requirements in different countries.

SLH. What do you think about the priorities of different organisations.

RT. How does that affect priorities?

HB. Educational institutions, distance leaning, not just government.

DS. A company with 100000+ employees. Different experience.

SLH. How does organization affect user test? How important is that in terms of usability testing? If we have users from different types of organisations are those people going to interact with the site differently?

DS. In my large company we have a vast reservoir of knowledge.

SAZ. I think the expertise is important. Managers look for one type of information, developers for other, not only because of the size.

DS. Disagree. For example in financial industry, in big company even a financial manager will be aware of issues, but in a small company no.

SLH. Depends on experience of user in company. Everything we want must be included in screener.

NL. Talking primarily about navigation?

SLH. Yes - navigation, design, layout, information architecture. Not content. Old site.

WL. If we decide that this is about usability. We don't necessarily need specialist users. Unless dealing with content, not just structure.

RT. Dealing with experience.

SLH. To summarise last discussion: Important is variety of experience levels, not of organisations.

HB. Some of the organisation, would be very similar, whether they come from central government or local government. Should include private enterprises and disability organisations. People in disability organisations have different viewpoint.

CC. If we include those four categories we would cover everything.

NL. We need to talk about the navigation and not the content. If we talk about roles, we are talking about content.

CC. Roles lead to question "What will I be doing on the site?"

HB. Structure is very closely related to this. If we look at navigation and structure, we need to look at different backgrounds, and what is this site presenting. Structure very important.

SLH. Leads us nicely into talking about tasks. If we talk more about tasks, it will help us define the rest. I sent an email with the tasks. [shows on screen and explains]

HB. Assuming that there is a voice recorder?

SLH. Will have a live facilitator, giving instructions.

DS. Like to see task to clarify difference in time for sighted and non-sighted.

SLH. With only eight users, not enough data. Dangerous to provide data with this number of users.

DS. Have complaints from blind users about time taken.

SLH. OK. Have seen data used negatively, so be careful. Any specific task?

DS. To find an embedded paragraph.

SLH. Deep within document?

DS. A linearised table. A visual user would have it easier, but a blind one finds it difficult.

LK. Different people, different levels of experience. Tasks. Some advanced, some novices. Find out what tasks each does, and base task on user's relationship to the site.

SLH. Hope to do the tests before, during and after, and show any improvement, so need structure. Will talk to people doing tests and ask their ideas.

JT. Blind user with screen reader. Excessive content makes it difficult to use.

SLH. Devise test to take that into account. Pages have breadcrumb trails. Test understanding of navigation mechanisms.

NL. Eight people are not enough. With heuristic studies, can be enough. If not going to publish results.

SLH. Might publish them.

NL. With free exploration at the end.

DS. Might be nice to ask some sighted users to use a screen reader.

SLH. Not in scope.

CC. Do tests have to be replicatable for new site.

SLH. Should generally be replicatable.

CC. So we need to decide what the objectives will be for the new design. To have comparison. So if we focus on existing design we need to plan now.

SAZ. Support what LK said that we need to match the tasks to the users.

HB. Navigation

WL. Is there any literature to convince me that this is science and not superstition?

RT. From my experience. Make the tasks very specific.

SLH. Yes. Group will review the tasks.

CC. We will pilot the tasks ourselves.

SLH. Any specific questions on defining the tests? In-page links. Specific questions about whether a design works?

DS. Leaving a page.

SLH. Anyone heard of Bunnyfoot Universality? Use eyetracking and work with people who are dyslexic. Would be interesting to try.

RT. Task that uses the right side of the page.

VC. So much on first page, that don't know where to go.

SLH. Impressions of home page.

LK. Ask whether the user has a particular part of the page to focus on.

SLH. We know there are problems with the site. Redundancy. We might ask someone where to find a list of resources.

SAZ. Do we want to find out which resources are more important?

SLH. I meant that we could find out which link the user clicks on.

SAZ. Wouldn't be better to access log statistics?

JB. We have access to good statistics, but not where people go.

SAZ. CC said we need replicatable tasks. May be other better ways.

SLH. It's on the to do list, and if anyone's especially interested...

LK. Ask people what they expect to see on the home page.

MK. Splash pages not so popular now.

DS. Looking at expander icon

ML. Find out what each user group is looking for.

SLH. User task role matrix. Better to do that after the UCD process.

ML. It's a result of the testing. Need to find out what users are trying to achieve.

SLH. Example of a specific task?

ML. Go to site and find a check list that you can use.

PG. What does the user ask the search engine to find the page?

SLH. We need to think why the person came to the site.

PG. It's important the way the search engine presents the list.

NL. Using meta tags.

SLH. Is a separate issue.

DS. Idea of HB's. Invisible link to skip navigation bar. Find out how good the ALT text is.

SLH. Wrapping up. Will have smaller group to focus on this, then full group can review. Pre-test and post-test questions to review soon.

Last updated $Date: 2003/07/31 21:46:20 $ by Shawn Henry <shawn @w3.org>