RE: Does a Navigation Block *really* need to be identified as such? (was RE: [WSG] Title attributes)

[John Foliot]

Steve Green wrote:
> John,
> 
> I would agree that there is little or no value in providing a heading
> for a single list. However, we often work on sites that have
> thousands of pages, that have at least two levels of navigation menus
> and sometimes three. There are often other lists at the top of the
> page, such as to the Sitemap, A-Z List, Accessibility Options etc.   

Steve, this is a fair point, but are they separate lists, or nested lists?
(The latter may not need a heading either).  No matter, your point is valid.


The key rebuttal that I am hearing from you however is that _in some
instances_ providing a heading to a navigational list may be of use - which
is far different than saying that *all navigational lists must have
headings* (which, BTW, the FAE alludes to).  And so as a "Best Practice" we
need to be careful what we say, how we say it, and when it may or may not be
applicable.  I don't want to seem like I'm picking a fight  (I'm not),
but I feel that this is very important to underscore.

 
> 
> These lists are usually styled visually so it is obvious what each
> list is, but it can be difficult to differentiate between all these
> lists when they are linearised in a non-CSS user agent such as Lynx,
> Webbie or some mobile devices. Rather than search for a specific link
> (assuming you know what you are looking for), it is easier to scan
> the page for headings, which most of these user agents style
> differently from the list items. 

Again, no disagreement.  I might comment though that there does come a time
when there may be too many navigational links on one page - I have used this
argument in the past regarding fly-out or drop-down menus.  163
"navigational" links on any page is too many, headers or not (IMHO).
 
> 
> I have to say that my opinion is based mostly on my experience of
> testing with mobile devices and Lynx rather than testing with other
> users, although some screen reader users have commented positively on
> the provision of these headings.

Steve, I was not for one minute questioning your authority to have an
informed opinion - far from it (we've been bumping into each other on these
lists for some time now).  I guess all I am saying is that from a
"standards" perspective (this *is* the Web Standards Group list), we need to
be careful about making blanket statements that are based as much on opinion
as factual data.  We lack this hard data today (AFAIK), but I for one would
welcome a conclusive study.

> When a screen reader user is
> navigating within a page, they benefit from having landmarks like
> this. I have had no adverse comments on the hidden headings, but they
> would not have been visible to most of the users we have tested with.

This is very similar to the argument of whether or not a "Skip to
Content"/"Skip Nav" link should be seen or hidden on a page.  I too use Lynx
for testing and demonstration purposes; I have also worked with daily AT
users, and I agree that logically structured pages are the best way to serve
these user-groups.  Again however, I must stress that we need to be careful
about making blanket statements about any user-group, and their needs or
what is "best for them".  That we need guidance and Best Practices for
developing content that "works" for all user groups, there is no argument,
but we need to be very careful about generalized statements, especially when
they are also partially based on opinion.

> 
> I really don't understand your objection and certainly don't see it
> as 'segregation'. We do all kinds of things to benefit specific user
> groups, and this is just another. 

Well... I simply feel that if you are going to be using headers to group
navigational blocks, they should probably also be "viewed" on screen by all
users - if they are of use to blind users, and Lynx users, and mobile device
users... Surely they will also be of use to older users, users with
cognitive disabilities, etc., etc.  In other words, if you are providing
these headings, why are you "excluding" one group? (even if that group is
the majority?) I of course suspect I already know the answer (usually it's
because it disrupts the visual design), but if it's good for one, why not
all?  I am from the Universal Design camp BTW .

Cheers!

JF
 




***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***

RE: Does a Navigation Block *really* need to be identified as such? (was RE: [WSG] Title attributes)

[Steve Green]

John,

I would agree that there is little or no value in providing a heading for a
single list. However, we often work on sites that have thousands of pages,
that have at least two levels of navigation menus and sometimes three. There
are often other lists at the top of the page, such as to the Sitemap, A-Z
List, Accessibility Options etc.

These lists are usually styled visually so it is obvious what each list is,
but it can be difficult to differentiate between all these lists when they
are linearised in a non-CSS user agent such as Lynx, Webbie or some mobile
devices. Rather than search for a specific link (assuming you know what you
are looking for), it is easier to scan the page for headings, which most of
these user agents style differently from the list items.

