Re: Downloading htmlds

Rohit Khare (khare)
Fri, 10 Feb 1995 13:10:28 -0800


Sounds like there was a misconception in Bruce's comments that I'd like to
clarify [it may or may not, however, get to the heart of his concerns, though]

.htmlds should not contain other .htmlds

It's not banned, because you may want to include such a document in the
user-content, jsut like dragging in an Improve worksheet, but authors and
authoring tools should NOT generate nested HTMLs. Subpart realtionships should
be maintained as links between atomic .htmlds.

Also, my original comments about downloading are more closely tied to "View
Source" options than the "normal" access process. To restate the problem:

After viewing an .htmld over the net normally (deferrable images, etc), a user
may choose to download that document. They're used to hitting View Source and
save to copy that page, but now we have a fairly substantial .htmld package.

Is the answer to 1) have a client request foo.htmld.tgz from the server, which
will automagically tar it up or 2) have WebStep-savvy browsers that create a
local foo.htmld directory and proceeed to copy each of the graphics, htmls,
etc, into it?

Not having such support is the root of Dan Grillo's request for
foo.htmld/foo.html, I think.

Comments?

Rohit