Off-topic: Plain text, XML or LaTex?
Mac
I apologize for this off-topic posting. But I really like to know how
you fellow Emacsien feal about his.
I like plain text, and have for a long time used outline-mode to write
both my own private notes as well as draft documents. I have also for
a long time been thinking about moving to LaTex so that I also can
produce my final documents in Emacs as well.
Now I have an opportunity to move to XML instead. So I would
appreciate your thoughts about this.
How can I get up-to speed with writing XML inside Emacs? Is it worth
it, or should I stick with plain-ol-text? Any pointers to tutorials,
books etc is appreciated.
Perhaps you have a completely different setup that you are willing to
share with me?
/mac
One reason for liking plain text is it's portability (but so are eg
.pdf these days). Another is version control, I'm a heavy user of cvs
(pcl-cvs is *great*).
Phillip Lord
Mac> I apologize for this off-topic posting. But I really like to
Mac> know how you fellow Emacsien feal about his.
Mac> I like plain text, and have for a long time used outline-mode
Mac> to write both my own private notes as well as draft
Mac> documents. I have also for a long time been thinking about
Mac> moving to LaTex so that I also can produce my final documents
Mac> in Emacs as well.
Mac> Now I have an opportunity to move to XML instead. So I would
Mac> appreciate your thoughts about this.
Latex is easy and quicker to write than any of the XML options that I
have tried (docbook/linux doc), and produces nicer documents anyway.
XML is fairly hard to read, and to use, for doing text anyway. XML
support for Emacs is okay (through PSGML mode), but there are many XML
specs it doesn't work with (namespaces, XML schema and so on).
Latex has it's irritations as well of course, but I think its a closer
match for what you want.
Cheers
Phil
Urban Gabor
I would NOT suggest to switch to XML/docbook. We are using
it at the company, and each team-member has a slightly
divverent version of docbook tools, what is making the lifa
a boiling hot hell.
Latex is very easy to start and really high quality...
Gabaux
Linux is like a wigwam: no gates, no windows, and an apache
inside!
Martin Stemplinger
I used LaTeX a lot during my time at the university and I'm using XML
now. Both are IMHO equally hard to learn at first but the XML/Docbook
combo is much easier to adapt to different layout needs (at least I
never managed to write my own LaTeX-Stylesheet).
A selection of tutorials for docbook can be found at
http://docbook.org/wiki/moin.cgi/DocBookTutorials. I found
http://ourworld.compuserve.com/homepages/hoenicka_markus/ntsgml.html
specially useful which describes the setup of emacs for XML processing
(although targeted towards Windows environment).
HTH
Martin
Ted Zlatanov
> I like plain text, and have for a long time used outline-mode to
> write both my own private notes as well as draft documents. I have
> also for a long time been thinking about moving to LaTex so that I
> also can produce my final documents in Emacs as well.
>
> Now I have an opportunity to move to XML instead. So I would
> appreciate your thoughts about this.
>
> How can I get up-to speed with writing XML inside Emacs? Is it worth
> it, or should I stick with plain-ol-text? Any pointers to tutorials,
> books etc is appreciated.
>
> Perhaps you have a completely different setup that you are willing
> to share with me?
>
> .pdf these days). Another is version control, I'm a heavy user of
> cvs (pcl-cvs is *great*).
I like plain text too. Generally I write a converter from my text
format to the desired format, using hand-coded parsing and templates.
Texinfo is actually very nice. You can embed Latex commands in it, I
believe, and for technical documentation it is great. The Emacs
texinfo-mode is very good. Also, Texinfo needs minimal markup, most
of it is plain text you write.
I have used XML with various setups, and the XAE (a package that
glues several others together) is my favorite. I wouldn't use
XML over plain text and Texinfo.
Ted
Jeffery B. Rancier
It really depends on your application. For specifications, reports,
documentation, etc. I like using DocBook in XAE
(http://xae.sunsite.dk) along with PSGML. The XML isn't as readable
as plain-text, but the power of using docbook is the fact that you
have your source written once in XML, then the theory is you can
generate various output formats (with possibe intermediate
translations), from the same source. I.e., HTML, FO, PDF, etc.
Jeff
--
Thanks,
Jeff
,----
| Jeffery B. Rancier
|
| Softechnics
| a METTLER TOLEDO company
`----
Roodw...@core.com
I've never used XML but I have used LaTeX. I'm a big fan of it. I not only
like its superior quality but the fact that the files are plain text.
