RE: FW: ALTifier

Looking good!  I tried

http://ssb22.joh.cam.ac.uk/scripts/access

To get the ALT tags I did needed to NOT check the "disable all sight..."
box.  I had gotten the opposite impression from what you wrote.

When I tried it on http://www.temple.edu, which leads off with a client
side image map that lacks ALT tags, It very nicely inserted the tags,
apparently by grabbing text from text links elsewhere on the page that
pointed to the same URL.

Just a couple of minor things:

It left spaces out of the words, e.g. it said "TEMPLEADMISSIONS" instead of
"TEMPLE ADMISSIONS"

Also, where it generates alt text from the URL it covereted

www.owlsports.com

to 

www.owlsports

It's nice that it dropped the .com.  I'd suggest dropping any leading
"www." as well.

Len

At 03:24 PM 2/9/99 +0100, Michael Vorburger wrote:
>Hey Silas,
>
>> Yes you're right, I'm not subscribed to that list.
>
>You should! ;-)  [WAI-ER-IG is pretty low-frequency, other than WAI-IG]
>
>> Your alt thing looks rather good.
>
>Thank you...
>
>> I've managed to temporarily hook up alt_filter.exe to my development
>> copy of the access gateway, on ssb22.joh.cam.ac.uk.  It gets the web
>
>Great, that's what I need to do here for a presentation of my project (at my
>university) next week! If you can, would be great if you could copy it to
>another machine which is permanently switched on and has a faster
>connection. - I hope you are fine with me announcing this to the list, it's
>just about 10-15 people:
>
>Go to http://ssb22.joh.cam.ac.uk/scripts/access,
>type an URL and enable "Disable all sight-related access options"
>to see what alt_filter.exe adds to a page!
>
>Can some people of the list tell me if the ALT added is of any use?
>
>> page, runs alt_filter through it and processes the results.  So you can
>> browse with it.  At the moment the user doesn't get any choice but to go
>> through the filter.  I don't think this way of doing it is a good
>> permanent idea though because the amount of process spawning that's
>> involved does slow it down.
>
>Yep, agree. Actually alt_filter would BENEFIT very much from a
>"non-respawning" model in the sense that the guessed ALT tags would
>gradually get better, because if it is permanently loaded then all pages and
>tags scanned so far could be added to the in-memory structure (linked list,
>ALT registry, "database") which would gradually produce better results for
>subsequent pages because it maybe already saw the IMG/link or other tag and
>required ALT. BTW: This is the reason why alt_filter and alt_report do NOT
>produce the same results, alt_report is BETTER because it uses information
>on a "site-wide" basis, not just the current page.
>
>> Also, calling a Win32 executable is not
>> portable to Unix platforms (most web servers run Unix).
>> May I ask what programming language it is written in?
>
>Sure. It's C++ ("my native tongue" as some call it ;-) with a LEXer HTML
>scanner. The C++ is pretty platform independant and should compile under
>UNIX as well. Actually I tried, some #includes have to be changed and I
>don't have much time now, but it shouldn't be a big deal.
>
>The Win32 alt_filter is a demo. The Win32 Front-End is just a few lines C
>for scanning a file, "guessing", and write it back and mainly calls the
>"back-end" engine. This ALT guess-engine could be integrated into whatever
>else... If you folks think this is really usefull, why not integrate it in
>Lynx, or an Apache post-processig module?
>
>> There is a slight bug in that consistently replacing VALUE with ALT
>> doesn't always work too well in forms, as in
>>
>> <input type="text" name="hpe" size="2" alt="10">
>> where the VALUE attribute is lost.
>
>Yup, thanks for point this out. This is a bug and will be fixed ASAP.
>
>Regards,
>Michael
>
>----
>Michael Vorburger <mike@vorburger.ch> & <michael.vorburger@epfl.ch>
>QUOTE: "Everything that does not kill you, makes you stronger."
>HOMEPAGE: http://www.vorburger.ch
>
>
>
-------
Leonard R. Kasday, Ph.D.
Universal Design Engineer, Institute on Disabilities/UAP, and
Adjunct Professor, Electrical Engineering
Temple University

Ritter Hall Annex, Room 423, Philadelphia, PA 19122
kasday@acm.org        
(215} 204-2247 (voice)
(800) 750-7428 (TTY)

Received on Tuesday, 9 February 1999 10:38:30 UTC