This is an informal, readable description of the current state of the draft FORTRAN 8x language standard. The authors say it is written as a textbook, but it lacks significant programming examples and follows a feature-by-feature layout rather than any pedagogic one. It seems technically complete, but one must seriously question its arrangement and usefulness. As for usefulness, since it does not describe an approved standard, it cannot be used for any educational or planning purpose, as the features described may be altered before the final standard is approved, assuming it is. The only likely audience, therefore, comprises those interested in the standardization process itself. As for arrangement, bearing in mind the limited audience, who are almost certainly knowledgeable FORTRAN programmers, the most useful arrangement would surely be one that highlighted the changes being contemplated in the draft. That is what standard-watchers are interested in, and they certainly do not wish to wade through an entire book to find the proposed modifications. (If they wanted to do that they would probably choose to read the draft itself.) Unfortunately, the differences between this draft and FORTRAN 77 are not stated anywhere.

It seems that the authors have confused their own interests as authors with the interests of their readers. They offer as reasons for their arrangement that it gives a complete view of the language, and that it enables them to easily produce future editions corresponding to further drafts and finally to the anticipated new standard. One feels that the second reason is the real one, yet in fact this is little more than a plea that the reader accept an inferior arrangement because it will make life easier for the authors when they write another book in the future. It does not seem to have occurred to them that when they do put out their next edition, the reader will have to read it all again from cover to cover even if only one or two points have been altered. A book about a work in progress is by nature quite different from one about an established standard, and if the authors intend to write both books, each should be given the particular treatment it deserves. Since the authors are among the designers of the new FORTRAN, it is perhaps not out of place to offer the opinion that the language being described seems to have lost its direction as well; its grotesque syntax changes are unlikely to appeal to the FORTRAN community, who might well decide upon reading this book that it is time to switch to C.

The presentation is reasonable, although the limited number of fonts fall down badly in the book’s few mathematical formulae, some of which are well-nigh indecipherable. The text is generally good, but a bit sloppy in parts. For example, “Any comment . . . may include lower- and upper-case letters.” Old FORTRAN hands will understand, but a reader treating this “just as if it were a totally new” language might wonder why he or she cannot have digits, commas, or other symbols in a comment. Finally, since this is envisaged as one of a series of books describing successive drafts, one would expect the official designation or date of the draft standard it describes to be mentioned in a very prominent place. It isn’t. If you read chapter 1, though, you will find that the time of writing of the book is mentioned. This is a wayward and deficient book that will satisfy very few readers.