At 12:32 PM 2/6/96 -0600, you wrote:
>First, the notion of a HTML-understanding user agent is vague. At
>what level does user-agent end? What if I wrote a script that pulled
>down HTML documents and then tabulating element count statistics. That
>script could be considered a user agent, ...
>Similarly, what about libwww? It understands HTML, but I wouldn't
>want it printf'ing an error message to the console when it sees a
>construct it doesn't understand.
"User agents must render highlighted phrases distinctly from plain text.
Additionally, EM content must be rendered as distinct from STRONG content,
and B content must rendered as distinct from I content."
I don't think that either your script or libwww are user agents because they
do not render anything (and cannot, therefore, render things distinctly.
Since user agents are required to "render" they can use their "rendering
mechanism" to report places that they inferred their rendering from