To: J3 Members J3/17-193r1 From: David Muxworthy & Steve Lionel Subject: Informal name of the revised language Date: 2017 October 19 Discussion: ---------- The following was discussed as a proposal for the UK vote on the second CD. It gained near-unanimous support but as we operate by consensus it was not included in the final vote. This is to bring it to the attention of J3 for information and possible action. Proposal: -------- The language should renamed Fortran 2017 since it has developed incrementally over the past two years and it is no longer true that the final definition took place in 2015. Rationale: --------- We have had: content informal published finalized name 1990 90 1991 1996 95 1997 2004 03 2004 2010 08 2010 A three-year gap is looking too far backwards and would make it look as if the language were out of date already. Fortran already suffers in some quarters from the perception that it exists only as a legacy language and to publish Fortran 2015 in 2018 could even attract derision. Other comments arising during the discussion were to use Fortran 2018, Fortran 2020 (the language for the future) or to use a naming scheme without a year in it. J3 Response ----------- J3 had an extensive discussion on this issue at meeting 214. Everyone agreed that there is a perception problem, but some felt it was too late in the process to change the name. It was noted that past Fortran standards have changed their year number (Fortran 88 became Fortran 8X and then Fortran 90, Fortran 2000 became Fortran 2003), and C++ was calling its revision "C++1X" until publication changed it to C++17. The name "Fortran 2015" is already in widespread use - changing it now could cause confusion, along with some rework by compiler vendors as well as book/paper authors. Among vendor representatives present there were proponents of both sides. A straw vote was held among those in the room. Four votes favored keeping 2015 in the name, five votes favored changing the year number to 2018, and three were undecided. This doesn't give an obvious sense of direction by J3. WG5 will be asked to vote on this issue. What everyone did agree on was that we should immediately start calling the next revision "Fortran 202X".