With most microcomputer applications – word processors, spreadsheets, databases – written in C these days, concern about standards for the language extends well beyond the Unix community. The time allowed for public comments on the proposed ANSI X3J11 C programming language standard has just closed, and as it dis, voices of doubt and dissent were being raised: the loudest from Dennis Ritchie, author (along with Brian Kernighan) of the original C language specification team at AT&T Bell Laboratories. Ritchie pointed to several flaws in the document inconsistancies caused by type qualifiers such as ‘const’ and ‘volatile’, and the use of ‘noalias’ pointers, which he said is a licence for the compiler to undertake agressive opimisations that are completely legal by the committee’s rules, but make hash of apparently safe programs; the confused attempt to improve optimisation by pinning a new qualifier on objects spoils the language. Ritchie’s comments were endorsed by Jeffrey Barth, a vice-president at Silicon Valley Software. There are too many unaddressed items, it doesn’t say a thing about the important issues. Barth said the standard lay somewhere in between the very good, very specific Fortran standard document, that has the answer to most questions, and the Pascal standard, which he rates as terrible: but every word has been treated as if its the Bible. Ritchie warns that the ANSI committee is in danger of ratifying a language specification that no-one can possibly embody in a useful compiler. According to Barth, most compiler developers, including Silicon Valley Software, will implement a compile-time flag that when set will conform to the new standard. For a long time the main demand will be for the traditional K&R C specification, he predicted.