Does anyone know what algorithm aural browsers use to render pages? I
would think it more sophisticated than FIFO. I've seen some things that
indicate everyone does their own thing, but there must be some general
rule of thumb or it would be utter chaos.
Better still, is there a utility to convert the html to a simple text
transcript.
William Gill wrote:
> Does anyone know what algorithm aural browsers use to render pages? I
> would think it more sophisticated than FIFO. I've seen some things that
> indicate everyone does their own thing, but there must be some general
> rule of thumb or it would be utter chaos.
>
I tested one screen reader (JAWS) combined with Internet Explorer.
The document is read interactively. There's a key to read the next or
previous paragraph, keys to jump to a section defined by a header
level.
There's a heading navigation dialog box to see the structure of the
document and jump to a specific place.
> Better still, is there a utility to convert the html to a simple text
> transcript.
This wouldn't be equivalent to a screen reader such as JAWS.
However, it may be possible to extract the data of a HTML document,
deleting navigation bars and decorative things. I don't have a
specific algorithm to recognize these things.
Moreover tables and divisions will be difficult to interpret.
--
Sincerely.
André Gillibert.
André Gillibert wrote:
>
> I tested one screen reader (JAWS) combined with Internet Explorer.
> The document is read interactively. There's a key to read the next or
> previous paragraph, keys to jump to a section defined by a header
> level.
> There's a heading navigation dialog box to see the structure of the
> document and jump to a specific place.
>
OK, This makes some sense, the browser uses headers to delimit each
logical section of the "copy", and lets the user decide the delivery
sequence. What happens in logical blocks without headings, like
navigation? Are they automatically delivered in logical units in the
order they are encountered (FIFO)?
>> Better still, is there a utility to convert the html to a simple text
>> transcript.
>
> This wouldn't be equivalent to a screen reader such as JAWS.
> However, it may be possible to extract the data of a HTML document,
> deleting navigation bars and decorative things. I don't have a
> specific algorithm to recognize these things.
> Moreover tables and divisions will be difficult to interpret.
I could use a text only browser like Lynx to get a rough extraction, but
I was hoping to be able to review my html and get an estimation of how
it is "seen" by users relying on this technology. A transcript would be
a text copy of what the user would hear, in the order the user hears it.
Guess I'll have to assume that if my code is fairly logical and
reasonably semantic, whatever algorithm they use, it should be rendered
in an understandable manner.
Thanks.
Gazing into my crystal ball I observed William Gill
@newssvr14.news.prodigy.net:
> I could use a text only browser like Lynx to get a rough extraction, but
> I was hoping to be able to review my html and get an estimation of how
> it is "seen" by users relying on this technology. A transcript would be
> a text copy of what the user would hear, in the order the user hears it.
>
Opera has a text only mode, and the small screen mode might be of help as
well.
--
Adrienne Boswell at Home
Arbpen Web Site Design Services
http://www.cavalcade-of-coding.info
Please respond to the group so others can share
On Dec 4, 5:01 pm, Adrienne Boswell
> Gazing into my crystal ball I observed William Gill
>
> @newssvr14.news.prodigy.net:
>
> > I could use a text only browser like Lynx to get a rough extraction, but
> > I was hoping to be able to review my html and get an estimation of how
> > it is "seen" by users relying on this technology. A transcript would be
> > a text copy of what the user would hear, in the order the user hears it.
>
> Opera has a text only mode, and the small screen mode might be of help as
> well.
>
Well you could do that:
Or you could just download the proper screen reading software.
http://www.freedomscientific.com/fs_downloads/jaws.asp
Note that their are other screen readers besides Jaws, it's just that
I use it and think it works the best for web based stuff.
--
Regards Chad. http://freewebdesignonline.org
Chaddy2222 wrote:
> Well you could do that:
> Or you could just download the proper screen reading software.
> http://www.freedomscientific.com/fs_downloads/jaws.asp
> Note that their are other screen readers besides Jaws, it's just that
> I use it and think it works the best for web based stuff.
Thanks Chad. I have found several such readers. Of course I have no
way of knowing which are better than which, or what algorithms they
follow.
I was hoping if I could understand the generic algorithms and "outline"
the page. Then I could incorporate that flow in the early design/layout
phase of development, instead of waiting to test the result. I guess if
I just use the readers regularly, it will eventually sink into my
initial planning phase instead on the final correction phase.
William Gill wrote:
>
> What happens in logical blocks without headings, like
> navigation? Are they automatically delivered in logical units in the
> order they are encountered (FIFO)?
Generally speaking, yes. CSS positioning doesn't really have much effect
on that, though visibility:hidden and display:none most definitely can
cause the reader to ignore those elements.
This site has a lot of good info on the subject of what screen readers
do, among other things:
http://www.access-matters.com/
--
Berg
On 21 Nov, 20:40, William Gill
> Better still, is there a utility to convert the html to a simple text
> transcript.
I'd suggest the very useful Fangs extension for Firefox - it tries to
emulate Jaws' behaviour closely, only with visual display rather than
speech and with free licensing for web developers.
Unfortunately it seems to have fallen off the development wagon for
Firefox. It's in the extensions sandbox, so you might find it if you
hunt.
Andy Dingley wrote:
> I'd suggest the very useful Fangs extension for Firefox - it tries to
> emulate Jaws' behaviour closely, only with visual display rather than
> speech and with free licensing for web developers.
I installed it and don't like how my pages display. I need to test them
against an actual reader for comparison. I also need to see the Fangs
output from a well done page, to better familiarize myself with fangs.