bobal@microsoft.UUCP (Bob Allison) (08/18/88)
Well, I was going to post a note anyway, indicating what I thought had happened at the last X3J3 meeting and this seems like a good opportunity. (The references talk about how deprecation could invalidate a lot of programs). The meeting started with eleven (!!!) language proposals, most of which were serious reductions in the language. By Wednesday there were four, by Friday, three or four depending on your point of view. None of these proposals contained the concept of deprecation (obsolescence is in some of the proposals). However, the proposals all remove the three lines in the text which refer to deprecation: none of them have attempted to integrate new features with old features (i.e. allow structures in COMMON and EQUIVALENCE stmts). However, I have high hopes that this will happen, and some efforts are going on independent of the proposals to figure out how to do this. This is not cause for unfounded relief: obsolescence is still in, and there is nothing preventing the 9X committee from putting COMMON and EQUIVALENCE on the obsolescent list (which is what deprecation was warning you might happen). We have simply eliminated the overt notice to the public of the possibility of COMMON and EQUIVALENCE being deprecated. I believe there are still some on the committee who were perfectly willing to remove the text since it did not require any change in their philosophy of language evolution. So, basically, the only thing that has changed is that users will not be warned before something is added to the obsolescent list, which was not the point of the users complaints. It is yet to be seen whether the concept of language evolution (which is what I believe most people were complaining about) has been eliminated. Aside: many people agree that FORTRAN 77 has mistakes (such as REAL DO loop variables and alternate returns); however, providing a mechanism for correcting those mistakes opens the door to removing anything and everything from the language (FORTRAN has a lot of mistakes, and some people believe COMMON is one of them). This is long enough. I'll try to indicate a little more of what happened in another message. Bob Allison