[Top][All Lists]

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

Re: LYNX-DEV problem page

From: Laura Eaves
Subject: Re: LYNX-DEV problem page
Date: Fri, 9 May 1997 20:00:39 -0400 (EDT)

> Date: Fri, 9 May 1997 17:42:05 -0400 (EDT)
> From: John <address@hidden>
> Yeah, I do also, and my point was (which I probably did not
> clearly state), that even when they do respond the usual
> message is something like 'why don't you upgrade your
> browser, you dumb fool' ... or whatever ...

Another reason to "upgrade" lynx...

> OK, here's the problem, and this is not a flame or even a
> spark ... Some of us work with old machines, in my case a
> 286-16 (upgraded from a 286-12).  Now with that I can surff
> the net to my heart's content, but what I had to do in order
> to do that is make a trip to waldenbooks and spend $29.95
> and a few hours reading. 

The more user-friendly a browser is the less background reading you need to
do to get started.  I actually considered buying a book when I first started,
but fortunately found help screens and web pages right away that described
pretty much all I needed to know.

> > And what's wrong with that?
> OK, What's wrong with that, is that IMHO, it would be more
> productive to teach people how to write proper code rather
> than to try to skirt around it, or try to 'emulate' the
> folks on the West Coast...

That brings us back to the point that a browser isn't used primarily as an
HTML validator.  As for lynx, when bad html is encountered, all it does is
print a one-second "Bad HTML -- use -trace" message on the status line.
(Sometimes it doesn't even do that.)
If you're a web author and do use -trace, the output isn't very halpful as
it contains mostly stuff that is of interest only to lynx developers.
As for emulating the stuff from the west coast, people do use Netscape and
MSIE to "validate" their html.  So it doesn't hurt to at least handle some
of the same non-html.

> ... Not all of us have eveything up
> and running properly nor permissions correcty made to all of
> our files, for example: 
> -------------------------
>       S.G. Sport long, full skirt has an all elastic waist for ease in
>    movement, a 71" sweep, a straight hem and two side pockets. Available
> -------------------------
> OK ... see what I mean?

No, actually -- this site was "up and running", and the page was generated
from a script that accessed their product database -- so all pages generated
by this script had the same bad html.  This was one of the pages
with the infamous form bug -- bad html that cuased lynx to close forms
prematurely...  Many websites display this bug, inlcuding some popular ones,
like lycos custom search.
(I should mention that I did report the above page to their webmaster, and they
have since fixed it, although it took them 2 months.  They again said I should
upgrade my browser.  I still have the bad page downloaded for lynx testing
purposes.  (Perhaps I should remove their name from it in case someone
complains about copyright...  I haven't publicized it for general use...)
Both Fote's and Klaus' versions of lynx handle the form bug, but
lynx2.7.1 and previous do not.)

> > Putting error recovery into lynx doesn't affect viewing of
> > pages with legal html. 
> I disagree, It compounds the problem because many users use
> many different platforms, and the so-called error recovery
> bug fix only lets people get into places where they should
> not be!

This doesn't make sense.  Where does lynx go that it shouldn't go
as the result of error recovery?

> Peace ... BYE ... John
> All flames are welcome ... It's cold here !

flame off :)

; To UNSUBSCRIBE:  Send a mail message to address@hidden
;                  with "unsubscribe lynx-dev" (without the
;                  quotation marks) on a line by itself.

reply via email to

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