RE: [3.1][2.4] Informative page titles: suggested solution

Hello list,

I will try to address some of Joe Clark's remarks about my suggestion:

> > Current phrase:
> > Page titles are informative
> > 
> > Suggested phrase:
> > All pages have informative titles
> 
> Not all Web structures are "pages." 

I know, that's a different problem which I didn't want to address because
that would cloud the issue.

> Further, <title></title> 
> can be empty and still validate; it is possible to publish 
> standards-compliant untitled pages.

Yes, so? Compliant doesn't mean accessible. For instance, it's also possible
to create standard compliants image links with empty alt-texts, which isn't
accessible either. If all you needed to do to create an accessible webpage
is follow the HTML specs we wouldn't have a WCAG working group :-) We define
guidelines that go beyond the standards all the time. 

> 
> 
> > I think "all pages have informative titles" should be at least at 
> > level 2, perhaps even at level 1, because of the 
> significance of the 
> > page title in orienting yourself in content. If you have 
> crappy link 
> > text but informative page titles, at least you'll know that 
> you ended up in the wrong place.
> 
> People keep bringing that up. I don't see how it's an 
> accessibility issue. 

People without disabilities will usually have plenty of visual cues within
seconds to determine what the page is about. If you don't provide a page
title, a blind person will have to search for other clues to 'scan' the page
for a general idea about the content. That makes the task of getting to know
what a page is about disproportionately harder for a disabled person than
for a non-disabled person, which in my opinion makes it an acccessibility
problem. 


> The page I link to is only my responsibility when I control 
> both pages. 

Totally agree with you there, I wasn't talking about providing page titles
for _other_ people's pages, only for your own.

> Further, the way to determine if you've reached the right 
> page or not is to read and understand it in its entirety. 

Frankly, I can't remember the last time I read a webpage with more than 100
words in it's entirety. I am from the 'fast-food' generation, maybe that has
something to do with it. If I visit a page I want information and I want it
NOW :-) Especially when visiting pages found by search engines, I usually
decide within two seconds whether the page is what I want or not. I look for
visual clues, headings, etc. to scan what the page is about and then decide
to read (parts of) the page. 

> Why 
> is there an effort to load everything onto the title element?
> 
> The HTML spec states:
> 
> <http://www.w3.org/TR/REC-html40/struct/global.html#edef-TITLE>
> 
> Authors should use the [90]TITLE element to identify the 
> contents of document.
> 
> 
> "Identify the contents" is a smaller and more appropriate 
> burden than "make sure every imaginable Web-surfer realizes 
> whether or not they've made a mistake."

If you identify the contents, that should give the web-surfer a good idea
what the page is about and whether or not that's the information he is
looking for. I never meant to suggest we should have long winded titles that
read like prose, just that it should be informative. There are too many
"Untitled document"'s out there which makes it hard to orient yourself. 

> 
> Also, um, how often, *really*, have you hit a link on a page 
> and "ended up in the wrong place"? 

Well, probably about a dozen times per week. I use search engines a lot and
very often I don't end up where I want to be. In my spare time I use search
engines to search for genealogical information. It happens all the time that
I look for two family names (of spouses) and the resulting link look
promising (both names present) but end up being a sporting match instead of
a genealogical homepage. Also, searching for certain technical/nerdy
information using search engines tends to result in a lot of links to pages
with porn or warez. 

But this is a special situation, since search engines use the page titles to
create their links. If people don't give their pages informative titles,
links from search engines will be non-informative and you'll end up in the
wrong place. This is because the link text is uninformative, which is in
turn caused by the fact that the page title is uninformative. 

In all fairness, it doesn't happen very often that I end up in the wrong
place when I follow human-made links. In my experience, most static links on
websites are pretty good in allowing you to predict where you'll end up,
either by link text or by context. 

> Just how often does that 
> happen on the real Web? You may need to ask someone who 
> experiences the real Web for an answer.

FYI: I have plenty of real-world experience using the web. I've been on the
web actively since Mosaic 1. In case you weren't: that was in 1993 ;-) I
built my first web site in that year, and have been doing it ever since.

Yvette Hoitink
Heritas, Enschede, the Netherlands
E-mail: y.p.hoitink@heritas.nl
WWW: http://www.heritas.nl

Received on Wednesday, 7 July 2004 15:45:27 UTC