>The idea of showing a circled portion of the original document is appealing
>but doesn't work when you can't say where a line break will occur or which
>words will be under each other. Just change the window size and it all
>changes.
I probably didn't make it clear in my message but we are aware of this
issue and are looking into ways of "circumventing" it. Even with the
possible limitations that might result, we still think a WYSIWYG
drawing tool on stacked layers would be an appealing way to do
annotations.
>This could be done by inserting the appropriate markup into document,
>in the same way that HTML+ handles change bars. The PANEL tag can be
>held separately, but relies on there
This looks promising. I don't see anything on change bars in my
version of the html+ document. I'll get the newest version.
>I think that a role attribute would clarify the use of the panel element
>e.g. role=annotation.
>We could extend the FIGT element to have an IDREF which names the figure
>to which it applies. This would allow these to be kept separately as well.
I think both of these would be useful.
Thanks for the comments.
Jay Glicksman