File formats come and go but from my experience, plain text is forever.
LaTeX isn't for everyone, though. If what you're producing fits into the way
it's set up, it'll be one of the easiest ways to go. If what you're
producing doesn't fit in, you may be in for frustration. You can probably
make it work but not without a lot of effort. Unless you like solving
problems, or, like me, just fall in love with it.
For more opinions you could also post the question to comp.text.tex.
--Rod
--
Author of "Linux for Non-Geeks--Clear-eyed Answers for Practical Consumers"
and "Boring Stories from Uncle Rod." Both are available at
http://www.rodwriterpublishing.com/index.html
To reply by e-mail, take the extra "o" out of my e-mail address. It's to
confuse spambots, of course.
Dmitri...@britishcouncil.ru
I'd say that it all depends on whether you can afford spending some time
tuning the XML tools to your needs. If yes, XML may be usable. Emacs is good
in editing XML (just as good as it is in editing anything else :) , there
are some XSL processors (like xsltproc) and XSL:FO formatters (PassiveTeX or
Apache FOP) of acceptable quality.
Anyway, you'll have either to choose some established standard (TEI or
DocBook) or to write your own XSL transformation from your own DTD to
XSL:FO, which means more time spent on tweaking the xsl files and writing
shell scripts.
The good side of XML is, of course, in XSL. The bad one is that it's still
far from being a ripe standard and some time will pass before, e.g.,
FO-formatters will fully comply with the current standard of XSL:FO.
See "XML, a new start for the Web"
(http://tex.loria.fr/xml-etc/goossens-xml2000.pdf) by M.Goossens and
"PassiveTeX: from XML to PDF"
(http://home.cern.ch/goossens/goossensrahtz.ps.gz) by M.Goossens and
S.Rahtz.
PS: Me, I prefer plain text on paper ;) I don't trust computers, you can
never know with 'em...
--
With best regards,
Dmitri Minaev
> -----Original Message-----
> From: Mac [mailto:oonpl...@sneakemail.com]
> Sent: Monday, June 23, 2003 4:22 PM
> To: help-gn...@gnu.org
> Subject: Off-topic: Plain text, XML or LaTex?
>
>
>
> I apologize for this off-topic posting. But I really like to know how
> you fellow Emacsien feal about his.
>
> I like plain text, and have for a long time used outline-mode to write
> both my own private notes as well as draft documents. I have also for
> a long time been thinking about moving to LaTex so that I also can
> produce my final documents in Emacs as well.
>
> Now I have an opportunity to move to XML instead. So I would
> appreciate your thoughts about this.
>
> How can I get up-to speed with writing XML inside Emacs? Is it worth
> it, or should I stick with plain-ol-text? Any pointers to tutorials,
> books etc is appreciated.
>
> Perhaps you have a completely different setup that you are willing to
> share with me?
>
> /mac
>
> One reason for liking plain text is it's portability (but so are eg
> .pdf these days). Another is version control, I'm a heavy user of cvs
> (pcl-cvs is *great*).
>
> _______________________________________________
> Help-gnu-emacs mailing list
> Help-gn...@gnu.org
> http://mail.gnu.org/mailman/listinfo/help-gnu-emacs
>
Kai Großjohann
> I like plain text, and have for a long time used outline-mode to write
> both my own private notes as well as draft documents. I have also for
> a long time been thinking about moving to LaTex so that I also can
> produce my final documents in Emacs as well.
>
> Now I have an opportunity to move to XML instead. So I would
> appreciate your thoughts about this.
I think the main advantage of using XML is that it is
machine-processable. And the advantage of LaTeX is that it can
produce good-looking output easily.
There are ways to produce, say, HTML from LaTeX, but they are
inherently fragile and difficult to get right. Or RTF, or whatever.
I guess that would be easier with XML.
But on the other hand, LaTeX has really great support for typography,
and I doubt that there is an XML processor that allows you to specify
whether a "." character is part of an abbreviation or the end of a
sentence. (In English, end-of-sentence spacing differs from
inter-word spacing.) In LaTeX this is quite easy.
Does this help?
--
This line is not blank.
Galen Boyer
> the power of using docbook is the fact that you have your
> source written once in XML, then the theory is you can generate
> various output formats (with possibe intermediate
> translations), from the same source. I.e., HTML, FO, PDF, etc.
