Re: Updated accesskey.next material

21.11.2014, 01:18, "John Foliot" <john@foliot.ca>:
> Léonie Watson wrote:
>>  The thing I like about author defined suggestions is that they're
>>  likely to be suited to the website/application, and therefore
>>  (probably) easier for people to learn.
>
> A "maybe" that I will grant you - it's not an unreasonable supposition

Yeah, that is the reason it is in the wiki already.

>>  The counterpoint is the
>>  likelihood of the author defined suggestions remaining active, given
>>  the conflict resolution chain.
>
> yep, coupled with i18n considerations: For example I've often used the
> example of "Search" - many authors would consider it 'reasonable' to map
> "Search" to "S", but what about if you are French (R for Rechercher) or
> Spanish (B for Buscar)... 

If the author declare a standard role, then accesskey isn't the right mechanism. You may add an accesskey as well, but the user agent would provide a default binding for e.g. search. Which means the accesskey is irrelevant.

If the authors adds accesskey instead of declaring a standard role, they're doing it wrong. But even so, they provide a *hint*. We are in violent agreement that there is no obligation for the user to agent to respect the author's choice - and in case of conflicts the user agent *should* bind the function to an unbound user interaction.

Whether the hint makes sense or not is ultimately irrelevant, since the actual interaction (key, gesture, etc) bound will be decided by the user agent (presumably with reference to the user themselves, but hopefully the defaults will be reasonable).

cheers

--
Charles McCathie Nevile - web standards - CTO Office, Yandex
chaals@yandex-team.ru - - - Find more at http://yandex.com

Received on Friday, 21 November 2014 13:27:35 UTC