From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,99ab4bb580fc34cd X-Google-Attributes: gid103376,public From: dewar@cs.nyu.edu (Robert Dewar) Subject: Re: Q: access to subprogram Date: 1996/07/15 Message-ID: #1/1 X-Deja-AN: 168543432 references: <4rb9dp$qe6@news1.delphi.com> <4rr5tu$sap@watnews1.watson.ibm.com> organization: Courant Institute of Mathematical Sciences newsgroups: comp.lang.ada Date: 1996-07-15T00:00:00+00:00 List-Id: Jon said "At the risk of starting another branch in this goofy thread, you are quite correct here in that to some extent this aspect of the discussion has been a case of "arguing in a vacuum. I suppose at this point Jay Martin would start blow torching academics for this, but it is curious that this sort of thing is not given much attention despite the fact that it does have real world impact and would appear to be very important." Actually this data DOES exist, but it tends to be proprietary. For example, IBM has *extensive* PL/1 execution traces that would certainly yield this information for a large set of existing PL/1 applications. Similarly you will find that Ada companies have this kind of informatoin available, but do not necessarily share it. I would not blow torch academics here. The trouble is that to get this information, you need widespread acccess to large real-world applications. Such access is usually only available to large companies, and not to universities, and measuremens on small non-typical programs may be worse than useless.