Re: ALT revisited

At 12:44 1997/06/30 -0400, Al Gilman wrote:
>to follow up on what Dave Raggett said:
>> 
>> One proposal quite a few years old is for authors to provide
>> a null value for alt as distinct from omiting alt, e.g.
>> 
>>   <img src=blank.gif alt="">
>> 
>
>This is not just a proposal.  Lynx responds reasonably well to
>this usage and there are authors using it.

This alt="" reliably informs the client that nothing of consequence is in 
that gif.

If the alt="..." were #REQUIRED, I would not have it default to "" in any
authoring tool, but rather to some serious reminder to the author that useful
description is required of information otherwise conveyed. Also the consequence
of  alt="" should be noted; it may get rid of some of that information-free 
noise.

>> User agents would be recommended to default alt based
>> upon the url's name, e.g.
>> 
>>   <img src=logo.gif> would yield  "logo" as the default.
>> 
>> Does this sound reasonable?
>>
That could give a "first-look" default while awaiting the response to
the head title content request Al proposes. 
>
>There are better defaulting rules, but they take more work.  Have
>you read
>
>http://www.access.digex.net/%7Easgilman/web-access/announce-two.html
>
>and the attached note about "metadata percolation"?
>

One of Al Gilman's proposals, as I understand it, is upon encountering a 
non-local URL with no alt="...", to get the head of that URL, and extract the
TITLE from it to provide the default value. That seems useful, at least 
for those documents that actually include the required TITLE element! 

>The proposal there reflects my reaction to several go-arounds on
>this issue on the lynx-dev list [not a Lynx consensus, may I make
>clear].  [More in reponse to Daniel...]
>
>WARNING: I don't think that we have time enough to reach a
>consensus on this issue here before you release this draft, even
>if it's not tomorrow.

Agreed.
>
>--
>Al Gilman
>
Regards/Harvey Bingham

Received on Monday, 30 June 1997 18:50:20 UTC