The latest article on A List Apart (welcome back!) offers some excellent tips on how to go about letting users know which accesskeys are available.
As it happens, my recent accesskeys post has also seen more comments of late. In particular, Geoff Hoffman has compiled and posted common browser shortcuts – essentially a handy list of accesskeys to avoid.
The post also led Joe Clark to start a fascinating thread on the WAI Interest Group mailing list which eventually boiled down to accesskeys not being that helpful afterall. The reason being (as hinted by Geoff’s post earlier) is that so many browsers and assistive technologies make use of the keyboard shortcuts being assigned to accesskeys. A key point raised on the mailing list is that this isn’t a problem with accesskeys, more a problem of their implementation in browser software. The accesskey issue does range wider though. As outlined in Using Accesskeys – Is it worth it? such issues as international keyboards also come into play:
So while it seems that accesskeys is a great idea in principle, implementation brings with it the possibility that it either will not be available to all users, or that the keystroke combination encoded within the web page may conflict with a reserved keystroke combination in an adaptive technology or future user agent.
Is this time for an ‘Accesskeys considered harmful’ essay?
Jon wrote:
Perhaps there are two separate issues:
If I visit a site daily or even weekly, I would definitely make use of an access key for the homepage. For example, what if [h] was always the homepage accesskey? I would use it all the time.
Rich wrote:
Jon – you may be right there. My feeling is that accesskeys were first introduced to ease form navigation (and make it more like GUI dialog boxes).
BTW the accesskey for the homepage on clagnut is [1]. Based on my brief research this seems close to a de facto standard.
Dave S. wrote:
I’ve done a bit of investigating into AccessKeys myself. That they break existing software shortcuts is a pretty big caveat.
Even numbered keys aren’t safe – from what I’ve been told, JAWS uses ALT + NumPad combinations, so the main group of people who’d benefit from AKs (the blind) will be locked out of their screenreader.
I like the idea of AKs, but there’s only so much you can do.
Simon wrote:
Dave S. writes that JAWS users may not be able to take advantage of numbered accesskeys due to the large number of alt+NumPad shortcuts the programme offers.
However, it should be noted that JAWS turns off the NumLock for the numeric keypad, so that alt+1 and alt+NumPadOne are actually different. Users can’t turn NumLock for the keypad back on whilst JAWS is running.
Thus alt+number AKs seem to me to still have their uses.
John Foliot wrote:
The Accesskey “debate” rages on.
In a non-scientific study conducted in the summer of 2002, we researched the availability of available Accesskeys which had not already been reserved by various other software technologies which might be employed by various users. The results indicated a real problem in that most ALT + keystroke combinations (assuming the Windows operating platform) have already been reserved by one type of application or another.
Given that at this writing there are few if any available ALT + keystrokes left, and conceding that built in or assigned keystroke combinations inherent in software applications should take precedence for the end user, WATS.ca continues to advocate that the ACCESSKEY element/attribute, while a good idea in principle, is fraught with so many potential problems that we have abandoned using them in the interest of true inter-operability.
To view our “listing” of reserved keystroke combinations, see:
http://www.wats.ca/resources/accesskeysandkeystrokes/38
Ruth wrote:
I’am having problems whit my accesskeys, when I press Alt and the key I want (161–162-163 example) I only get little squares. Can you help me?
Small Paul wrote:
Late to weigh in on this, but I’ve been mulling a “Hey, access keys – they’re pretty good” article, to try and sum up all the useful material that’s going around about the issue – and to mitigate against the fact that most influential bloggers don’t use them.
When access keys clash, I think it’s more of a problem if the user agent gives precedence to the HTML access key (e.g. ALT + D in WinIE) than if it doesn’t. According to a webaim discussion (http://www.webaim.org/discussion/mail_thread.php?thread=1584&id=3631), JAWS, IBM HPR and WindowEyes don’t support HTML access keys. So although those users won’t be able to access them, they won’t have their own keys overridden. I don’t see this as a reason not to use access keys – let user agents oriented towards the visually impaired stick with their richer, more useful access keys.
From what I saw from the wats.ca piece, the numbered keys are used by WindowEyes for user-defined shortcuts, and nowhere else in the user agents they examined (except for 5 in HPR). This suggests to me that the numbered keys are reasonably safe.
I’d agree with others who’ve said that only a small amount of access keys are useful, and that a cross-site standard would be nice. So full marks to Rich for the research he did.
The 5 keys used on clagnut all link to pretty important/useful things. Judging from the wats.ca research, they clash minimally with application access keys. And if we all did it, it could be relied upon. This would help users with low motor skills, and would make some of my web browsing faster.
So why not?
Small Paul wrote:
Just one thing though: I like how on Mark Pilgrim’s site ALT+4 takes one to the search box. Unfortunately here, it’s on a link to the search page, and the search page expects to be called with a search term in the query string. So the user gets an error. Just a little thing, reely.