[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

reproducible format



This has reached pin-dancing. Apparently academia needs a reproducible standard for algorithm proofs. Fine, but contemporary hardware differs so much in the details that any completely reproducible implementation will in fact be a software library with performance to match.

Consequently any sensible hardware vendor will ignore the reproducible standard and wait for an open source implementation of the library. There will of course be great hoorah when the library implementation discovers ambiguities in the English text, but the tempest in the teapot will be ignored not only by the hardware vendors but also by those practitioners who are not standards lawyers, i.e. all but perhaps 50 people in the world.

Given this practical and political reality, the committee could avoid a great deal of trouble by simply producing that open-source library itself and blessing it as the standard.

754 | revision | FAQ | references | list archive