I have to say that my opinion is based mostly on my experience of testing
with mobile devices and Lynx rather than testing with other users, although
some screen reader users have commented positively on the provision of these
headings. When a screen reader user is navigating within a page, they
benefit from having landmarks like this. I have had no adverse comments on
the hidden headings, but they would not have been visible to most of the
users we have tested with.

I really don't understand your objection and certainly don't see it as
'segregation'. We do all kinds of things to benefit specific user groups,
and this is just another.

Steve

-

Steve,

Is this based on your user-testing feedback (no downside)?  My only concern
is that we're hiding the heading via CSS for the majority of "mainstream"
users, yet leaving it in for "the others" - I find this hard to accept.
This segregation feels like a downside to me...

Who here really has a problem understanding the following:

SUPER-DUPER WEB SITE
 * Home
 * About this Site
 * Frequently Asked Questions
 * Contact Us
 * Site Map

...? I suggest that even the newest beginner, sighted or otherwise, will
quickly and easily grasp both the concept and the function of that list - it
is, after all, the foundation of the web itself - click on those link-words
and that's where you go.  Screen readers in particular will announce each as
a link, whereas un-styled sites/user-agents will also indicate that they are
links (blue underline, etc.).  Do we really need to hit them over the head
harder than that?

As I was thinking about this, it occurred to me that one person who might
also have some insight into this would be Jonathan Chetwynd of Peepo
(http://www.peepo.com/), who has done some extensive and valuable work with
Downs Syndrome users, and has a very clear grasp of users with severely
cognitive impairments.  Yet a check of his site (which is essentially a list
of navigation links) shows that he has not bothered announcing that his list
of navigation links is a list of navigation links. (Even just typing that
out makes it seem kind of redundant).

To be sure, a consistent placement and treatment of site navigation on each
site is important, perhaps even critical.  As Roger's OzWai presentation[1]
alluded to however, even here the "location" of the primary navigation block
was less critical than the consistency - that said it also left me with the
feeling that all things being equal, newer, less experienced users showed a
slight preference for site navigation at the "top" of the document. (And
Roger's testing panel was very small).  With this in mind, and convention
being what it is, it would seem (to me) that for the majority of users, the
initial list of links at the top of any page are used for navigation - we
don't need to keep telling them that (after all, if we include it on the
first page, it will be on *every* page, and I'm sure your non-sighted
userbase have comments about overt verbosity...)

I think at this time we are still very much in the realm of *opinion*, and I
respect that many in the field of web accessibility and web standards are
well meaning, well versed in the "issues", and want to do everything they
can to improve and maximize the user-experience for all; but I also honestly
think we need imperial data and proof that this *is* the best practice
before we start floating it as such - I for one remain skeptical.

JF

[1] If you missed the original link:
http://www.usability.com.au/resources/ozewai2005/




***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***



***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***

Does a Navigation Block *really* need to be identified as such? (was RE: [WSG] Title attributes)

[John Foliot]

Steve Green wrote:
> The use of hidden headings for navigation is of benefit to anyone
> whose user agent does not support CSS, not just screen reader users.
> We are seeing an increasing number of sites built that way and there
> isn't a downside that I can think of so perhaps it should become
> standard practice. 

Steve,

Is this based on your user-testing feedback (no downside)?  My only concern
is that we're hiding the heading via CSS for the majority of "mainstream"
users, yet leaving it in for "the others" - I find this hard to accept.
This segregation feels like a downside to me...

Who here really has a problem understanding the following:

SUPER-DUPER WEB SITE
 * Home
 * About this Site
 * Frequently Asked Questions
 * Contact Us
 * Site Map

...? I suggest that even the newest beginner, sighted or otherwise, will
quickly and easily grasp both the concept and the function of that list - it
is, after all, the foundation of the web itself - click on those link-words
and that's where you go.  Screen readers in particular will announce each as
a link, whereas un-styled sites/user-agents will also indicate that they are
links (blue underline, etc.).  Do we really need to hit them over the head
harder than that?

As I was thinking about this, it occurred to me that one person who might
also have some insight into this would be Jonathan Chetwynd of Peepo
(http://www.peepo.com/), who has done some extensive and valuable work with
Downs Syndrome users, and has a very clear grasp of users with severely
cognitive impairments.  Yet a check of his site (which is essentially a list
of navigation links) shows that he has not bothered announcing that his list
of navigation links is a list of navigation links. (Even just typing that
out makes it seem kind of redundant).

