ASCII was an early standard, and ANSI extended it. What Jordan and I need for web pages is now a fairly mature standard that supercedes both. It's called UTF-8, and it's likely to be in use for a long time.
The UTF-8 entry at Wikipedia explains it pretty well, but suffice it to say that ASCII is fully preserved within UTF-8.
Here's an analogy for folks of my vintage: Remember the old IBM Selectric typewriters, and the way you could change fonts by replacing the metal ball? The ASCII character set is like having one of those balls. The ANSI character set is like having an additional one or two, and UTF-8 is like having about a dozen of 'em.
Jordan, if you're using Microsludge's word processor, you can easily convert your document to UTF-8 by saving as TEXT, then specifying UTF-8 instead of ANSI or ASCII as the character set, then check the little box that says "allow character substitution" before hitting the Save button in the dialogue box. I use this all the time to prep documents sent to me for conversion to HTML.
For browsers to display any character set correctly, the character set you're using SHOULD (and sometimes must) be specified in the of the HTML page. So Jordan's issues with that might be solved simply by specifying ASCII as the character set used on a given page. I'm happy to convert to UTF-8 for the sake of consistency, though, and for safety's sake I always specify it in the HTML page header as well.
Robert and Carl are correct that ISO-8859-1 is the default character encoding for most browsers, but I think it's limited to Latin-based languages. UTF-8 replaces that in the XHTML specifications, I believe, and is finding increasing favor because it's more universal.
If I had to choose between Front Page and XyWrite for building web pages, I'd use XyWrite in a heartbeat. Not that XyWrite is so good for the task, but Front Page is THAT bad! It throws in so much extraneous code, it's a complete joke ... which I guess makes it an exemplary product for Microsoft.
(Jordan said):
I still get a bunch of things like "é," left
over (sometimes with #s like 127 included), after I've
run a page of HTML through DELTAGS. I don't know if
it's from ANSI, or just some MS FRONTPAGE detritus
that proved indigestible to the U2 frame. (If this
was ever explained to me before, sorry but I just
don't recall.)
In any case, I can readily turn those left over marks
into the appropriate apostrophes, quote marks, or the
occasional accented character, by means of a few
successive CI requests dropped into my CI template at
the Command Line. Because of that, I'm inclined to
think that these items are fairly standard and
recurrent, and could be coded into DELTAGS.