Will that be paper or PDF?

I am evaluating accessible-document work flows for my school and several others. For the most part, this means documents in the PDF file format. Like so many aspects to this topic, none of the solutions are easy and/or cut and dried. It would seem that applying an existing process for web accessibility would do the job, but this isn't so.
The web work flow is a poor fit because of the disparate skill sets between the document creators. Most colleges have professional web designers, especially for their main pages, but the skill of those creating PDF documents vary greatly across the campus. For every skilled designer using inDesign, you have dozens of instructors creating quick, unstructured and inaccessible PDF documents with MS Word. You certainly can achieve a level of accessibility with MS Word, but most of the instructors on my campus don't take the necessary extra steps.
One interesting response I've heard on several campuses is just a flat out dismissal of PDF as the wrong tool for the job. It is hard to find evidence, however, that colleges are decreasing their dependence on PDF documents. Duff Johnson (a person worth following if this topic interests you) wrote Why PDF? outlining reasons for PDF's entrenchment in our electronic-document systems. He makes very good points. Don't bet on PDF disappearing anytime soon, in spite of what some experts are saying.
It's doubtful that one work flow is going to work on most campuses for PDF production. Certainly professional designers need to add accessibility to their skill sets so that what they create for their university works for all users. Faculty, on the other hand, need to create short documents and frequently do so quickly.

A Work Flow for Faculty

What then is a good work flow for faculty? As someone who was an instructor thirty years ago and remembers creating mimeographed syllabi, handouts, tests etc., I understand the notion that the faculty creates course material from start to finish. I think it may be time to reconsider this idea, particularly in the electronic age. A PDF document is so much more versatile than its paper counterpart. It can be searchable, which is real benefit for a long handout. Carefully made it can be accessible to people with a wide variety of print disabilities.
Faculty have become accustomed to creating these course materials in a word processor, usually MS Word, and then hitting "save to pdf." That's quick, but the resulting PDF is unstructured and certainly not very accessible. Ironically, making an accessible syllabus PDF from MS Word is not that difficult. Unfortunately, compared to typing it out quickly and hitting "save to pdf" it seems more complicated. I recently asked a departmental communications director, "Are you after wide-spread behavior-modification of your faculty or do you want accessible syllabi and handouts?" The response was a reflective pause.
So what am I proposing we consider? What if faculty could "knock out" a syllabus or handout in Word and then a departmental staffer creates the accessible PDF? Remember, I said that going from MS Word to an accessible PDF is not that difficult. Adding staff usually raises an eyebrow or two, but in this case one or two well-trained work-study students should be able to handle it. What about more problematic elements such as tables? They're not extremely problematic it is just that they cannot always be rendered accessible in Word alone and sometimes require some additional work in a tool like Acrobat Pro.

The computer gave us a quick and simple way to create paper documents. Electronic documents are vastly different from paper. Paper is rigidly locked into its presentation format. Electronic documents are fluid and dynamic. They can be "read" in various ways and soon can be translated into other languages.

The problem is we continue to use a "paper" work flow for an electronic document.

Planning Enterprise Level Web Accessibility

I was invited to attend a meeting between a communications committee and the new IT director at a college. Among the new director’s responsibilities was the college website and accessibility was on the agenda. What I heard were fairly vague reassurances that the website would be accessible, but really no substantive plans to make that happen.

After the meeting I sent the chairperson a brief list of the sorts of things one should be hearing when accessibility is going to be integrated into the web production process. I thought I’d share those here. Bear in mind that this not the only approach and someone else could come up with different list. But these are the main building blocks to publishing an accessible enterprise level website.

Accessibility Guidelines
The foundation to accessible web delivery is identifying the guidelines you intend to comply with. The government’s 508 and W3C’s Web Accessibility Content guidelines are cited most frequently. 508 is concise, while W3C tries to be complete and overarching. W3C recently updated their guidelines and 508 is in the final stages of revision as well. “Being accessible” is rather vague, but complying with a set of published standard should be concrete and verifiable.

Implementation
How do you intend to ensure the guidelines are followed in your web production? This goes to how pages are created. Will a professional developer or development team create maintain the site or will a content management system (CMS) allow for all level of employees to contribute content? If you use a CMS, then the more accessibility the CMS can automate the less testing and verification may be required later. Automation is attractive, but it still may require thoughtful response. A CMS may prompt for an alt attribute, but what your employee creates for that attribute may require some careful consideration.

Any web elements routinely used on your site and not handled by the CMS may require either training for the staff or outsourcing. For example, a complex data table can be completely accessible, but coding it is not trivial and difficult to automate. Either staff will need to know how to create these or the production can be outsourced.

Verification
How do you test for compliance of either your live site or a pre-live test site? This is where independent validators are attractive. See: http://www.w3.org/WAI/ER/tools/complete.html. Validators have their supporters and their detractors. I tend to stay out of that debate. But no matter what, you need an articulated process for ensuring that your accessibility guidelines are making it into you web site. Even a process in which random pages added within the past two months are spot checked for accessibility is better than no process at all.

