X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil] ["1849" "Sun" "6" "October" "1996" "21:43:03" "+0200" "Hans Aberg" "haberg@matematik.su.se" nil "41" "Re: General configuration of LaTeX" "^Date:" nil nil "10" nil nil nil nil] nil) Received: from listserv.gmd.de (listserv.gmd.de [192.88.97.1]) by mail.Uni-Mainz.DE (8.7.5/8.7.3) with ESMTP id AAA03811; Mon, 7 Oct 1996 00:28:52 +0200 (MET DST) Received: from listserv.gmd.de by listserv.gmd.de (LSMTP for OpenVMS v1.1a) with SMTP id <14.FA1E9D9E@listserv.gmd.de>; Mon, 7 Oct 1996 0:28:51 +0200 Received: from RELAY.URZ.UNI-HEIDELBERG.DE by RELAY.URZ.UNI-HEIDELBERG.DE (LISTSERV-TCP/IP release 1.8b) with spool id 204449 for LATEX-L@RELAY.URZ.UNI-HEIDELBERG.DE; Sun, 6 Oct 1996 20:44:23 +0100 Received: from mail.nada.kth.se (root@mail.nada.kth.se [130.237.222.92]) by relay.urz.uni-heidelberg.de (8.7.6/8.7.4) with SMTP id UAA06442 for ; Sun, 6 Oct 1996 20:44:22 +0100 (MET) Received: from [130.237.37.58] (sl38.modempool.kth.se [130.237.37.58]) by mail.nada.kth.se (8.6.10/8.6.10) with SMTP id UAA19569 for ; Sun, 6 Oct 1996 20:44:07 +0100 X-Sender: su95-hab@mail.nada.kth.se Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Message-ID: Reply-To: Mailing list for the LaTeX3 project Date: Sun, 6 Oct 1996 21:43:03 +0200 From: Hans Aberg Sender: Mailing list for the LaTeX3 project To: Multiple recipients of list LATEX-L Subject: Re: General configuration of LaTeX Status: R X-Status: X-Keywords: X-UID: 1767 >Configuration.... > >There are arguments both for and against configuration. > >The main argument against is that it makes document portatbility >much harder. For instance if there was a latex.cfg then people would >add things like a4 paper size, German.sty style easy "o umlauts, >and perhaps load graphics package or whatever. > >This would mean that a document produced at that site would not run at >any other site. (Yes you could document what stuff was loaded locally, >but who reads documentation:-) > >The graphics package configuration files are rather different, they >don't change (much) the top level user interface, they just change the >back end so that say a file produces on unix/dvips can run unchanged >on Mac/textures. So they *increase* rather than decrease document >portability. > >Since this is latex-l list (for long term suggestions) it would be >possible to come up with a configuration scheme if a mechanism could >be devised such that a `portable' document could always be reliably >produced (by the system, not by the user). Any ideas??? Actually, an idea that comes to my mind, is that a \NeedsPackage{} function should contain information about a URL where the needed package can be fetched. Thus, if you do not have the package available when compiling a document, you would get an error message "This document needs the file foo, look at URL bar", or something. This URL may then be document with information of where the file can be found, as URL's may change more frequntly than the LaTeX distributed files. When compiling a file "foobar.tex", it could generate a file "foobar.url", with the URL info, which then can be included with the document distribution. Packages could also have distinctive names, that could be used in a net-search, in case the URL has moved. -- Or something. Hans Aberg