Return-Path: owner-linux-activists@joker.cs.hut.fi Received: from joker.cs.hut.fi (root@joker.cs.hut.fi [130.233.40.32]) by keos.Helsinki.FI (8.6.4/H45) with SMTP id LAA03080 for ; Wed, 9 Feb 1994 11:45:20 +0200 Received: from joker.cs.hut.fi by niksula.hut.fi id <50801-3>; Wed, 9 Feb 1994 11:43:44 +0200 From: "Linux Activists" To: "Linux-Activists" Reply-To: "Linux-Activists" X-Note1: Remember to put 'X-Mn-Key: DOC' to your mail body or header Subject: Linux-Activists - DOC Channel digest. 94-1-9-6:12 X-Mn-Key: DOC Sender: owner-linux-activists@joker.cs.hut.fi Message-Id: <94Feb9.114344eet.50801-3@niksula.hut.fi> Date: Wed, 9 Feb 1994 11:43:34 +0200 Status: RO X-Status: Topics: Re: Linux-Activists - DOC Channel digest. 94-1-8-5:5 Linux-Activists - DOC Channel digest. 94-1-8-22:23 Linux FSSTND 1.0 almost completed ---------------------------------------------------------------------- From: Thomas Dunbar Subject: Re: Linux-Activists - DOC Channel digest. 94-1-8-5:5 Date: Wed, 9 Feb 1994 03:27:03 +0200 > > 3. Using SGML, HTML, or some other notation is not an option for me > > either, at this time: I have no tools, and no time to install or > > learn such. However, experimenting with other formats is a good thing, > > and after others have done it, I'll be happy to follow. > > I'm going to start looking at SGML. Once I have it all worked out, I > can write up a short tutorial (it looks quite easy), even get a Linux > binary distribution of the software you'll need. > > It looks like the HOWTOs are going to move to SGML, so that will be > my guinea pig. :) > how about a little more precision here: i suspect that when people say SGML, they actually mean the subset HTML+ and in particular the subset of HTML+ understood by Mosaic. dont mean to be pedantic, just want to understand what others mean. imo, current implementations of HTML+ can't produce good looking books. What is needed is something more sophisticated than Mosaic to run the html thru, say TeX. but then why not just keep the doc in LaTeX, taking care to limit oneself to what's convertable by ones latex2html. "is that what you're doing with your book?" no :( for one, i dont use latex. but primarily, i dont think it possible to get decent hardcopy and screencopy for the same document. or at any rate, if possible, no easier than just doing two different, but related, docs. "so what are you doing with your book?" nothing right now..but i intend to convert it (now, before i get into writing the main part) for good screencopy bias, and not worry about hardcopy. that moves away from TeX, regretably, but works out better wrt my emphasis on interacting with METAFONT. That change fits nicely with Linux also, i think. after all, the global network is an essential feature of linux development..shouldn't documentation for it be woven into the net, too? course that means i wont see my book in print..no big deal. besides, probably wouldn't anyway! "so what?" guess my main point is to argue that one must choose between good hardcopy and good screencopy. what html+ offers is, for the first time, good screen copy (formatting, links, annotations, graphics) is possible in a portable format. -- Thomas Dunbar phone: 703 231-3938 Virginia Tech fax: 703 231-3714 100C Sandy Hall WWW: http://milieu.grads.vt.edu/ ------------------------------ From: nick@nsis.cl.nec.co.jp (Gavin Thomas Nicol) Subject: Linux-Activists - DOC Channel digest. 94-1-8-22:23 Date: Wed, 9 Feb 1994 06:00:35 +0200 >I believe that the best choice for hypertext on-line documentation >would be HTML because there already exists a variety of tools that >can use it and there is also an editor for it in tkWWW I think. I would recommend HTML+, because: 1) It supports tables 2) It will be the future HTML standard However, it does not support figures (except inline stuff). There are other alternative SGML DTD's. One can choose DocBook, QWERTZ (which Matt favours), TEI's one, and a whole host of others. The advantages of using HTML+ however are: 1) You get online browsing and fine-grained hypertext for free. 2) Converters already exist for it, and more will come. The biggest disadvantage is that it is not all that pretty. I mean using things like

instead of . Anyway, if anyone is really planning to do something with SGML, two tools I would recommend looking at: 1) sgmls + qwertz This gives you a good idea of what an SGML publishing system should look like. 2) The Amsterdam SGML Parser This is a *very* flexible system for working with SGML. It converts a DTD into an LL(1) grammar, and then generates an executable parser for it. Like sgmls, it uses replacement files for conversion to other formats. With these two tools alone, or possibly with addition scripting language support, any conversion is possible. One way or another, I think SGML is the way to go, and I must admit, I'm biased toward HTML+, because I think it'll save a lot of work in the long run. I don't see anything wrong with extending it somewhat to support extra functionality, because the draft standard for HTML+ browsers states that they should ignore entities and elements they don't recognise. This will allow the LDP to add whatever hacks they desire, and *still* get SGML and hypertext in one file, which can be easily converted to LaTeX for printing. We had a similar discussion 2+ years ago, we wouldn't be having this discussion now if people thought ahead a little then. ------------------------------ From: quinlan@spectrum.cs.bucknell.edu (Daniel Quinlan) Subject: Linux FSSTND 1.0 almost completed Date: Wed, 9 Feb 1994 06:52:02 +0200 While it isn't exactly documentation, the first public version of the Linux filesystem standard is to be released within about a week. Since it is already being used as a reference by many developers and some Linux documenters have followed our progress, this notice is being posted here in consideration of the DOC channel's readers. In short, anyone who is interested can take an early look at this standard from the FSSTND group (formerly of the FSSTND channel). For those who are not yet aware of this long-term and comprehensive effort, this abstract taken from the current draft should explain the essence of it: The open and distributed process in which the Linux operating system has developed fosters rapid growth of the operating system, applications, and integrated distributions. This decentralized process, however, has created a need for standardizing the structure of the filesystem. This standard aims to define locations and specifications for files and directories in Linux systems. This will allow users, developers, and distributors to assemble parts of the system from various sources that will work together as smoothly as if they had been developed under a monolithic development process. It will also make general documentation less difficult, system administration more consistent, and development of second and third party packages easier. The draft is available through anonymous ftp at tsx-11.mit.edu:/pub/linux/docs/linux-standards/private/fsstnd In the "fsstnd" directory there is an ASCII copy (named according to the date of the revision -- "draft-yy.mm.dd") and PostScript/DVI copies of the most current draft ("draft.ps.gz" and "draft.dvi.gz"). Also, please note the "private" in that pathname (which won't remain for long). If you have any questions or comments, please feel free to direct them towards me or any contributor listed in the standard. Thank you. Daniel Quinlan FSSTND Coordinator -- Daniel Quinlan ------------------------------ End of DOC Digest ***************** ------- .