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=-0.4 required=5.0 tests=AC_FROM_MANY_DOTS,BAYES_00 autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,ac39a12d5faf5b14 X-Google-Attributes: gid103376,public X-Google-ArrivalTime: 2002-04-29 16:34:10 PST Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news.mailgate.org!newsfeed.icl.net!psiuk-p2!psiuk-p3!uknet!psiuk-n!news.pace.co.uk!nh.pace.co.uk!not-for-mail From: "Marin David Condic" Newsgroups: comp.lang.ada Subject: Re: Grace and Maps (was Re: Development process in the Ada community) Date: Mon, 29 Apr 2002 10:57:50 -0400 Organization: Posted on a server owned by Pace Micro Technology plc Message-ID: References: <3CB46975.90408@snafu.de> <4519e058.0204180800.44fac012@posting.google.com> <3CBF0341.8020406@mail.com> <4519e058.0204190529.559a47ae@posting.google.com> <3CC1C6B3.6060306@telepath.com> <3CC21747.5000501@telepath.com> <3CC59ED2.1000803@home.com> <3CC5B286.6FE61551@san.rr.com> <3CC5B9EE.32F3060@san.rr.com> <7vvgadecxc.fsf@vlinux.voxelvision.no> NNTP-Posting-Host: dhcp-200-133.miami.pace.co.uk X-Trace: nh.pace.co.uk 1020092272 6495 136.170.200.133 (29 Apr 2002 14:57:52 GMT) X-Complaints-To: newsmaster@news.cam.pace.co.uk NNTP-Posting-Date: 29 Apr 2002 14:57:52 GMT X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4522.1200 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200 Xref: archiver1.google.com comp.lang.ada:23242 Date: 2002-04-29T14:57:52+00:00 List-Id: I liked the article. I'm not sure I agree that the only two alternatives are to a) build a perfect system that takes forever to get there or b) build a lousy system that gets out quickly. I *do* agree that a lousy system that gets out first and still manages to fill a void is going to generally become the dominant thing and not allow the "perfect" system that's still in the lab to get a toe-hold in the marketplace. But why does "Quick" have to equate to "Lousy"? Couldn't it equate to "Small-And-Incomplete-But-A-Solid-Basis-On-Which-To-Build"? My contention would be that it is better to get some useful piece of the "perfect" solution out there early. When we're talking about the Grace components, this would be some version of "Get a couple of useful data structures out there that more-or-less are accepted as the 'conventional answer' and then build from there." (If Grace.Lists and Grace.Maps got accepted and shipped with compilers, we'd have A Good Start. The interfaces are not weak and the design is not poor - just not the answer to all problems, everywhere.) This is a similar situation to the AdaOS discussion going on in another thread. If the AdaOS project had enough software to build a successful bootstrap and kernel that allowed someone to load one or more programs & execute, then it would constitute A Good Start that would likely mushroom from there. If the RTEMS kernel could be used as a kick-off point, maybe that's a good thing. So what if it doesn't have all the security one might want or that its a realtime scheduler (on which a non-realtime scheduler could be built!) - its there and it works. Get a "kit" built that would let someone compile it with Gnat for a PC and build a bootable floppy. Add on to it so it could load/execute some rudimentary programs from a floppy. At that point, you've got a project going that could be enhanced over time. Maybe RTEMS isn't the right answer - it does present a POSIX interface so its Just Another Realtime OS rather than being something unique and Ada-ish. But if it were used to simply provide the RTK (and some device drivers) for a monolithic realtime Ada program (sounds like an apt description of an OS to me!), it would allow building a more "Ada-ish" OS on top of it. Possibly there are other answers that could be made to work even better. But the point is that getting something out there *NOW* is a whole lot better than waiting around forever for a thousand other things to get done so that a "perfect" system can be eventually built. Get the scope of the project down to something that would be A Good Start & there's no reason that this small piece can't be well designed and well implemented. It doesn't have to be a turd to succeed - it could be a diamond-like jewel. Just make it a *SMALL* jewel that a beginning jeweler can afford to buy and put in a setting. Add more gems later. MDC -- Marin David Condic Senior Software Engineer Pace Micro Technology Americas www.pacemicro.com Enabling the digital revolution e-Mail: marin.condic@pacemicro.com "Ole-Hjalmar Kristensen" wrote in message news:7vvgadecxc.fsf@vlinux.voxelvision.no... > > For an interesting discussion about perfect vs. good enough, see the > link below or do a search on "worse is better".... > > http://www.ai.mit.edu/docs/articles/good-news/subsection3.2.1.html