Terrence Wood wrote:
> Thierry Koblentz said:
>>> Both. You have misinterpreted the articles, and have formed an
>>> opinion based on that misintrepretation.
>> I disagree.
> So you keep saying, but your actions are different.

I'm not sure I agree with that.
Please see the bottom of this message.

>>> you use skip links on your site but are argueing here that every
>>> link must load an entirely new document.
>> What I'm using on my site has absolutely nothing to with the way I
>> interpret the USEIT articles.
> I never said it did. What I said is you are practicing the opposite of
> what you are preaching.

I think you may be misinterpreting.
I am *not* preaching anything, I am *not* saying what should be done or
used, I am only arguing about the fact that "jump links" have been presented
in this thread as a usability-free solution. That's all.

>>> If you replaced it with "chocolate orange cake" it would make sense
>>> according to your logic, but it becomes glaringly obvious just how
>>> wrong that logic is.
>> I disagree
> in fact, you *do* agree with me, you just seem unable to see how it
> relates to the argument you are putting forward... Read on
>> FWIW I find your analogy pretty silly.
> Exactly. As is your assertion that a recommendation against opening
> new windows is a recommendation against using in-page anchors.
> Substituting anything in Nielsens recommendation distorts the
> recommendation: it *is not* what he said, and it *does not* make
> sense. "using jump links" is not the same as "opening new windows"
> and it clearly isn't "chocolate orange cake".

I respectfully disagree for the following reasons:
Following Jay's excellent advice, I wrote an email to Jacob Nielsen telling
him about my interpretation of both of his articles.
I explain the little "issue" we were discussing here and asked him if he
could find the time to answer this question:
"My interpretation of these articles is that the reference to popup windows
does not exclude "jump links" from being an issue themselves in regard to

Less than half an hour later, I received his brief answer:
You are right: links within the same page almost always cause confusion in
user testing, and it is almost always best to avoid them.

>>> Hypertext links are the foundation of the web.
>> That's the W3C talking, AFAIK, it has absolutely nothing to do with
>> usability/accessibility.
> Web Standards. Consistency. Platform conventions. The thing that
> defines the web.
>> It is about how things are supposed to work
> Exactly. If things work the way they are supposed to, then you can't
> get much more usable than that.

I respectfully disagree for the following reasons:
In an ideal world, things work the way they are supposed to; unfortunatley
the Web is not an ideal world and in my opinion there are 2 paths:
recommendations and best practice. The latter usually address
usability/accessibility issue.

>> [accesskey's have] usability/accessibility issues attached
> Yes they do, but that is a browser implementation issue, not a markup
> issue. e.g. Macs browser's and Opera's accesskey implemenatation do
> not conflict with the OS like other PC browsers.

What about assistive devices? Setting accesskeys that clash with user's
shortcut keys? What about implementing redundant mechanisms that break
keyboard navigation (onkeypress)? What about about implementing skip link
that follow the recommendations but also break keyboard navigation. These
are usability issues that often come up simply because authors implement
techniques relying too much on recommendations.

>> So how can you say that "jump links" in a document are consistent
>> with the navigation links for example?
> They don't have to be, in the same way that main nav, secondary nav,
> and in-content links are generally easy to distinguish and
> understand: they should be consistent within the context in which
> they appear (internally consistent within a block?).
> That said, you might have to hack in a div or heading here and there.
> See: http://en.wikipedia.org/wiki/Hyperlink#toctitle
>> Actually, I believe the key is to let the user *know* what's about to
>> happen... a FAQ page that says "clicking on the Qs will reveals the
>> As below" is less an issue than "jump links" that do not warn the
>> user of what's gonna happen next.
> When a user clicks on a link they *know* they will be taken to the
> resource described by that hypertext link. It doesn't even need an
> explantion because it is so fundamental.

Apparently Jacob Nielsen doesn't agree on these points...

Thierry | www.TJKDesign.com

The discussion list for  http://webstandardsgroup.org/

 See http://webstandardsgroup.org/mail/guidelines.cfm
 for some hints on posting to the list & getting help

Reply via email to