I'm still using Microsoft's
Publisher (97 & 98) to generate some HTML code, but I've struck out
on my own and am writing quite a bit of the code. ("The best html-writing
program is Notepad.") Also I've started using html generators included
in many latest-version applications: Publisher 98, Netscape Communicator
4.5, Lotus WordPro 97, etc.
Some of the early problems
are solved. Microsoft's Internet Explorer doesn't go to a name anchor,
<a NAME=name></a>, properly if it's located within a table.
(Netscape Navigator navigates nicely.) The strange patterns in the
water and sky are peculiarities of the .gif format. (Wayne Fulton's
web page is outstanding for explaining these peculiarities. GO
THERE) Microsoft tends to clutter its html code with really weird,
extraneous tags which, like dead teeth, should be carefully extracted.
I have a certain morbid fascination for some of this cluttered code and
have let it stay. Eventually, I'll clean it up, but it's been interesting
trying to figure out the "rationality" that underlies it.
I haven't yet found how to
invoke some sort of "password" function in HTML. (My ISP doesn't
allow cgi-bin's at my level of service.) We need that to create
"Adventure cave" structures. The random-dot stereogram and quilt-pattern
hyperlink hot spot tricks work well, but anyone can, of course, keep an
eye on the browser's active-link line to tell where and what the hot spots
are. As long as I'm revealing some of the routes to clues here, be
it known that there are clues in the source code as "<!--" lines.
HELP!! HELP!! In my
opinion, the bouncing ball problem has as great a potential for helping
learners understand, usefully understand, physics as anything we
can use. I believe that a simple(?) applet is needed. Something
that would let a person apply a digital force on a digital mass and watch
the digital acceleration. A graphics object would be assigned a mass
and a force on a joy stick would be interpreted as a force on that object.
Variable friction should be allowed, but the limit to no friction is an
essential option. The task would be to follow a route on the screen.
Possibly race over that route.