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=0.7 required=5.0 tests=BAYES_00,MSGID_RANDY autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,447bd1cf7a88c198 X-Google-Attributes: gid103376,public X-Google-ArrivalTime: 2001-01-06 09:30:50 PST Path: supernews.google.com!sn-xit-02!supernews.com!216.227.56.88.MISMATCH!telocity-west!TELOCITY!news-out.usenetserver.com!newsfeed.skycache.com!Cidera!news.maxwell.syr.edu!nntp2.deja.com!nnrp1.deja.com!not-for-mail From: Robert Dewar Newsgroups: comp.lang.ada Subject: Re: Do we need "Mission-Critical" software? Was: What to Do? Date: Sat, 06 Jan 2001 17:24:29 GMT Organization: Deja.com Message-ID: <937kc7$ssq$1@nnrp1.deja.com> References: <3A4F5A4A.9ABA2C4F@chicagonet.net> <3A4F759E.A7D63F3F@netwood.net> <3A50ABDF.3A8F6C0D@acm.org> <92qdnn$jfg$1@news.huji.ac.il> <3A50C371.8B7B871@home.com> <3A51EC04.91353CE7@uol.com.br> <3A529C97.2CA4777F@home.com> <3A53CB9E.EA7CF86C@uol.com.br> <3A5466DE.811D43A5@acm.org> <932aol$ikc$1@nnrp1.deja.com> <932mi6$r2k$1@trog.dera.gov.uk> <9343b1$3g5$1@nnrp1.deja.com> <934iuf$eqv$1@nnrp1.deja.com> NNTP-Posting-Host: 205.232.38.14 X-Article-Creation-Date: Sat Jan 06 17:24:29 2001 GMT X-Http-User-Agent: Mozilla/4.61 [en] (OS/2; U) X-Http-Proxy: 1.0 x51.deja.com:80 (Squid/1.1.22) for client 205.232.38.14 X-MyDeja-Info: XMYDJUIDrobert_dewar Xref: supernews.google.com comp.lang.ada:3706 Date: 2001-01-06T17:24:29+00:00 List-Id: In article , Ole-Hjalmar Kristensen wrote: > Btw., c2ada shows that the automatic generation of thin > bindings is indeed feasible, but the c2ada implementation > lacks a lot in terms of stability. Yes, interestingly there seems to be very little demand for working on this kind of tool. At one point Ada Core Technologies had a fairly high priority project to adapt the SGI binding generator (which is far more advanced and robust than c2ada, and also handles C++) for general use (it currently uses a proprietary C++ front end), but there simply has not been any significant demand for this tool, so the priority of the project got reduced, and many more important tasks are ahead of it at this stage. > It seems to me that your position is that lack of bindings to > foreign libraries is not a problem, since you do not see any > demand for tools. I hope you are right, but I do not believe > so. It is a problem, but I don't think it is a blocking problem to the use of Ada in most cases. Through the history of Ada we have had people saying "If only we did X, then Ada would be more widely used" But most of these claims have proved experimentally false, and I think this is another such case. If someone produced a much improved c2ada, I doubt this would suddenly make Ada popular. Indeed, most of the small scale development of which N.Brunot talks about is done in Windows environments, and here the issue is not translation of C headers, it is automatic binding to type libraries. I think the gnatcom approach is FAR more interesting, valuable, and promising in this context, since it allows for automatic reliable generation of thick Ada bindings from language independent specifications (it is quite fortunate that Visual Basic is so much more widely used than C, since it means that Microsoft has to take a language independent approach, and Ada is quite happy to take a place at the table). Sent via Deja.com http://www.deja.com/