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=0.2 required=5.0 tests=BAYES_00,INVALID_MSGID, REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,39e272d357c68416 X-Google-Attributes: gid103376,public From: kilgallen@eisner.decus.org (Larry Kilgallen) Subject: Re: Is Apex dead as an environment for Ada & Java? Date: 1999/12/01 Message-ID: <1999Dec1.072849.1@eisner>#1/1 X-Deja-AN: 555268907 References: <11f733ec.57d88b68@usw-ex0107-042.remarq.com> <384450A7.CAF2722F@averstar.com> <00844d40.d48fa2f6@usw-ex0107-043.remarq.com> X-Trace: news.decus.org 944051332 4583 KILGALLEN [216.44.122.34] Organization: LJK Software Reply-To: Kilgallen@eisner.decus.org.nospam Newsgroups: comp.lang.ada Date: 1999-12-01T00:00:00+00:00 List-Id: In article <00844d40.d48fa2f6@usw-ex0107-043.remarq.com>, jim_snead writes: > If this is alluding to the problem of brittle libraries in > Ada 83 code, I can see why someone might be gun-shy of having > a large subsystem. I have heard many stories of "ada.lib" > libraries in the traditional sense that can easily get > out of sync or get corrupted and lead to massive rebuilds. > If you spread ada.lib's around, not as much can get damaged > at once. A clever deceit on Rational's part but I do not buy it. > > What I rather do like is the idea of source based libraries. > Apparently more Ada 95 compiler vendors are turning to > source-based libraries, and they have a reputation of being > very stable. > Therefore I conclude that a lighter-weight directory structure > may be more useful for future growth. Other merits of various library structures aside, the notion that Ada implementors should avoid traditional libraries because they cannot build a reliable one is a sad statement indeed. "Careful write" strategies can protect a library against operating system crashes, so I see no excuse why correct library code should not result when an Ada compiler implementor attempts such a feature. Larry Kilgallen