Date: Wed, 07 Mar 2001 20:18:21 -0800 From: Jack Rusher <jar@integratus.com> To: arch@freebsd.org Subject: Re: Configuration management Message-ID: <3AA7080D.3A450101@integratus.com>
next in thread | raw e-mail | index | archive | help
Wes Peters wrote: > > There is an interesting opportunity here for somebody adventurous and > not afraid of hard work to set about converting FreeBSD (or any other > source-available operating system) to a test case for a new management > framework. Gathering up all the various configuration files or data > stores on a system and replacing them with, say, an XML representation > of the same data would be an interesting project. Providing legacy APIs, > like getpwent, etc., that used the XML format, as well as providing the > new API you would like to see would be a straightforward, if somewhat I've had this bug in my ear for awhile now. Unfortunately, there seem to be some pretty strong human engineering difficulties with this sort of work. Some time ago, I took a couple of days of my personal time and converted one of our products to use an XML configuration file format. The other engineers voted this down because they felt that the presence of tags in the files made them less readable than the positional configuration files (!) we had before. I can only imagine the grief we might receive if /etc/passwd were to wake up one morning in a format that differs from what people have come to expect. On the other hand, I would be willing to donate some time and effort, write some DTDs, write some format converters, & produce a stripped down bsdxml library to help experiment with this... if enough people actually expressed interest in the idea. I would hate to experience my previous disappointment on a larger scale after doing a lot more work. I would like to see this effort combined with the work being done on both SonOfSysinstall and the various network directory and configuration database projects that are happening in the world. The parsers for each of these things could come together into a single BSD license XML parser. I'm pretty sure BSDi & Apple would be interested in this work (perhaps to the point of helping out with some funding). Lastly, in an attempt to preempt the "why XML" bikeshed, XML gives you a language with which you may define other little languages within constraints that work especially well for this sort of problem domain. We could come up with our own meta language, but why? XML isn't the great savior the pundits would suggest, but it is a standardized technology that people understand, and it would work for this purpose. Comments? -- Jack Rusher, Senior Engineer | mailto:jar@integratus.com Integratus, Inc. | http://www.integratus.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3AA7080D.3A450101>