[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Lynx-dev] Sitemap Warning
From: |
David Woolley |
Subject: |
Re: [Lynx-dev] Sitemap Warning |
Date: |
Thu, 29 May 2008 14:44:37 +0100 |
User-agent: |
Thunderbird 2.0.0.14 (X11/20080421) |
Elizabeth Pittis-Moffitt wrote:
To Whom It May Concern:
This is a peer support public mailing list.
> Re: [Lynx-dev] Sitemap Warning
The site in your signature doesn't appear to have a site map!
I use an Apple PowerBook G4. I installed leopard and my Mac OS X is
Version 10.5.2
I created my Web-site using iWeb 08. Do you support Apple and can you
help me?
Some of the Apple tools, and this appears to be one of them, generate
extremely bad HTML. What they do is use the minimum of constructs to
get the intended visual appearance on typical graphical browsers; they
are not suitable if you intend to view with tools that expect proper
HTML markup. They do not represent the true document structure, which
would allow a real HTML viewer to display them on any imaginable medium.
The result is a very messy display when one actually gets to the right
page with Lynx.
It looks as though the page uses a malformed URL in the meta refresh
element. The specification for meta refresh requires the use of
absolute URLs, whilst Lynx has error recovery for properly formed
relative URLs, there must be something else wrong. As the simplest fix
is to conform to the meta refresh specification, it's not worth my while
to work out exactly why the error recovery is guessing wrongly.
Note that the HTML specification strongly discourages the the use of
refresh for simulating redirects, especially with a timeout of 0. You
should use a proper server redirect. However, as noted above, Apple
don't seem to be interested in producing good HTML.
This is how Lynx perceives the Refresh link:
Linkname: http://www.elizabethparadiserentals.com/
URL: http://www.elizabethparadiserentals.com/
Either it has been confused by the " " after the ":", or the error
recovery is limited to URLs that start with at least one "/".
I am not sure if common search engines even follow meta refresh, but if
they do, they are not guaranteed to use the same error recovery as the
graphical browser you used for testing.
Both factors compromise the use of the page with assistive technology,
for the disabled.
When I do manage to load the real page in Lynx, I notice that most of
the alt text for your images is the rather long URL to the image. That
is simply disruptive to anyone trying to read the page without images,
whether with Lynx, screen readers for the blind, or graphical browsers,
with images off for performance.
I also noticed, in Firefox, that there are rather poor colour contrasts,
particularly involving green.
You have violated basic graphical web browser user interface conventions
by using underlined blue text for things which are not links.
Using "Click here" for every link makes it difficult for assistive
technology users, who will often request a list of links, to avoid
having to trawl through the whole page with a screen reader.
The big picture to the left and above the main title looks like abstract
art. Firefox says it has no alternative text (it probably detected that
it matched the URL) so I can get no clue that way, either.
validator.w3.org finds 30 errors with the homepage proper. The page
claims to use XHTML, but some of these errors would cause a real XHMTL
browser to abort the page (they are well-formedness violation).
It finds 25 errors on the page in your signature, some of which would
abort a real XHTML browser. Note that it doesn't vaildate the attribute
values in meta elements, so it would not pick up the problems that cause
Lynx to not follow the refresh.
--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.