Plus, the sales factor to your fellow compadres and upper
management includes the letters, XML. My company is on the verge
of moving to docbook as its publication format (based solely on
my publications with it) and the fact that it has XML in its
description has the sexiness needed. I could have never sold
some version of TEX, no matter how much better it might have
been. Either way, one source for the documentation is the key,
and tagged documentation is the way to go.
--
Galen Boyer
Galen Boyer
> Hi,
>
> I would NOT suggest to switch to XML/docbook. We are using
> it at the company, and each team-member has a slightly
> divverent version of docbook tools, what is making the lifa
> a boiling hot hell.
Hm... I will be enforcing this by having all of docbook coming
from CVS. Whether one uses Emacs, notepad or ArborText, they
must use the docbook distribution in CVS. Are you doing this and
still having issues?
--
Galen Boyer
Mac
After reading your posts, I think I'll start by trying out the
"XML/docbook in XAE" solution.
I like the fact that XML is machine-processable, and the theory that
it can be converted to other formats (HTML, PDF, hmm, should it not be
able to generate "reader friendly" plain-text as well?;-), not
forgetting the "sales factor" it has.
I'll just have to sacrify the typhographic defects for now. I'll also
take a note of the issue regarding different versions of XML/docbook.
Thank you all for your comments and links, I really appreciated
everything.
/mac
Summary of your links:
XML Authoring Environment for Emacs
(http://xae.sunsite.dk)
XML, a new start for the Web
(http://tex.loria.fr/xml-etc/goossens-xml2000.pdf)
Robert Mecklenburg
> After reading your posts, I think I'll start by trying out the
> "XML/docbook in XAE" solution.
>
> I like the fact that XML is machine-processable, and the theory that
> it can be converted to other formats (HTML, PDF, hmm, should it not be
> able to generate "reader friendly" plain-text as well?;-), not
> forgetting the "sales factor" it has.
You should take a very serious look at texinfo. From a single texinfo
source file you can generate docbook, xml, html, TeX, info, and plain
text. Obviously, the "sales factor" you'll get will be equivalent to
straight xml, but you'll also have the flexibility of the other
formats. Texinfo is very pleasant to read and write (unlike docbook,
IMHO). Emacs support for texinfo is excellent. The ability to view
your "formatted" document as plain text is quite convenient also as
you edit in emacs. Finally, it is an "industrial strength" solution,
quite a few books have been done in texinfo.
Cheers,
Robert
Steve G
> On Mon, 23 Jun 2003, jeff.r...@softechnics.com wrote:
>
>> the power of using docbook is the fact that you have your
>> source written once in XML, then the theory is you can generate
>> various output formats (with possibe intermediate
>> translations), from the same source. I.e., HTML, FO, PDF, etc.
> Plus, the sales factor to your fellow compadres and upper
> management includes the letters, XML. My company is on the verge
> of moving to docbook as its publication format (based solely on
> my publications with it) and the fact that it has XML in its
> description has the sexiness needed. I could have never sold
> some version of TEX, no matter how much better it might have
> been. Either way, one source for the documentation is the key,
> and tagged documentation is the way to go.
Gene
The `again' in `not again' must certainly qualify.
When people were discussing XML versus LaTex in 2003 I'd liken this to the clay tablet and papyrus days of digital publishing.
Now BOTH of those formats seem to qualify as derivative output formats a half step removed from whatever the reader ends up `consuming'.
With org-mode alone one may transform org-mode markup into BOTH XML and LaTex ... as well as other formats.
And if one starts out with either XML or HTML one may parse either of them into a lisp format to facilitate automating processing ... possibly also using outline mode.
I'd like to see more pendulous `slosh tank' processing going on, wherein bidirectional transformations between mark-up meta-level meta-text generates any number of target formats ... then each of these are transformed back into one-or-more mark-up formats for any number of reasons.
Personally, the biggest problem I have with LaTex and Tex is their slavish, obsequious pandering to Surface Structure ... how things LOOK.
Theoretically, one may use the conditionals of lisp to craft a dynamic semantic web more akin to what peeps like Ted Nelson notion of hypertext and Alan Kay's notion of a Dynabook imagined in the 1960s.
It offends me to the core that we're using digital devices functioning at the performance level superior to the supercomputers of the 1960's to display quasistatic images which Gutenberg would recognize as not all that different from what he could produce with movable type and wood blocks
Why use an AI-capable language such as elisp to generate something as sophisticated as a papyrus scroll or a clay tablet?
It's obscene.
Since 2003 -- when this thread was started -- org-mode mark-up has become available on emacs, and it can be used to generate both XML and LaTex from the same source file.
