Re: [webcomponents] Custom Elements Spec

On Tue, May 8, 2012 at 1:48 AM, Tab Atkins Jr. <jackalmage@gmail.com> wrote:

> For the same reason that jQuery plugins are authored by a
> significantly smaller set of people than "jQuery users".
>

Significantly smaller, but not necessarily significantly more capable.

*Theoretically*, every jQuery user can write their own plugins - it's
> really easy, and very useful.  In practice, almost everyone uses ones
> written by someone else.  Part of the *point* of Components is to
> reach exactly this model, and have a safe (from an author-usability
> perspective) way to include someone else's HTML in your own, so we're
> not all constantly reinventing the same markup patterns.
>

Staying with the jQuery comparison, there are TONS of plugins that solve
the same problem. Some in much better ways than others. There's a large
potential for problems when the learning curve for the basics is low.
There's an even larger problem when developers start blogging about their
recent discovery (which no doubt will lack any form of accessibility) and
other developers start learning from there.

> And it's not exactly about "making the right choice". Component
> > authors will have to do a significant amount work.
>
> Again, the example of jQuery is instructive.  A lot of plugins do
> *not* create accessible markup.  However, the most popular ones *do*,
> and because of this, you end up with much more accessible markup
> overall than if every user was inventing things themselves.
>

I'd say some do, and an extremely minuscule amount actually use ARIA. I'm
not convinced that even the top 10% of developers writing plugins are
always thinking about accessibility.

Accessibility is hard. I can speak personally about how difficult it is to
design widgets for jQuery UI so that absolutely no choice is left to our
users about how accessibility works. Developers will find ways to bend
features to their will; if they can get something to work the way they
want, they'll do it, regardless of whether it breaks accessibility.

Received on Tuesday, 8 May 2012 12:34:07 UTC