laser-lovers@uw-beaver (05/17/85)
From: Les Earnest <LES@SU-AI.ARPA> After Brian Reid's generous offer last Saturday to provide print samples from his Laserwriter, I dispatched a self-addressed envelope to Brian on Monday via special courier (he is just 100 yards away). So far I have not gotten anything back. Yesterday I managed to get some current Laserwriter samples from another source. Let me say that on the whole I was very pleasantly surprised.* The letter spacing was much less glitchy than earlier samples I had seen. There were a few odd lumps and bumps on certain characters at certain sizes, but on the whole the letter spacing was rather good. Given that the samples were in a widely used typeface (Times Roman), I also obtained a sample of the same text in the same font in the same point size (10 pont) from another laser printer. One thing that struck me immediately was that these were quite different fonts! While they bore a strong family resemblance to each other, one had relatively small openings in letters such as "o" and "e", while the other was more open. The character spacings were also quite different. For example, a line of text that was 5.5 inches wide on one machine was 6.2 inches wide on the other. This contrast gave me an idea. A few years back I used to hold weekly winetastings in my home for upwards of 20 people. We managed to drink our way around the world several times, focussing mainly on European and California wines. We sometimes collected rankings of wines from all present in order to measure the degree of consensus. Just for fun I wrote a computer program to instantly analyze these data. What we learned right away was that while there were often strong disagreements about the rankings of wines by the group, there were often correlations between the rankings of individuals. In other words, if two people agreed about the relative goodness of one wine they were quite likely to agree about others. This seemed to confirm certain theories about genetic differences in taste. Some tastings were run "blind" -- that is, the wines were decanted into containers marked with single letters and were tasted and voted on using secret ballots, so that no one would be influenced either by the name on the label or by the opinions of their colleagues. We noticed that "name" wines often faired less well in these tastings than when their labels were visible. Anyway, it occurred to me that it would be interesting to run a blind "tasting" of fonts. Given that there is a suitable forum here today, I am going to try it. On most Fridays there is an informal meeting here at Stanford, hosted by Don Knuth, called the "Metafont for Lunch Bunch." There are usually several heavy-duty typographers present, several others who are skilled in the art, and a smattering of computerniks with interests in typography but relatively little background. I plan to spring this tasting on them, but with the request that anyone who can identify the source of either print sample disqualify him/herself. Someone will report the results. Cheers, Les Earnest Notes: * For the record, I own no stock in either Apple or Adobe.