I don’t disagree. My point is that if the mangled code causes accessibility issues (such as with data tables), then this will already be a failure somewhere else in WCAG 2.0. For every type of error that does affect accessibility, there are 100 that have no impact at all on accessibility. Let’s keep WCAG focused on accessibility, not on coding best practices.

Best regards,

Bill Tessore

It is critical that certification is intended for professionals from all sectors. Examples would include those from industry, government, commerce, education,
NGO groups, and health care. If certification were to be customized to any small segment, it would lose the importance of broad applicability. Any employer
should be able to turn to this certification as proof of skill level, and not skill level customized for the needs of a narrow constituency.

While most would agree with this premise, it comes into play as other issues are addressed, for example, when talking about funding a system of professional
certification. While those in industry may have membership fees paid by their employer, such a practice is not allowable in education and most NGO’s. If
the membership fee is a couple of hundred dollars a year this may be a non-issue, however, what could be the effect if it were $600 a year or more? I would
predict that over time this organization would contain greater numbers of members from industry. If this were to happen what would be the impact to cross-sector
applicability over time as the membership demands greater customization to their unique interests?

Response:
I think you need to consider building this step-by-step. Boiling the ocean, and requiring a single certification to cover every area where accessibility can come up might be more than nearly anyone is ready to take on, or even actually perform. Focused certifications can be packaged in to hierarchies, and allow for building a certification set by demand and preference.

You wrote:
Issue number two

The certification that results from this effort must be based on an individual’s performance and not on attendance or completion of any training or education
experience. While most would agree with this, we must consider plans for how performance-based appraisal would occur. For example, where do these activities
take place? It is unlikely that a professional member group would host site-based testing opportunities all across the globe. Moreover the personnel seeking
certification would likely expect asynchronous online testing to obtain certification. Yet such testing creates the very context that continues to rock
the distance-education world. Namely, e-verification of the person who is taking the test or completing the tasks. To the extent that potential employers
require certification – a desired goal to be sure – pressures may exist that could lead to practices antithetical to the verification of individual professional
skills.

Response:
cisco and Microsoft have certification programs which operate remotely, and I think such can be accomplished for accessibility as well. Cheating could happen, but shouldn’t be considered as a primary obstacle to success. I think hands-on training is invaluable for educating people about IT accessibility, assistive technologies especially, but with sufficient creativity, remote learning can do most of this.

You wrote:
Issue number three

While we are talking about performance-based testing, there are several issues that affect quality testing as the basis for certification. Let us be clear,
if this certification is used as any part of employment decisions, it must be rock solid; we would want it so even if not used in employment. Whatever
is developed must contain the highest levels of both internal and external validity. Those not familiar with creating valid, robust testing items probably
should not do so. Central to a quality test is a quality-testing standard. Stakeholders representing divergent sectors should create the specific testing
standard and individuals with disabilities must be part of this group. Now don’t get me wrong, when I talk about testing standards my starting point is
WCAG 2.0, as the international standard. However with that said, exactly what is being tested? What is the scope, depth, and diversity of this performance-based
measure?

Here are examples of what I mean: Would we ask an individual to identify accessibility issues, repair an issue given to them, or create content free of
issues? Since each of these requires a different level of skill we need to know what we want; With form labels mapping to 6 different success criteria
within WCAG 2.0, which is tested? Or is each tested in some redundant fashion? Would we judge the use of ARIA for labeling to be OK if they could have
used HTML?

As you can see, a great deal of thought must be brought to the specifics of the testing standards. We cannot afford to be reductionists, stating that we’ll
test for WCAG 2.0. It is much more complicated than that.

We also think it vital that these testing standards are continuously developed as our understanding and use of technologies expand over time. Moreover,
we feel strongly that any resulting testing specification would be publically available, at no cost. This is important so that preparation for certification
could come from many sources, including the ability for an individual to prepare him or herself if they so choose. Public availability of the testing standards
would also provide the opportunity for continuous feedback into the system and keep it relevant to the needs of professionals.