To be sure, a consistent placement and treatment of site navigation on each
site is important, perhaps even critical.  As Roger's OzWai presentation[1]
alluded to however, even here the "location" of the primary navigation block
was less critical than the consistency - that said it also left me with the
feeling that all things being equal, newer, less experienced users showed a
slight preference for site navigation at the "top" of the document. (And
Roger's testing panel was very small).  With this in mind, and convention
being what it is, it would seem (to me) that for the majority of users, the
initial list of links at the top of any page are used for navigation - we
don't need to keep telling them that (after all, if we include it on the
first page, it will be on *every* page, and I'm sure your non-sighted
userbase have comments about overt verbosity...)

I think at this time we are still very much in the realm of *opinion*, and I
respect that many in the field of web accessibility and web standards are
well meaning, well versed in the "issues", and want to do everything they
can to improve and maximize the user-experience for all; but I also honestly
think we need imperial data and proof that this *is* the best practice
before we start floating it as such - I for one remain skeptical.

JF

[1] If you missed the original link:
http://www.usability.com.au/resources/ozewai2005/




***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***

RE: [WSG] Title attributes

[Steve Green]

The use of hidden headings for navigation is of benefit to anyone whose user
agent does not support CSS, not just screen reader users. We are seeing an
increasing number of sites built that way and there isn't a downside that I
can think of so perhaps it should become standard practice.

Screen readers do not read 'title' attributes by default. You can configure
some to read 'title' attributes instead of the on-page text but no one is
going to have that as a permanent setting. You can also read the 'title'
attribute for a specific element but that presupposes the user knows which
elements have 'title' attributes.

Tooltips of any kind can be a nuisance for screen magnifier users because
even a small one can obscure a large proportion of the screen at modest
magnification levels. It is even worse when the tooltip is caused by the
'title' attribute for a structural element such as a paragraph or a div
because the user does not know where to move the mouse to get rid of it. It
may not even be possible if the element fills the entire screen. For this
reason I would not recommend using a 'title' attribute for a list.

Steve Green
Director
Test Partners Ltd / First Accessibility
www.testpartners.co.uk
www.accessibility.co.uk





From: listdad@webstandardsgroup.org [mailto:[EMAIL PROTECTED]
On Behalf Of Antonios Sarhanis
Sent: 24 January 2007 23:00
To: wsg@webstandardsgroup.org
Subject: Re: [WSG] Title attributes


I give headings to my navigation, as well as other areas on the page, but
the headings are hidden (position: absolute; left: -100em) so that they can
be read by a screen reader.

>From what I've read, title attributes should only be sparingly and in
special cases where more information might be helpful rather than annoying. 

Having the title say exactly what a piece of text says is completely
useless, 
and having the title say something slightly different to what a piece of
text says only makes things annoying for users with a screen reader that
might have to read both instances of the very similar text. 



***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***

Re: [WSG] Title attributes

[Barney Carroll]

Katrina wrote:

Your post really reminded me of an older post that talked about this:
http://www.usability.com.au/resources/ozewai2005/

May I asked which automated accessibility test?

Hope that helps.
Kat


http://fae.cita.uiuc.edu/ - it's pretty good generally, but as with all 
automated tools it is only really a tick box. Validity for pseudo-human 
readability if you will.


That link of yours is pretty good. The source order debate is one worth 
having again and again - although I think the writer puts too much faith 
in what people have come to expect (which is varied and incongruous 
enough) and not enough in what people might want.


Superfluous headings: Sure, repeated information of any kind is 
irritating. But whenever I see 'Site Navigation:', I can't help but feel 
it's incredibly inane.


Incidentally, I did not fail absolutely on that particular criteria - 
the site in question has the first two levels of navigation in separate 
columns on the left, while the third level takes the shape of an inline 
ul right-aligned at the top of the main content... Because of possible 
ambiguity what with the different format, this is headed Inside 
[section name]:. In that instance I think it's 
forgivable, and necessary, because it rests inside different block 
elements, and for users with CSS and a visual display, the semantics of 
position may confuse. Otherwise I think there's no confusion over a list 
of links (provided they have conventional text) being a navigation tool 
of some kind, and the general one need not be labeled.


