WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: End of page notification

for

From: Jukka K. Korpela
Date: Dec 17, 2012 1:33PM

2012-12-17 21:39, Chagnon | PubCom wrote:

> In traditional print publishing, either a square box (such as ■),
> the "End of Proof/Halmos" symbol (∎) or 3 hash marks (###) signal
> the end of the story's content.

I have never seen the latter, but I have seen a square box, but it's far
from ubiquous. Many other symbols are used, such as a rectangle or a
lage bullet - namely to indicate end of an article. A web page might
contain several articles. An end-of-article marker is mostly needed in
publications that contain several articles, and an end-of-page indicator
is something different. After all, on the web, "page" means a document,
which might be anything from a very short note to a book-length
presentation.

I mostly find the issue a non-problem. Billions of pages lack any end of
page indicator in content, and end of page is really something to be
noticed by a browser rather than expressed explicitly. Especially if a
page ends with a footer containing creation date, last update, author
name or company name, and contact information, it should be pretty clear
that the page ends there.

> Personally, I think it would help AT-users to have some marker that
> tells them they have reached the end of the main story content on a webpage.

Main story content is different from end of page. And the end of main
story content is probably best expressed by indicating that what follows
is not main content any more. That is, the problem would then be how to
signal the start of a footer.

> Professionally, it should be the Halmos symbol [...] but in our testing,
> none of the screen readers with default settings interpreted any of these symbols correctly.

The best we can reasonably expect from screen readers is that any symbol
that is otherwise unknown is spoken by uttering its Unicode name. What
you suggest is named "end of proof", so it would hardly quality, since
mostly the main content of a page is not a proof. The symbol is clearly
meant for mathematical use; its General Category is "Symbol, Math". So
in that sense, it would be better to use a semantically vague rectangle
or box character.

And since support to special characters is so limited in screen readers,
an image would actually improve accessibility. You could then use a
symbol of your preference, with descriptive alt attribute value such as
"End of page" or "End of main content". But I don't see why that would
be needed.

> (And I wonder, how many of our list members were able to read the characters
> inserted in this email? They're all Unicode characters.)

I had no difficulty in reading them, and I suppose most email programs
can display the "end of proof character". But it's a different issue
whether people understand its meaning, and yet another issue what screen
readers do.

Yucca