W3C home > Mailing lists > Public > w3c-wai-er-ig@w3.org > August 1999

Bobby non-handling of "*.shtml"

From: Harvey Bingham <hbingham@acm.org>
Date: Fri, 20 Aug 1999 14:36:16 -0400
Message-Id: <4.1.19990820143555.04cbccf0@pop.tiac.net>
Message-Id: <4.1.19990820143555.04cbccf0@pop.tiac.net>
To: <bobby@cast.org>, <w3c-wai-er-ig@w3.org>, <w3c-wai-au@w3.org>
Trying to submit following URL to the public version of Bobby,

    www.cast.org/bobby

I get message:

    Can't connect to: http://www.esperanza-lutheran.org

I question this message, that led me astray. I have found the problem.
That URL gets appended before the suffix with file type "shtml":
 
    /index.shtml

I can run that URL against the downloaded version of Bobby 3.1.1 build 6.

From that local run, I see that all 36 files on that site have filetype 
*.shtml. I know not why. Glad I could run the local version. I believe 
there is a better warning message than just 

     Can't connect to: http://www.esperanza-lutheran.org.

You should certainly be able to recognize the "*.shtml" or I suppose
"*.sht" if that is a 3-character surrogate, like "*.htm" is for "*.html".
Possibly respond with:

     Cannot analyze file with privacy suffix "*.shtml"

Or even less specific:

     Can only analyze HTML files with a case-insensitive suffix: 
     ".htm", or ".HTML".

Suggest that authoring tools make more explicit when files should be 
marked private. [Microsoft FrontPage 4.0 was the authoring tool credited
with causing this problem. I don't use it so don't know what control
the author has.]

Suggest that user agents recognize this bar to display, sometimes
opened by a request for a password, though not in the cited case.
Is there other metadata that prevents analysis by Bobby?
This may be a significant bar to public checking of the corporations
who try to protect themselves by "*.shtml".

Regards/Harvey Bingham 
Received on Friday, 20 August 1999 14:39:46 GMT

This archive was generated by hypermail 2.2.0 + w3c-0.30 : Thursday, 9 June 2005 12:10:33 GMT