So why not generate a variety of file formats THEN view them for the superficial factors one wishes to employ to impress the troglodytes who are still immersed in the ink-on-paper paradigm as if one paradigm after another hasn't shifted and surpased it in many ways.
And after viewing several of those formats it would seem that one could go back and tweak one's org-mode markup to start another iteration in the process of pandering to those who embrace static images, stone tools, and bronze age weapons with equal delight?
You all DO know that we've put a man on the moon and that so-called `smart' phones have quad core processors running at over a billion instructions per second ... right?
It doesn't strike you as absurd that with all the potential for dynamism that the human race is STILL playing with coprolith and we pander to so-called `natural' language, don't you?
Why not just use a 3D printer to print fossilized dinosaur shit, then micro-engrave or otherwise chisel out words in some culture's `mother tongue' on the neo-coprolith?
This begs for a theatre-of-the-absurd send up, if you asked me.
Cheers!
Steve G
> On Friday, December 10, 2021 at 5:20:18 PM UTC-5, Steve G wrote:
>> Galen Boyer <galen...@hotpop.com> writes:
>>
>> > On Mon, 23 Jun 2003, jeff.r...@softechnics.com wrote:
>> >
>> >> the power of using docbook is the fact that you have your
>> >> source written once in XML, then the theory is you can generate
>> >> various output formats (with possibe intermediate
>> >> translations), from the same source. I.e., HTML, FO, PDF, etc.
>> I cannot believe I am reading this in a lisp group... Please continue.
>> > Plus, the sales factor to your fellow compadres and upper
>> > management includes the letters, XML. My company is on the verge
>> > of moving to docbook as its publication format (based solely on
>> > my publications with it) and the fact that it has XML in its
>> > description has the sexiness needed. I could have never sold
>> > some version of TEX, no matter how much better it might have
>> > been. Either way, one source for the documentation is the key,
>> > and tagged documentation is the way to go.
>
>> Oh no... not again.
> I'm reminded of that old Latin quote translated into English which
> asserts that one can never step into the same river twice.
> The `again' in `not again' must certainly qualify.
> When people were discussing XML versus LaTex in 2003 I'd liken this to the clay tablet and papyrus days of digital publishing.
> Now BOTH of those formats seem to qualify as derivative output formats a half step removed from whatever the reader ends up `consuming'.
> With org-mode alone one may transform org-mode markup into BOTH XML and LaTex ... as well as other formats.
>
> And if one starts out with either XML or HTML one may parse either of them into a lisp format to facilitate automating processing ... possibly also using outline mode.
>
> I'd like to see more pendulous `slosh tank' processing going on, wherein
> bidirectional transformations between mark-up meta-level meta-text generates
> any number of target formats ... then each of these are transformed back into
> one-or-more mark-up formats for any number of reasons.
sql, comma seperate values (CSV), for years. I think the new thing is to
use the JavaScript ? Notation (JSON). I just had a look at JSON; it
seems to be very popular with GO (programming language) and I think SQL
now. It seems like it would be relatively easy to parse into lisp SEXP.
> Personally, the biggest problem I have with LaTex and Tex is their slavish, obsequious pandering to Surface Structure ... how things LOOK.
not discuss the subject.
> Theoretically, one may use the conditionals of lisp to craft a dynamic semantic web more akin to what peeps like Ted Nelson notion of hypertext and Alan Kay's notion of a Dynabook imagined in the 1960s.
PDF. I think it was made at AT@T
> It offends me to the core that we're using digital devices functioning at the
> performance level superior to the supercomputers of the 1960's to display
> quasistatic images which Gutenberg would recognize as not all that different
> from what he could produce with movable type and wood blocks
> Why use an AI-capable language such as elisp to generate something as sophisticated as a papyrus scroll or a clay tablet?
> It's obscene.
> Since 2003 -- when this thread was started -- org-mode mark-up has become available on emacs, and it can be used to generate both XML and LaTex from the same source file.
> So why not generate a variety of file formats THEN view them for the
> superficial factors one wishes to employ to impress the troglodytes who are
> still immersed in the ink-on-paper paradigm as if one paradigm after another
> hasn't shifted and surpased it in many ways.
adventure with my hatred of using San Serif typfaces on the screen
@96dpi when people where throughing out 1200dpi laser printers.
