gillies@m.cs.uiuc.edu (08/11/89)
I'm using "ptroff" on NeXT to generate postscript output from troff.
Unfortunately, there is a problem with the symbol font. All the greek
(upper & lower case) characters and numerals comes out in a normal
face. However, the standard appearance in troff is italic for these
characters. I have two questions:
(1) Does adobe plan to fix this bug in their driver? Is there some
way I can fix the driver by changing the data files? I doubt we have
a source license with a vanilla NeXT box.
(2) Probably the simplest fix is to change the psdit.pro files to CONS
up a new font with italic greek and numerals, and then replace symbol
temporarily on the server. Can someone suggest a few Postscript
incantations to accomplish this, so I could add it to our psdit.pro
file? Is this even possible?
Any help is greatly appreciated.
Don Gillies, Dept. of Computer Science, University of Illinois
1304 W. Springfield, Urbana, Ill 61801
ARPA: gillies@cs.uiuc.edu UUCP: {uunet,harvard}!uiucdcs!gilliesrcd@ico.ISC.COM (Dick Dunn) (08/12/89)
In article <36700007@m.cs.uiuc.edu>, gillies@m.cs.uiuc.edu writes: > I'm using "ptroff" on NeXT to generate postscript output from troff. > Unfortunately, there is a problem with the symbol font. All the greek > (upper & lower case) characters and numerals comes out in a normal > face... It is an upright face, yes. > (1) Does adobe plan to fix this bug in their driver? It is not a "bug in their driver". It doesn't have anything to do with a driver; it's the way the symbol font was designed. I don't like it either, simply because (as you noted) the conventional style of Greek for use with mathematics is slanted. (I hesitate to say "italic"--somehow "italic Greek" doesn't sound quite right.:-) > (2) Probably the simplest fix is to change the psdit.pro files to CONS > up a new font with italic greek and numerals, and then replace symbol > temporarily on the server... > ...Is this even possible? It is possible to cobble something together, but it's not easy. You can modify the font transformation to use the existing outlines but slant them a bit. However, note that you want to do this *only* for the Greek alphabet--you don't want slanted mathematical symbols. This means you need to split out references to the symbol font according to whether they're Greek or not, or else have two symbol fonts. I played around with this a while back and, for a test case, I used a font transformation matrix of [24 0 6.6 24 -2 0] (obviously to test it in 24-pt; you'll need to rescale). The 6.6/24 was a roundish number that fairly closely matches the italic angle of Times Italic. The -2 was a bit of italic adjustment I was using in the test. Since eqn does italic adjustment on its own, better play with mixtures of text to see what adjustment, if any, is needed. This will be a lot easier to figure out (and more useful for the future) if you're using a recent troff--the DWB version usually called ditroff. This doesn't really give the font you want--slanting a font is not the same as making an italic version--but it's less jarring. I wish that Adobe would produce another math-symbol font with a Greek alphabet specifically designed slanted. -- Dick Dunn rcd@ico.isc.com uucp: {ncar,nbires}!ico!rcd (303)449-2870 ...Are you making this up as you go along?
