From ... Path: supernews.google.com!sn-xit-02!sn-xit-03!supernews.com!xfer13.netnews.com!netnews.com!fr.clara.net!small.fr.clara.net!news.tele.dk!148.122.208.68!news2.oke.nextra.no!nextra.com!news01.chello.no!Norway.EU.net!127.0.0.1!nobody From: Erik Naggum Newsgroups: comp.lang.lisp Subject: Re: corba or sockets? Date: 30 Oct 2000 20:31:36 +0000 Organization: Naggum Software; vox: +47 800 35477; gsm: +47 93 256 360; fax: +47 93 270 868; http://naggum.no; http://naggum.net Lines: 39 Message-ID: <3181926696765923@naggum.net> References: <3181895804626114@naggum.net> <8tjujd$oa5$1@reader1.fr.uu.net> <3181904960820128@naggum.net> <8tkf8h$4ka$1@reader1.fr.uu.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: oslo-nntp.eunet.no 972939901 22024 195.0.192.66 (30 Oct 2000 21:05:01 GMT) X-Complaints-To: newsmaster@eunet.no NNTP-Posting-Date: 30 Oct 2000 21:05:01 GMT mail-copies-to: never User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.7 Xref: supernews.google.com comp.lang.lisp:2933 * "Marc Battyani" | This is why I ask you if you still think so if the protocol is a | simple lisp reader friendly text based protocol when 1) the task is | really easy and 2) you master both sides of the communication? The likelihood that novices will do worse than CORBA is still very, very high. Look at how people design their input languages from simple files that are under total programmer control, and often they blow it so bad applications crash and burn, a missed version update causes serious bit rot, and changing your mind about something means you lose a _lot_ of old information. Using Lisp for this does not really help. No networked task is really easy. If it were, you wouldn't be doing it. If you're still doing it, how could that possibly impact any decision on how to design protocols for tasks _worth_ doing? I actually _favor_ Lisp-based protocols, but not because "it's text, so it's easy" is even close to relevant (it isn't -- syntax is an important issue to humans -- it is _not_ important to representing the data), but because it means I have a stable and proven framework to work from. It is actually quite important to realize that just because it's text doesn't make it any less fraught with all the dangers of other network protocols. For one thing, network communications is subject to a failure modes that novices never consider. Even with the best of checksum algorithms, you _may_ get rotten data once in a while. Deadlocks may occur whether you use a hariy binary protocol or a Lisp-based protocol. All these things are unlikely to happen if you run on a local machine, and they don't happen if you work within the same memory. (Unless you run without error correcting memory, of course, but then again, if you do that, you have already proclaimed to the world that you don't care, so nobody else should bother.) #:Erik -- Does anyone remember where I parked Air Force One? -- George W. Bush