Regards,
Barney


***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***

RE: [WSG] Title attributes

[John Foliot]

Katrina wrote:
> Barney Carroll wrote:
>> I was recently told by an automated accessibility test that my
>> navigation was not up to scratch because it simply consisted of a
>> plain ul at its highest level. It penalised me for not having a
>> preceding heading to give some kind of indication of what the ul was
>> about. Now I've never seen a navigation marked 'navigation', and
>> even in the most of eccentric sites I've always been correct in the
>> assumption that the first list of links with category denominations
>> is some kind of navigation tool. 
>> 
>> I'd like to know what people think about that first of all - does
>> anybody give headings to their navigation?

While using a heading to identify your navigation *may* be useful in some
instances, it is neither an "official" requirement, nor to my knowledge even
a documented Best Practice beyond what the tool (which I suspect is the FAE)
is suggesting (http://cita.uiuc.edu/html-best-practices/nav/menus.php).  I
have had this discussion with the FAE developer, and it really is an
opinion.  The opinion is grounded in experience, to be sure, however the
omission of an  was originally marked as a fail, and now is simply
marked as a warning. 

My *opinion* is that even the most inexperienced non-sighted visitor will
very quickly understand a navigational block of links for what they are - to
presume differently is a tad patronizing to me.  I think that this result
also under-scores the critical pint that no single automated testing tool
can determine beyond question the accessibility of any given site - it
requires a thinking brain to make that final determination. 

> 
> 
> Your post really reminded me of an older post that talked about this:
> http://www.usability.com.au/resources/ozewai2005/

Presumptions are just that.  To categorically state X, Y or Z regarding any
user or user-group is bad juju, as Roger's presentation illustrates.

> 
> May I asked which automated accessibility test?


I will guess that it is/was:
  http://fae.cita.uiuc.edu

Register for a free user account at:
  http://fae.cita.uiuc.edu/register.php.


As always, just my $0.02.

JF




***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***

Re: [WSG] Title attributes

[Antonios Sarhanis]

I give headings to my navigation, as well as other areas on the page, but
the headings are hidden (position: absolute; left: -100em) so that they can
be read by a screen reader.


From what I've read, title attributes should only be sparingly and in

special cases where more information might be helpful rather than
annoying.

Having the title say exactly what a piece of text says is completely useless,
and having the title say something slightly different to what a piece
of text says
only makes things annoying for users with a screen reader that might have to
read both instances of the very similar text.

On 1/25/07, Katrina <[EMAIL PROTECTED]> wrote:


Barney Carroll wrote:
> I was recently told by an automated accessibility test that my
> navigation was not up to scratch because it simply consisted of a plain
> ul at its highest level. It penalised me for not having a preceding
> heading to give some kind of indication of what the ul was about. Now
> I've never seen a navigation marked 'navigation', and even in the most
> of eccentric sites I've always been correct in the assumption that the
> first list of links with category denominations is some kind of
> navigation tool.
>
> I'd like to know what people think about that first of all - does
> anybody give headings to their navigation?


Your post really reminded me of an older post that talked about this:
http://www.usability.com.au/resources/ozewai2005/

May I asked which automated accessibility test?

Hope that helps.
Kat


***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***





***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***

Re: [WSG] Title attributes

[Katrina]

Barney Carroll wrote:
I was recently told by an automated accessibility test that my 
navigation was not up to scratch because it simply consisted of a plain 
ul at its highest level. It penalised me for not having a preceding 
heading to give some kind of indication of what the ul was about. Now 
I've never seen a navigation marked 'navigation', and even in the most 
of eccentric sites I've always been correct in the assumption that the 
first list of links with category denominations is some kind of 
navigation tool.


I'd like to know what people think about that first of all - does 
anybody give headings to their navigation?



Your post really reminded me of an older post that talked about this:
http://www.usability.com.au/resources/ozewai2005/

May I asked which automated accessibility test?

Hope that helps.
Kat


***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***

Re: [WSG] Title attributes

[Micky Hulse]

Barney Carroll wrote:
I'd like to know what people think about that first of all - does 
anybody give headings to their navigation?


Sometimes I like to use a DL for my primary/secondary navigation... 
using the DT I can give the nav a title, and then position it off-screen 
using absolute positioning and monstrous negative-left value. :)


For some Yahoo! tabbed navigation I dinked with a while back, they used 
heading tags that got positioned off-screen and a couple UL's (this was 
a multi-level tabbed nav.)


Hth,
M


--
 Wishlist: 
   Switch: 
 BCC?: 
   My: 


***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
***