From owner-freebsd-arch Mon Mar 26 23:42:29 2001 Delivered-To: freebsd-arch@freebsd.org Received: from winston.osd.bsdi.com (winston.osd.bsdi.com [204.216.27.229]) by hub.freebsd.org (Postfix) with ESMTP id 50A2E37B719 for ; Mon, 26 Mar 2001 23:42:25 -0800 (PST) (envelope-from jkh@osd.bsdi.com) Received: from localhost (jkh@localhost [127.0.0.1]) by winston.osd.bsdi.com (8.11.3/8.11.2) with ESMTP id f2R7fT034079; Mon, 26 Mar 2001 23:41:29 -0800 (PST) (envelope-from jkh@osd.bsdi.com) To: Cy.Schubert@uumail.gov.bc.ca Cc: jar@integratus.com, areilly@bigpond.net.au, jonathan@graehl.org, freebsd-arch@FreeBSD.ORG Subject: Re: configuration files, XML? In-Reply-To: <200103270707.f2R77SR08693@cwsys.cwsent.com> References: <3AC007FF.1D9F60A3@integratus.com> <200103270707.f2R77SR08693@cwsys.cwsent.com> X-Mailer: Mew version 1.94.1 on Emacs 20.7 / Mule 4.0 (HANANOEN) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-Id: <20010326234129J.jkh@osd.bsdi.com> Date: Mon, 26 Mar 2001 23:41:29 -0800 From: Jordan Hubbard X-Dispatcher: imput version 20000228(IM140) Lines: 44 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG I think the point which has been emerging from this discussion is that one can't really win in this game. In scenario A, you have an /etc/foo.conf and all the Unix die-hards ("That's *always* been in /etc/foo.conf!") and heterogeneous environment folks ("I have 437 different Unix variants here and 391 of them use /etc/foo.conf") are happy. Everyone else, who wonder just why the heck it's called /etc/foo.conf and why the heck its syntax looks like some bad late-70's ASCII terminal control character set (answer: "It is!"), is unhappy. In scenario B, you have /var/db/localhost/config.pgsql with an complete documentation set for interface programmers and set of access utilities for everyone else which totally front-ends the configuration process so that Joe User doesn't even have to know where the data is kept or what format it's in. All the non-Unix and not-so-techy people are happy. The Unix die-hards and heterogeneous folks are hunched over a toilet bowl being explosively sick and screaming for their Solaris or SCO installation media. In scenario C, you have the abstract configuration database AND you have /etc/foo.conf with some sort of fancy two-way auto-merging thingamajiggy which propagates any change made from one data representation to the other. The joe users are still happy because they can use the configuration database thingy, though they still sometimes stumble over /etc/foo.conf and wonder just why the heck it's there. The Unix die-hards, on the other hand, are highly skeptical about viability of this whole thing and proceed to prove their point by getting themselves wrapped around the axle numerous times while attempting to use both interfaces at once or getting into odd race conditions between the two interfaces. Between that and occasionally managing to turn their systems to spam by having the wrong edit back-propagate to the correct copy, well, pretty soon they're calling your scenario C solution a grossly over-engineered and highly dysfunctional hack. "Bring back scenario A!", they will cry. To put it somewhat more succinctly, I think we're all pretty wedded to our ways and only accept change when it's a significant order of magnitude improvement beyond what we currently have or it doesn't conflict with any existing scenario. Thus Unix slowly accretes functionality to itself without changing shape too much. :) - Jordan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message