- From: Jim Ley <jim.ley@gmail.com>
- Date: Tue, 15 Jun 2004 13:21:38 +0100
On Wed, 16 Jun 2004 00:05:44 +1200, Matthew Thomas <mpt at myrealbox.com> wrote: > On 15 Jun, 2004, at 8:51 PM, Jim Ley wrote: > > ... > > I don't this ends up with people just leaving it at defaults, this > > isn't much good, we don't know if the first choice was a deliberate > > choice, or just because they didn't bother, > > You don't know that anyway, since (as described in my citation) browser > behavior differs. If it's important, e.g. a vote, you are definitely > better to default to an explicit null choice (e.g. "(*) I am > abstaining"). For Radio elements? I'd not noticed. However I thought we were looking to sort out the mess, not perpetuate it - adding an "I am abstaining" option isn't any good either, it could be because they're abstaing simply because they've not read the question. The only way to force answering in a sensible way IMO is make it required, but have nothing selected by default. That seems the natural way to me when a deliberate choice is required. Consider "do you want me to spam you?" as developers we'd like to encourage yes's here, but if we default to no (as we have to AIUI for EU regs and things) then some people may simply miss the question, and we end up without permission to sell their soul. > True, I should have said "as standards-compliant browsers do for SELECT > elements". (IMO platform UI standards are just as important, if not > more important, than W3C standards.) Platform UI standards are much more important I would say, which is why we shouldn't do anything that mandates behaviour different from platform UI standards - so if my UI doesn't default to the first if nothing is selected - neither should the web version. Jim.
Received on Tuesday, 15 June 2004 05:21:38 UTC