Maintenance and Complaint Resolution
Maintenance is either checking new pages for compliance or for evaluating the accessibility of new “widgets” that heretofore have not been used on the site. I acknowledge that “complaint resolution” comes across heavy handed, but the reality is this is an area that could easily turn into a legal issue. If we are informed that a person feels that they cannot use one of our websites on the basis of their disability that can become a serious issue. Simply put, articulate the process of maintaining accessibility, testing new web “mini apps” and assessing end-user complaints about accessibility.

The Take Away
The process of ensuring accessibility of enterprise level web sites is a deliberate and involved process. The process should be clearly articulated from: standards to development to verification and finally deployment and maintenance. Anything less, will more than likely result in a site that is unfriendly and not useful to people using assistive technologies.

AT and the Evolving DS Service Model

Like many Disability Services offices my department is examining how it does what it does. This is not new and many offices have been doing this for several years now. The basic thrust of these evaluations is to move from a medical model definition of disability to a social model definition. Without getting too into this (and it is a field of study) the medical model focuses on the individual and treats disability as an impairment, while the social model attributes “impairments” to an inadequately designed environment. The medical model leaves “prescriptive” decisions to “experts” rarely consulting the person with a disability; while the social model puts the opinion of the person with a disability first and foremost. As I said, this is a field of study and my two sentences hardly do it justice, if you want to learn more, then web search “disability medical model” and “disability social model.”

What this means to most DS offices is a move from “clearing paths” for individual students with disabilities to promoting a campus environment in which most paths have had the barriers removed. For example, rather than hunt down a classroom for a student in a wheelchair, invest time in ensuring all classrooms are wheelchair accessible. If this sounds like something you are not going to accomplish overnight, you’re right. And if it sounds like you’re putting your campus on the path toward Universal Design, you’re right again.

There are a number of reasons for doing this. Not the least of which is higher education’s commitment to inclusion and diversity. Read through your school’s mission regarding inclusion and how can you not want to revamp how students with disabilities are treated? There are, however, other reasons for considering this change. Not only are the service models built on social and cultural assumptions that are being challenged, and the older models are not really sustainable. The old model cannot handle the increasing numbers of people and the broadening definitions of disability.

Of course a change like this does not occur on a campus easily or without the support of other departments. And we have found that there are quarters on our campus that are quite comfortable with the older definitions and service models. To be fair, there are also departments who ask, “What took you so long?” Just remember this is a transitional process and incremental gains are good gains.

What We Can Learn from AT

Now having set the stage, let me discuss AT’s place in all this. At my school AT has already started down this path and the reasons are rather pedantic. Fifteen years ago the typical personal computers in student labs ran a handful of programs and replicating them in an AT lab was not too difficult. Now, however, the typical computer is running dozens of programs in general student labs and the departmental labs run highly specialized software dedicated to an academic field. There is simply no way an AT lab could keep up with all this diverse technology. And even if you could technically, you would not have the lab assistants that could help students to the depth that a specialist in a departmental computing lab could.

Sustainability has forced AT out of our AT lab and onto computers all over campus. Basically I have had to learn to partner with the IT people on campus. And my campus is highly decentralize which means I’ve had to deal with various personalities and attitudes over the years. Add to that the systems change efforts that I and other AT coordinators have had to promote around accessible web and electronic documents and you get a fairly good dress rehearsal for what many DS offices themselves about to embark on.

Another preview of things to come is the advocacy of accessible web and electronic documents. AT coordinators find themselves trying to convince campus communicators to produce accessible electronic publications, but it is frequently a steep up-hill battle.

One problem with the new model (as far as “others” are concerned) is that it shifts the work and responsibility from one office (the DS office) and distributes it among the campus community. I have actually had web designers suggest that I should retrofit accessibility on their campus sites rather than them build it in to begin with! That would be hilarious if it if it hadn’t been a sincere suggestion.

Universal Design includes, but also extends well beyond, technology. The deployment of AT into mainstream computing labs does serve, however, as a great map for at least one method of moving toward access or all.

Universal Design and Access is the obvious direction for human interface development. It is inevitable, but DS programs can quicken the pace of adoption. Colleges and universities are immense organizations, but they’re usually more flexible than business organizations in implementing new ideas. DS offices are great motivators for the campus implementation of UD and the AT area may have some insight into dealing with some of the essential stakeholders on campus.

There are two take-aways from this. One, is that your AT people may have some useful input and advice as your office embarks on modernizing “disability services’ on our campus. And two, is that the old rationales of legal consequences, higher moral obligations etc. just aren’t as persuasive as they once were. Spreading Universal Access is going to take more sophisticated organizational change methods.

---

WebAIM Training
I'll be attending WebAIM's accessibility training on February 17 and 18. I plan to "twitter" from the training. Follow me at jbailey on Twitter during the training.