With the current Unisys/Compu$erve lock on .gif format, we
are going to see a LOT of variation in inlined images, of
whatever style, that use the LZW compression algorithm.
I'd suggest that the WebStep group allow the .gif and .xbm
which are currently in such widespread use, but promulgate
jpg compressed tiffs as a replacement standard.
> For browsers/clients:
> * Must be able to "open" .htmld by tacking on
> "index.html", not by displaying the .htmld as a directory
> listing.
What about htmld's in URLs?
> How interested are y'all in actually defining a standard
> icon or look? If this is possibly going to be standardized,
> why don't we go even further and define a standard set
> of dwrites for
> Live link color
> "Been there" color
> TagFonts -- CODE, BLOCKQUOTE, ADDRESS, Hn
> TagRulers -- code in 1/4 inch tabs, centered H6s
IF dwrites are to be defined, they should be carefully
checked against current usages, and GLOBAL settings
encouraged.
> ... I, too, would like to move the WebStep community past
> GIF, but that's not a practical vision for the Web as a
> whole. Also, I doubt jpeg is the right answer for imaged
> text (even in lossless mode).
I agree that raw jpg/jpeg are not, but the tiff spec has
plenty of room for specification of overlay text.
> Inline jpeg should be supported, as well we should keep
> a close eye on the gif-24 replacement that is currently
> being worked out in comp.graphics and elsewhere.
Also has anyone seen anything about the ".GEF" format
which uses lzh compression in lieu of lzw? Legals
about it?... Availability of the standard?
(Ref TeleGrafix Communications Inc., Huntington Beach, CA,
and the RiPscrip drawing "standard" - comparable to PostScript
plus some live rendering (buttons/popups/radio/selection)
PLUS pseudo-inline bitmapped images. Existing DOS/Mac-OS/2
apps handle inlined jpeg, (gif removed), sound, Adobe(tm) type-1
and TrueType(tm) fonts, BMP/DIB-device independent bitmaps,
contact address <PATCLAWSON@delphi.com>)
> 4) a standard DO-based server object ([[NXApp webspace]
> getIconForURL:])
Where the "get" may launch a Mach request at a proxy
or gateway server!
> 8) Define filter services that actually "process" urls;
> take a WebSTEP url type and return HTML or formatted RTFD or ...
This one gets sticky unless the Text object handling
the rtfd can understand URLs, as well.
> The DOS filename mapping would now be .HTM and .HTD, per
> Scott's question of cross-platform compatibility.
> Any debate? I don't mean to sound dogmatic...
Can anyone speak to VMS compatibility. From what
I've seen the 8+3 conventions seem to all too often
carry over to VAX/VMS - and what I've seen is too
small a sample to ensure my comment.
> I'd like to modify NCSA and CERN to look for .htmld as an
> option if ..html is not found for a file..
A better answer, if you ask me! Going the other route
we could all end up with W4WG files in place of HTML
Bruce