Date: Tue, 10 Jul 2001 06:54:50 -0700 From: Dima Dorfman <dima@unixfreak.org> To: Nik Clayton <nik@freebsd.org> Cc: doc@freebsd.org Subject: Re: State of the Handbook Message-ID: <20010710135450.EA8803E2F@bazooka.unixfreak.org> In-Reply-To: <20010710135753.G16152@clan.nothing-going-on.org>; from nik@freebsd.org on "Tue, 10 Jul 2001 13:57:53 %2B0100"
next in thread | previous in thread | raw e-mail | index | archive | help
Nik Clayton <nik@freebsd.org> writes: > On Tue, Jul 10, 2001 at 04:32:47AM -0700, Dima Dorfman wrote: > > Do we have a mechanism so that we can keep the ASCII art and display > > it where necessary? I.e., I'd rather not see the ASCII art thrown > > away; not everybody wants to start a graphics-capable browser to read > > the docs (I know I don't). > > Yes we do. See section 4.2.6 of the primer. > > We've previously talked about adding another format to the set that's > generated; html-no-images (or similar). That's trivial to generate > (since we already generate it in order to produce the plain text version > of the docs). > > However, it brings up an additional issue -- currently, the output formats > are all produced in the 'current' directory. This works OK for the > formats that we have, since none of them have filenames that 'collide' > with one another. > > It won't work with a hypothetical 'html-no-images' format. The likes of > 'book.html' is perfectly legitimate in this format, which would collide > with the files generated by the 'html' format. It's also feasible that > people would want an 'html-split-no-images' format as well, the output > files for which would collide with the output from the 'html-split' > format. Okay, so it's possible but not exactly trivial. Well, that's good enough. At this point, it'd just be nice if whoever replaces the ASCII art with something else leave the ASCII art intact, perhaps in a comment. > > I don't see a problem with individual copyright holders as long as > > they're using a reasonable license. =20 > > We still have to define 'reasonable' in this context. There are a half > dozen or so 'documentation' licenses out there; the GNU Free > Documentation License, the Open Content License, and whatever licenses > people like Apple are using for Darwin. Then there's the BSD-style > license I (fairly abitrarily) slapped on the Handbook. Berkeley stuck their BSDL on the man pages that shipped with 4.4BSD. I mean they stuck it on them verbatim; it still mentions distribution in source and binary form :-). At least you modified the one you stuck on the Handbook to actually refer to documentation rather than source code. Well, I guess we need to come up with one de-facto license and ask that new documentation use it, and try to convince the other copyright holders to adopt it. > > I'm also uncertain how things like the copyright notice on chapter 13, > or appendix F, 4,3, Which chapter 13 ;-)? On my local build, 13 is l10n; on www.freebsd.org (which is messed up right now), it's x11. Neither has anything that resembles a license at the top. > I guess it's not so much copyright as license that I'm concerned about. > > > Their name in the copyright > > notice is also a form of credit; not everybody thinks it's appropriate > > to clutter the main text with "I rewrote this" type of attributions. > > You snipped the bit where we discussed attribution. Any thoughts? I like the idea of sticking attributions somewhere away from the mainline text. Since this is what you proposed, I felt no need to comment on it. > > Erm, the bibliography should really stay with the document (book) that > > it goes along with. However, the Handbook's Bibliography section > > should be spelled Other Links. > > I was envisioning a 'meta' bibliography. Books/articles that need a > bibliography include this one, and the stylesheets can be configured to > only include information about books that are referenced in the main > text. > > At the same time, we could then have a separate 'bibliography' book that > just contained details about every other document listed in the > bibliographies of all the other books. Okay, that makes sense. "Bibliography of all FreeBSD Documents". I'll see about doing that. > > And while WRS is sponsering this stuff, perhaps they can get another > > computer to act as www.freebsd.org; having freefall do that isn't good > > for security. Oh, and maybe they can help fix the "broken Handbook" > > problem that comes up every so often on the lists; it's quite real, > > and is happening more and more often now :-/. > > Working on it (I've just pinged admins@freebsd.org about this). I hope you mean about the first part, not the part about the corrupted Handbook. I think the latter is caused either by a malfunction in Jade, or, more likely, a malfunction in the way Jade is run. I recall logging into freefall and seeing two instances of jade running; this either means that the build took more than 12 hours, that the build scripts used the -j option to make(1), or some other weirdness. Since this happens quite often on freefall but never for anybody else (that I know of), it's likely that it has something to do with the build environment. Dima Dorfman dima@unixfreak.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-doc" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010710135450.EA8803E2F>