From owner-freebsd-doc Tue Jun 4 14:53:17 1996 Return-Path: owner-doc Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id OAA17338 for doc-outgoing; Tue, 4 Jun 1996 14:53:17 -0700 (PDT) Received: from gatekeeper.fsl.noaa.gov (gatekeeper.fsl.noaa.gov [137.75.131.181]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id OAA17333 for ; Tue, 4 Jun 1996 14:53:14 -0700 (PDT) Received: from emu.fsl.noaa.gov (kelly@emu.fsl.noaa.gov [137.75.60.32]) by gatekeeper.fsl.noaa.gov (8.7.5/8.7.3) with ESMTP id VAA10567; Tue, 4 Jun 1996 21:53:04 GMT Message-Id: <199606042153.VAA10567@gatekeeper.fsl.noaa.gov> Received: by emu.fsl.noaa.gov (1.40.112.3/16.2) id AA020905183; Tue, 4 Jun 1996 15:53:03 -0600 Date: Tue, 4 Jun 1996 15:53:03 -0600 From: Sean Kelly To: jfieber@indiana.edu Cc: grog@lemis.de, doc@freebsd.org In-Reply-To: (message from John Fieber on Tue, 4 Jun 1996 16:01:51 -0500 (EST)) Subject: Re: How do I write this SGML stuff? Sender: owner-doc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >>>>> "John" == John Fieber writes: >> Third was the DTD itself. John> Gee, and didn't you just about puke? ;-) I definitely felt ill. But since I'm hardly an SGML expert, it wasn't too bad! I didn't even call in sick the next day ... although I can almost hear my boss saying ``I saw a disgusting DTD yesterday and won't be in today.'' :-) John> In pondering what to do about it, I had visions of a totally John> new system and ditching the linuxdoc dtd. You're not the only one. For one, it has the substring `linux' in it. Okay, I'll admit that that particular string occurs often in quite a lot of FreeBSD. More importantly, it's a much better DTD for writing journal articles than for technical documentation of a computer system---but that's far more because of its lineage in all likelihood. Anyway, yes, I'd love to see a DTD that really reflects the needs of technical writing. Sure, it needs section headings and code blocks, but it also needs warning boxes, and tricks/tips areas. And it'd certainly be nice to have figures of some sort (pbmtoascii, anyone?) but tables even more so. And if not tables, than at least two-column lists. John> Unfortunately with the time I've been able to devote to it I John> don't see that happening very soon Yes, time's the problem. I unfortunately have quite a deficit of it as well :-( John> I'll give a fair warning that

tags will NOT be implied John> by a blank line after the de-crufting. No problem. John> a headache downstream in the processing. And while I've got John> some troff experts on the line, how do I tell troff that I John> want the first character of a line to be a period, not a John> troff command? Given that we already can go to LaTeX which can go to professionally typeset documents, why do we need to also go to troff which can go to professionally typeset documents? (And while I'm dreaming, how about ``sgmlfmt -f ps'' or ``sgmlfmt -f pdf''?) John> Actually, its part of the translation, not the dtd, and it John> could be trivially changed. See! Told ya I wasn't an SGML expert! :-) John> Would you like it changed? Frankly it makes more sense to John> me to NOT assume an extension. Absolutely. John> The HTML translation could probably be mashed into an href="file">file. The file would have to manually be John> added to the makefile to get installed with the html files. Or perhaps sgmlfmt could do gs -dNOPAUSE -dSAFER -sDEVICE=gif -sOutputFile=- -c quit for each referenced EPS file in the SGML source during the HTML conversion. Maybe even add the alt="Image Here" for the benefit of Lynx users or the ASCII version. But then there's the time problem again ... sigh. -- Sean Kelly NOAA Forecast Systems Laboratory kelly@fsl.noaa.gov Boulder Colorado USA http://www-sdd.fsl.noaa.gov/~kelly/