Response:
We have been doing Section 508 testing certifications at DHS for a couple years now. We currently do a 4 day class with hands on training and a final exam of hands on testing following our testing procedures. We have experimented with doing online practice and exam but it needs a lot more work to be considered as a replacement–but should be feasible with much more work. To certify people in testing you have to set a test process you certify them for, the more complicated the process the more difficult training and certification exams are. For each identified test condition, e.g. “do images have alternate text”, with sub-sets, pass/fail ranges and criteria must be defined and taught. If ARIA is an acceptable success criteria to meet a test condition, then it must be taught with clear pass/fail examples and hands-on exercises. Its not easy, and can’t be reduced to WCAG 2.0 certification. Note, we require studentes to take intro to section 508 first, and have minimum HTML knowledge or they never pass.

You wrote:
Issue number four

We feel that for accessibility certification to be taken seriously it needs to be operated by an existing respected independent professional organization
and insured to do so. Any group with suspect motives, or one that springs up just to take on this initiative, would probably not get the needed participation
or collaboration from those currently in the field. This group then needs to figure out who is evaluating the tests that come in from those seeking certification.
Clearly, the personnel used to judge performance must, themselves, be beyond reproach in the field. Moreover, legal challenges to the operation of testing
and certification are likely to follow as they have in just about every other professional certificate; this makes sense in the context of certification
as a gate-keeper for some employment opportunities.

Response:
W3C/WAI is a logical candidate for a start of one certification name recognition, but what else has to take place is a common agreement on what is taught across the community of practice. Getting more consensus is what will take hard work, more than developing and deploying training. Alternately as long as someone can get this started without suspect motives I think it can take off. We are receiving enormous interest here but can’t train the world.

You wrote:
Issue number five

Costs need to be reasonable to attain, and maintain, certification. Everyone in this discussion agrees that there should be an initial certification with
professional paths that would allow for continuation of certification in regular cycles. In other words, this should not be a one-shot deal. However, I
have not heard anyone talk about the costs of doing this – both for the person who receives certification and the group who does the work to offer it.
This will be important as cycles of continued certification are defined; for example, if you must pay to recertify every two years versus every five.

Response:
By building individual courses to get certified costs can be broken down but not always minimized.

You wrote:
Issue number six

Finally, WebAIM believes that this certification process must be a living entity. Technologies and techniques change what it is that a professional does
to ensure accessibility. The testing standards and the certification(s) themselves must reflect this fluidity. Failure to create a mechanism for continuous
updates, or changes, to certification would be its ultimate death as the certificate would quickly be out of touch with what is needed in the field. With
that said, would this require new performance-based testing or evidence of education on the matter, and in what cycles (e.g., 6 months, yearly)? Depending
on the profession you choose to look at you’ll see different mechanisms. While many educators retain licensure by working in the field and obtaining continuing
education credits, physicians who wish to use new surgical procedures must not only be trained, but undergo a practice-based evaluation. I suspect that
some areas of the field are more like educators and others are more like surgeons.

Response:
Agreed, but keeping consensus on what is “right” to teach is going to very very hard at first.

Finally:
Certification is needed by government at the minimum, and employers who develop for us in particular. Employers can’t know if people they hire can meet contract requirements when developing or testing products, and government can’t have any level of comfort with acquisitions without growing our ow testers either. We need this to move from the chef to the fast-food level of availability and the only way to get there is through common agreed upon baseline accessibility education for various stakeholders including developers, testers, project managers, content creators, etc. There are many areas of focus which can be developed with certifications, but I don’t vebelieve an all encompassing certification is reasonable, because its too much, and few if any would actually do everything.

So, in my opinion this needs to be broken in to small doable components, e.g. web tester certification, could even be website tester and web ap tester sub components. Common agreements on what is taught need to be developed with the right set of folks, and then training developed, exams certified, and operators identified. I would stress operators, as in plural to get this considered standard educational work as opposed to a single organization’s certification. Certifying organizations to deliver the training and exams would help move it out to the masses. Pricing might vary but that is a market force that I think needs to be left to what works.

OK, enough for now but I’d love to chat with you about thisfurther.