Re: Compromise on Numbering: Changing 11 AAA numbers solves the level problem

> Numbers are a terribly brittle way to ID something, and we have much
better IDs already in the spec.

That's interesting. I hadn't thought about the HTML ID numbers which were
previously XML ID numbers as ways to refer to success criterion. I'm quite
certain that these IDs are not used in common practice anywhere where
people are referring to SC's in the public. Usually records in a database
are identified by number rather than an ASCII text string.

I was using the term ID as the way that automated tools and accessibility
professionals would refer to the success criterion rather than the way that
we would internally or in our code for the document.

The concern you're raising is interesting and will have to address that at
some point.

Detlev, I think Michael was concerned about using the current success
criterion numbers as IDs rather than the four digit scheme which I also
think is a poor choice for us. The mockup is actually a good way to
demonstrate that it's doesn't work.

Cheers,
David MacDonald



*Can**Adapt* *Solutions Inc.*
Mobile:  613.806.9005

LinkedIn
<http://www.linkedin.com/in/davidmacdonald100>

twitter.com/davidmacd

GitHub <https://github.com/DavidMacDonald>

www.Can-Adapt.com <http://www.can-adapt.com/>



*  Adapting the web to all users*
*            Including those with disabilities*

If you are not the intended recipient, please review our privacy policy
<http://www.davidmacd.com/disclaimer.html>

On Wed, Sep 20, 2017 at 9:41 AM, Michael Cooper <cooper@w3.org> wrote:

> On 20/09/2017 9:10 AM, David MacDonald wrote:
>
> If we do that I think should start referring to the numbers as ID#s. Its a
> change in layout because WCAG 2 used the numbers as "Outline" mode to order
> them. The new layout would be changing that "ID" mode as unique identifiers
> but not the common way of referring to them by lay people. I'm OK with that
> change but I think we should articulate it.
>
> We should not refer to numbers as IDs. Numbers are a terribly brittle way
> to ID something, and we have much better IDs already in the spec. In WCAG
> 2.0 the ID for SC 1.1.1 is "text-equiv-all"; in WCAG 2.1 we base the ID on
> the SC title so it's "non-text-content". In both cases there is a lot of
> infrastructure built around those IDs, and no infrastructure built around
> the numbers.
>
> I know I'm going to lose the debate on numbers, where my position is that
> they are meaningless and we should number things as appropriate to *this*
> spec, but we should not attempt to solve concerns with numbers by declaring
> them as IDs when they are not and we already have better, more stable IDs.
>
> Michael
>