ath@helios.prosys.se (Anders Thulin) (08/14/89)
In article <16009@vail.ICO.ISC.COM> rcd@ico.ISC.COM (Dick Dunn) writes: >I wish that Adobe >would produce another math-symbol font with a Greek alphabet specifically >designed slanted. Lucida-MathItalic (yes, from Adobe) looks fine to me. -- Anders Thulin, Programsystem AB, Teknikringen 2A, S-583 30 Linkoping, Sweden ath@prosys.se {uunet,mcvax}!sunic!prosys!ath
gillies@m.cs.uiuc.edu (08/15/89)
>>I wish that Adobe >>would produce another math-symbol font with a Greek alphabet specifically >>designed slanted. > >Lucida-MathItalic (yes, from Adobe) looks fine to me. > Why doesn't this font come with the transcript software? Adobe, are you listening? Don Gillies, Dept. of Computer Science, University of Illinois 1304 W. Springfield, Urbana, Ill 61801 ARPA: gillies@cs.uiuc.edu UUCP: {uunet,harvard}!uiucdcs!gillies
bobm@agsm.unsw.oz (Robert Marks) (08/29/89)
In article <16009@vail.ICO.ISC.COM> rcd@ico.ISC.COM (Dick Dunn) writes: >I wish that Adobe >would produce another math-symbol font with a Greek alphabet specifically >designed slanted. Well, Adobe needn't jump -- It's already possible, using the \S troff function. For instance, here at the Australian Journal of Management, we use \S, \H and the define function of eqn to automatically get oblique, synthetic Greek letters in equations. For instance, to get oblique beta, we define: define beta % "\S'+15'\s-1\H'+1'\(*b\H'0'\s+1\S'0'\h'0.2n'" % Then beta inside .EQ/.EN or eqn delimiters will be printed with a 15 degree slant, so it's really oblique rather than italic, but the change in the aspect ratio I suggest above improves the appearance, I believe. I can post all our eqn definitions if there's a demand. [Posted again in case the first item didn't escape.] Robert MARKS, Australian Graduate School of Management, University of New South Wales, PO Box 1, Kensington, NSW 2033, Australia. Phone: +61 2 662-0271 Fax: +61 2 662-2451 Internet: bobm%agsm.unsw.oz.au JANET: agsm.unsw.oz.au!bobm@ukc BITNET: bobm%agsm.unsw.oz%uunet.uu.net@WISCVM BITNET: bobm%agsm.unsw.oz%uunet.uu.net@HARVUNXT CompuServe: >internet:bobm@agsm.unsw.oz.au UUCP: {uunet,ukc,mcvax,ubc-cs,nttlab}!munnari!agsm.unsw.oz!bobm
brown@astroatc.UUCP (Vidiot) (09/05/89)
In article <678@agsm.unsw.oz> bobm@agsm.unsw.oz (Robert Marks) writes: <In article <16009@vail.ICO.ISC.COM> rcd@ico.ISC.COM (Dick Dunn) writes: <>I wish that Adobe <>would produce another math-symbol font with a Greek alphabet specifically <>designed slanted. < <Well, Adobe needn't jump -- <It's already possible, using the \S troff function. For instance, <here at the Australian Journal of Management, we use \S, \H and the <define function of eqn to automatically get oblique, synthetic Greek <letters in equations. For instance, to get oblique beta, we define: < <define beta % "\S'+15'\s-1\H'+1'\(*b\H'0'\s+1\S'0'\h'0.2n'" % Sorry, but \S and \H do not exist in the generic 4.3BSD troff. Maybe you have ditroff, or someone's enhanced troff, but otherwise it does not exist. Reference 4.3BSD USD manual, page 24-6. -- harvard\ att!nicmad\ Vidiot ucbvax!uwvax..........!astroatc!brown rutgers/ decvax!nicmad/ ARPA/INTERNET: brown%astroatc.UUCP@spool.cs.wisc.edu
sakkinen@tukki.jyu.fi (Markku Sakkinen) (09/05/89)
In article <678@agsm.unsw.oz> bobm@agsm.unsw.oz (Robert Marks) writes: > [...] >Well, Adobe needn't jump -- >It's already possible, using the \S troff function. For instance, >here at the Australian Journal of Management, we use \S, \H and the > [...] >I can post all our eqn definitions if there's a demand. Yes, please! The eqn definitions as they are in DWB 2.0 are unusable with PostScript postprocessors and printers even generally: a lot of special characters become bad or unrecognisable. Well, that is not the only problem with DWB 2.0 - I wonder when AT&T plans to distribute an up-to-date, production-quality release. (DWB = Documenter's Workbench (tm), includes device-independent Troff, preprocessors and some associated stuff.) Markku Sakkinen Department of Computer Science University of Jyvaskyla (a's with umlauts) Seminaarinkatu 15 SF-40100 Jyvaskyla (umlauts again) Finland