[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: lynx-dev Still learning about what Goes Wrong.

From: Vlad Harchev
Subject: Re: lynx-dev Still learning about what Goes Wrong.
Date: Sat, 29 Jul 2000 12:25:34 +0500 (SAMST)

On Fri, 28 Jul 2000, Martin McCormick wrote:

>       This is a copy of a message I posted to another list,
> today, so those who are on this list and the Blind-L list can
> shoot me, but there may be good answers from both groups hense
> the cross posting.
>         A number of our web sites at Oklahoma State University
> have recently been upgraded and have become just awful as far as
> accessibility goes.  A few have not one single link that works
> under lynx and I have noticed that every one of them was
> generated with Microsoft Frontpage of various versions.

  Since it's not-very-commercial ogranization, chances are big that after some
finite number of complaints they will fix this. You should try to notify them.

>         I thought I had figured out that anything generated with
> this software was dead on arrival, but I just got the shock of my
> life, yesterday.
>         There is a web site operated by one of the Oklahoma City
> television stations whose url is
> if anybody wants to try it out.  I have always thought it worked
> quite well with lynx and I have used it on occasion for several
> years to look up things like the Oklahoma City Bombing trial
> transcripts and the Starr Report on Bill Clinton's activities as
> well as lighter fair.
>         I have always been able to read anything on that site
> that was text to start with and I always knew what link was
> highlighted, etc.  In other words, it works as well as one would
> expect a very busy web site to work.
>         Curious about what they use to generate their web pages,
> I used the lynx -source flag to download the html so I could read
> it.
>         I was shocked to see that it was generated with Microsoft
> Frontpage 2.0 or something like that.
>         The problem appears to be that many sites use the
> "onclick" javascript mechanism to select links and lynx can
> neither understand nor generate responses to javascript.
>         Is there some sort of flag or switch in MS Frontpage that
> controls whether mouse clicks or some other device is used?
>         It seems that a number of sites would probably work quite
> well if the older system was used for selecting links.  It is my
> understanding that both methods can coexist so there doesn't need
> to be a problem.

  Yes, both methods can coexist (I guess directly specifying <a href=URL
onclick="some_handler()" will cause javascript-enabled browsers to use
javascript, and non-enabled will use href=URL directly).
>         The bottom line is that I don't know enough about this
> subject to even know what I am looking for, but if we can find
> out what it is, it might be possible to head off a lot of trouble
> by just telling all the web teams and designers around campus to
> be sure to do this or not to do that.
>         One non working site was made to work when one of the
> programmers gave me another link to the material which bypassed
> the home page.  It worked like a charm and I read everything that
> was there.
>         Another dead site has no alternative link that I know of
> so it is still death by javascript for that site.
>         Assuming that there is a particular mechanism or the
> abuse there of that is involved, does the w3c site say anything
> about it?
>         Knowing what "it" is would make it easier to search for,
> also.
>       For the lynx list;  I know, I could read all the code,
> but does the client simply transmit the url of the highlighted
> link back to the server when one hits Enter?

 Yes, the URL of the highlighted link is passed to the server in the request,
and server returns that document.

>       On a lot of sites, the real problem is a breakdown of the
> ability to read valid links and select one of them.  It is my
> understanding that javascript that causes images to move and does
> similar graphical things is totally ignored by lynx so it does
> not hurt anything.  It is what is missing that breaks
> everything.

  As others said, it would be nice to have javascript-filtering proxy. I think
this is nearly impossible thing to implement (it's 30 times more difficult
than just writing javascript interpreter). I afraid that the only workable
solution would be using scripts tuned for that particular site, that will 
substitute those "onscript" with href=URL in all links, and probably do some 
other things). Such functionality doesn't exist in lynx yet.
> Martin McCormick
> ; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to address@hidden

 Best regards,

; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to address@hidden

reply via email to

[Prev in Thread] Current Thread [Next in Thread]