discuss: LinuxDoc vs. DocBook (was Re: Staff Election)
Subject:
Re: LinuxDoc vs. DocBook (was Re: Staff Election)
From:
Binh Nguyen ####@####.####
Date:
3 Apr 2005 09:34:17 -0000
Message-Id: <1081abc9050403013352c44ff9@mail.gmail.com>
> > Why is it that this conversation always brings the worst out in people?
>
> I have seen worse. You have not been flamed.
>
> > I never insulted anyone....
>
> Sure? You just said "this conversation always brings the worst
> out in people".
Fairly sure.... Just to make sure though, if I've caused any offense
to anyone, I would express to you my sincerest apologies. Sorry.
I agree to disagree on our definitions for 'theory' and 'hypothesis'
and that we have had a misunderstanding. However, you're reducing the
scope and any possible input to this discussion by utilising purely
scientific methods. How about using Kuhn's defintion for a theory?
That, is a theory must exhibit five characteristics - accuracy,
consistency, scope, simplcity and fruitfullness.
> > ****
> > How do you avoid this conversation when David brings up LinuxDoc in such a
> > significant number of his emails.
> >
> > A search via Google for
> >
> > "david lawyer site:lists.tldp.org" returned 640 hits
> >
> > "david lawyer linuxdoc site:lists.tldp.org" returned 857 hits
> >
> > "david lawyer docbook site:lists.tldp.org" returned 575 hits
> >
> > Clearly, something strange is occuring here!
>
> I see no strangeness. He simply uses LinuxDoc and joins in
> on these discussions. When he makes his points I do not make
> "mee too"-posts, I try to limit needless traffic.
Once again, I agree to disagree. Joins in these discussions? He
starts them doesn't he? Also, look again at the numbers. He records more
LinuxDoc discussions than he does general discussions!
> > raised such as metadata, tool availability, the popularity of Docbook and
>
> I have brought up the issue on metedata previously and have even
> made working prototypes that demonstrate we do not need them yet.
> There are occational threads on metadata and the semantic web but
> so far I have not yet seen any demonstrations from the proponents.
> It is up to them to prove their point, preferrably with a real
> prototype.
However, if we don't choose Docbook then we well be locking out any
people in the future who may wish to use metadata....
> If you choose DocBook you also have the 5 choises below still
> as well as choices in tool suite (jade v. openjade etc).
I got the tool suite working with one line on my Debian machine :) I
just can't remember what it was though.... :(
> You might remember that I proposed TLDP should host RPM and DEB
> files for a tool chain set in order to help people get started.
> My focus is on improving things, not limiting choices.
So is mine, but the manner in which we go about it are clearly different.
> This issue is of interest to those that would like to make a
> career of documentation. At work I have helped people in
> getting information on DocBook for use in technical documentation
> and in the end they decided on using it.
>
> I do not wish to make a professional career of documentation
> so to me LinuxDoc is sufficient: it does what I want and it
> does fullfill my requirements.
Nor do I have any intention of making a career out of documentation (my
most recent jobs have been in business administration/management). However,
on how many other major documentation projects do you see LinuxDoc
being used as the primary language?
A search for Google for the terms 'mailman docbook' and 'mailman
linuxdoc' 202,000 and 141,000 respectively.
> Incidentally I see LaTeX returns about 9,580,000 hits.
Its also been around for longer....
http://www.xent.com/FoRK-archive/feb98/0307.html
http://www.oasis-open.org/committees/docbook/intro.shtml
Mid-80s versus early-90s.
> A full statistics would be interesting to see.
I'd be interested as well. Not sure how to get at them without too
much pain though....
> > In the case of the smaller HOWTOs I am willing to convert the documents
> > over to DocBook by myself (if need be).
>
> Is there a need currently? My impression is that there are
> other areas in far more urgent need of attention.
'if need be'
> > Earlier in this conversation it was stated :-
> >
> > 'For paragraph separation it uses infinitely more tags.'
> >
> > This is clearly untrue. Infinite implies limitless/boundless. This argument
>
> This is the result on 1/0.
>
> > is biased and should be stricken from the record.
>
> I disagree.
The definition of 1/0 is undefined
(http://mathforum.org/library/drmath/view/55764.html (Drexel
University),
http://alumnireview.queensu.ca/pastissues/holiday2002/letters.htm
(Queen's University)).
However, once again you're reducing the scope of this discussion to the field of
mathematics. Remember how Popper said that adding further conditions
to ensure that a theory can not be falsified is absurd?
> If such numbers alone were sufficient argument we should
> immediately switch to LaTeX.
The only time I've seen Latex used of late is in the academic environment
and even then people use it with derision. It suffers from the same issues
that LinuxDoc has in that it does not produce a well-formed document,
doesn't have support for metadata and also handles whitespace in a
somtimes very peculiar manner (not sure about LinuxDoc on this issue)
Most of the time I've seen people just copying and pasting to
templates.
> That guide is a work in progress and is likely to remain so for
> quite some time. Irrespective of preferred format we should
> improve it whenever we see we can make it better.
Agreed! Woohoo! Yippee! :)
> > If you'd like to argue in terms of reasoning and logic then an argument
> > is only valid if its premises are true (http://www.answers.com/topic/valid).
> > Hence, in order to lay the foundations for an LDP with a better future
> > and one that
> > has relevance in the prevailing Open Source world I believe that it
> > will be necessary
> > for us to utilise technologies that have been established as the
> > 'standard' as they
> > will be the technologies for which new tools and innovations will be
> > created for.
> > ****
>
> TLDP should make decision based on the situation for TLDP and
> I believe it benefits TLDP to offer LinuxDoc as a choice.
>
> I am all for innovations and tools and if you were to think
> back on past announcements you would realise I have made
> - templates
> - HOWTO Generator
> - NetHelp based HOWTO reader demo
> and more, all based on LinuxDoc.
If you're so determined to keep LinuxDoc than so be it. However, I
would also like to endorse
http://lists.tldp.org/index.cgi?1:mss:8956:200503:mjolkcjnjbginkhagjki
Scot W. Stevenson's idea of incorporating AsciiDoc
(http://www.methods.co.nz/asciidoc/)
into our process. Further, I must endorse Martin Wheeler's suggestion
and ability to be
able to run (did you volunteer Martin?, I think you did?) a
Wiki/database that is within the requirements of the LDP and which
should hopefully allow us to expand our collection of documents at a
far greater rate.
Regards,
Binh.
