Letters Letter to the Editor

The good name of TEX One of the many wonderful things about TEX is that its behaviour is essentially the same, no matter where it runs. TEX is a fixed point, identical on all machines. The same goes for METAFONT and the Computer Modern fonts. The author of TEX, Donald Knuth, has made it perfectly clear that he does not object to anyone revising TEX(or METAFONT) just as long as the resulting program is called something else. However, he also says “nobody is allowed to call a system TEX or METAFONT unless that system conforms 100the TRIP and TRAP tests”. He also asks us “to help enforce these wishes, by putting severe pressure on any person or group who produces any incompatible system and calls it TEX or METAFONT or Computer Modern — no matter how slight the incompatibility might seem”. (Both quotations are from TUGboat, 11(4), p489, reprinted in Knuth’s Collected Papers in Digital Typography). In a recent article in TUGboat (issue 19(4), p366–371), Petr Olsak describes encTEX, a not completely compatible revision to TEX. It seems to me that Olsak has not followed Knuth’s wishes in a consistent manner. Although he calls his new program encTEX, in his article he talks about TEX this and TEX that when he is referring not to Knuth’s TEX, but to his own encTEX. For example, he describes the creation of encTEX the program as a “new compilation of the TEX binary” (p367), and throughout the article he talks of iniTEX when in fact he means iniencTEX. On page 369 he writes “the production version of TEX” when in fact he is referring to his encTEX program. Olsak is tackling a real problem faced by TEX users in his own country, and he deserves credit for this. His solution requires an incompatible revision of TEX the program. If this has to be, it has to be. But more care is required in the documentation.