Okay, folks, it's standards update time again. I'm still jet lagged from several grueling weeks of travel. In late June, I convened the X3J11/WG14 (C) meeting in Amsterdam. Got home for the week of the Fourth, which also included a quick overnight trip to Washington DC. No sooner had I readjusted to Eastern Standard Time but I had to go back to Europe for the X3J16/WG21 meeting (C++) meeting in Stockholm. At the time this schedule was set up, I didn't think it would affect me. Had I But Known, I would have used my limited powers as WG14 Convener to make it more humane.
In any event, the C meeting went smoothly, as it usually does. About 20 attendees, representing six countries, were present. Most are old friends by now and work smoothly together. The committee looks to be on schedule to have the technical particulars of C9X decided by the end of 1996. A formal standard should still occur in time to make that 9X an honest designation.
Without going into a lot of detail, I'll simply say that C9X looks to be the Standard C you've come to know and love with a number of features added primarily for numerical programming. That's largely because the Numerical C Extensions Group was for several years the principal clearing house for extensions to Standard C. The C committee is very conservative, turning down essentially all proposals for which there is no existing practice somewhere. The sad thing, to me, is that any hope of adding class extensions to C, a la C++, is probably dead. The principal champion for this ambitious addition, Bob Jervis at Sun Microsystems, is now caught up in the Java frenzy and can no longer attend meetings.
The C++ meeting went fairly smoothly too, at least through Thursday evening when I left. About 50 attendees, also representing six countries, were present. The goal of this meeting was to complete the changes in response to the first Committee Draft ballot, and to vote out the draft for a second such ballot. Actually, it was the goal of the previous meeting to do so -- the March meeting in Santa Cruz couldn't quite handle all the issues in time, however.
As far as I know, the committee didn't actually vote out the draft this time. A few issues remain to be handled. More important, several national bodies wanted time to review the updated draft before sending it up the line to ISO. But the committee did commit to a feature freeze. If they hold to their promise to themselves, then Standard C++ may actually be pretty well nailed down by now.
Pete Becker and Dan Saks were also at the meeting, so you should be reading more detail in these pages, from various perspectives, in the coming months. Do stay tuned.
P.J. Plauger
Senior Editor