[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bye bye dvicopy+ ?
- To: PEB@dmumpiwh.bitnet, andrew.trevorrow@anu.edu.au, math-font-discuss@cogs.susx.ac.uk, rokicki@cs.stanford.edu, zlatuska@glingol.ics.muni.cz
- Subject: bye bye dvicopy+ ?
- From: ls@mathp7.jussieu.fr (Larry Siebenmann)
- Date: Mon, 4 Apr 94 06:24:36 +0200
Dear Jiri,
Oversight indeed. It seems that font change and
"vertical moves" both zero drift. Do you agree?
Under drivers that do not keep drift near zero, this
invalidates dvicopy+ for any virtual font that uses more
than one raw font or makes use of vertical moves (vskips).
My envisaged use of dvicopy+ for archived dvi's uses
only one raw font per virtual font. But it did explicitly
envisage vertical adjustment of accents. Bang. dvicopy+
dies (!?).
Andy Trevorrow is perhaps the last person to analyse
this problem closely. I wonder what he thinks of this mess.
Standard atomic fonts remain promising for
standardizing the use of vf's in the non-PS TeX world. It
is the atomic .dvi notion that is in trouble.
Yours
Laurent Siebenmann
PS This correction crossed with some first reactions of
Jiri; on which more later. One important point I was quite
unaware of is that Peter Breitenlohner has a sheme for
overcoming the drift problem...