Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site uw-beaver Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!vax135!cornell!uw-beaver!laser-lovers From: laser-lovers@uw-beaver Newsgroups: fa.laser-lovers Subject: "Oh no" recursion Message-ID: <1156@uw-beaver> Date: Mon, 13-May-85 04:51:09 EDT Article-I.D.: uw-beave.1156 Posted: Mon May 13 04:51:09 1985 Date-Received: Tue, 14-May-85 20:27:23 EDT Sender: daemon@uw-beaver Organization: U of Washington Computer Science Lines: 59 From: Les Earnest As a BWM owner and Imagen shareholder I am naturally interested in Brian Reid's generous offer to provide print samples from his Scribe-Laserwriter system, given in his "Oh no" message of 11 May. [To Brian: will you settle for my sending you a self-addressed Stanford interdepartmental mail envelope or must I really provide a stamped envelope? If the latter, how much postage will be needed?] Brian chastises Paul Rubin for saying about the Laserwriter that "the intercharacter spacing is uneven and sloppy-looking." Brian says, "This has nothing to do with the LaserWriter. . . . there is nothing wrong with the intercharacter spacing on the LaserWriter." Inasmuch as the Postscript software on the printer, the fonts, and presumably the intercharacter spacing tables were all developed by Adobe, this is a bit hard to swallow. Mr. Reid seems to be telling Mr. Rubin that he should not believe his eyes. Brian goes on to say that perhaps the problem is caused by obsolescent software on the Macintosh. This leaves unanswered the question of who is responsible for system inadequacies of this sort and when they are likely to be fixed. The thesis that this problem is Apple's fault would be plausible except for one thing: the print samples developed and distributed by Adobe show the same "uneven and sloppy-looking" intercharacter spacing. A strong clue as to why this has happened appeared in the April 27 message from Ed Taft of Adobe to Laser-Lovers, where he draws a distinction between "geometric scaling" and "typographic scaling." He then remarks that "In PostScript, geometric scaling is the default, because it minimizes the burden on composition programs." Unfortunately, the rounding errors intrinsic in geometric scaling produce the uneven intercharacter spacing that various viewers have seen. It is reasonable to ask why Adobe does not normally do "typographic scaling," whatever Mr. Taft means by that. I would speculate that the reason is that it is hard to do within the neat "transform anything" model on which Postscript is based. Actually, the term "typographic scaling" is rather ambiguous. I know of five different classes of scaling that are in use in modern composition systems and there are likely some more in use in systems that I have not reviewed. Most of them produce much better looking text than geometric scaling. If Brian Reid has succeeded in producing well-spaced text with his Laserwriter, I will be very interested in seeing it and learning more about how it was done. (For what it is worth, a couple of weeks ago I privately offered Brian some print samples of new fonts that I had just filched from Imagen. He responded that he already had all that stuff and that his informant inside Imagen was keeping him well supplied.) Responding to Brian's remark about defensive BMW owners, though I have purchased five BMW cars so far, I decline to defend them totally because my kids keep totalling them. As a consequence, I cannot yet afford to have my own Laserwriter at home, let alone one of the better laser printers. :-) Cheers, Les Earnest