I am now in recovery. I have accepted the sans serif type faces. I now
emplore them regulary with great delight. It is no more difficult than
accepting a medium weight or the use of a semibold. The new san serif
fonts are usually OTF with fine hinting these days. You can use the same
font format OTF (open type font format) for type1 typefaces, quadratic
splines, bitmaps, and the combination. The format seems to have settled
some of the microsoft vs apple problem.
GOD Bless the people of the freetype project! Remeber what X used to
look like with those crappy bitmap fonts that it came with...
> And after viewing several of those formats it would seem that one could go
> back and tweak one's org-mode markup to start another iteration in the process
> of pandering to those who embrace static images, stone tools, and bronze age
> weapons with equal delight?
> You all DO know that we've put a man on the moon and that so-called `smart' phones have quad core processors running at over a billion instructions per second ... right?
> It doesn't strike you as absurd that with all the potential for dynamism that the human race is STILL playing with coprolith and we pander to so-called `natural' language, don't you?
> Why not just use a 3D printer to print fossilized dinosaur shit, then micro-engrave or otherwise chisel out words in some culture's `mother tongue' on the neo-coprolith?
> This begs for a theatre-of-the-absurd send up, if you asked me.
> Cheers!
elho
Richard Smith
I'm a user as it serves Trade welding, engineering and science work.
I use flat text wherever possible. Emacs. With everything good
including its bookmarks, its abbreviation expansions (I didn't just
type
"Thermo-Mechanically Controlled-Processed High-Strength Low-Alloy"
in describing the steel!)
I'll just spin a calculation for a 1016x305x584UB ("Universal Beam")...
(ma2nd-z-plt-ibeam-prettyprint 314.0e-3 1056.0e-3 36.0e-3 64.0e-3)
"I-H : b=0.314 h=1.056 t_w=0.036 t_f=0.064 : I=1.229913e-02 Z=2.329381e-02"
"I is the 2nd moment of area; "Z" is the section modulus. Beam
intrinsic geometric properties.
My beam spans 20 metres and is centrally loaded...
(/
(beam-fmax-ibeam-simple-cload 314.0e-3 1056.0e-3 36.0e-3 64.0e-3 20 275e6) ;; 1281159.3955555558 ;; N
9.81 1e3) ;; 130.59728802808928
It will theoretical according to Euler-Bernoulli beam calculation take
130 Tonnes force of concentrated central load at its midspan, 10m from
each end of its 20m span.
I do my technical webpages in HTML as if flat text. I have some
functions to set-up a document with headers and footers, insert
images, links, etc. to expedite typing at the keyboard.
"LaTeX" is have used for 25 years.
It's been a success.
No two ways about that.
No matter the critical analysis of markup, typesetting and all that
good stuff.
I have stayed sane (?!) producing big documents, where the same
observation would not be made for others using "user-friendly software
applications".
The constancy of the fundamental methods means years can go by then a
new need can have you pick up where you left off doing your next task.
I appeal to you to think of the pragmatic as I present it as this
"metal-basher".
My Doctoral thesis was in typeset PDF version 22 years ago - all 240
pages.
A few years later I blanked the "space box" code for where I glued
photos and un-commented-out the image typeset instructions whose code
was already known at time of writing and re-ran the LaTeX typesetter.
The result was accepted as authentic and went to the archive as a PDF
with all images incorporated
http://bura.brunel.ac.uk/handle/2438/4617
Ten years later I did some more research, this time in a more
engineering discipline
http://www.weldsmith.co.uk/big_files/thesis_fatgproj_rev111029/fatigue_perform_Tjoint.pdf
Again - systematic working; hard work but under smooth control.
By the way - back to flat text - in that work I had to set the fatigue
testing machine. The console offered only the cycle-rate (eg. 10
cycles per second the machine maximum), the minimum force and the
maximum force. The testing machine couldn't "see" the sample. All it
new was when the sample broke (all force disappeared) so it stops and
holds the number of cycles at which that event happened (typically
hundreds of thousands to millions).
So I had to know the sample size so convert force to stress; know what
stress that should be for a short harsh or longer more typical service
stress test, etc. Know that my test should fit into the time before
the next machine booking - so tweak the stress to get information
needed which was achievable in the time.
Cascades of calculations.
Likely when fraught, tired, and possibly near building locking-up
time, etc.
So - suite of functions in elisp. As ever - grew from the bottom up,
until gathered at the "consumer item" level with a grand
invoke-all-in-a-cascade function. Which - as ever - pretty-prints its
inputs with labels so a person cross-eyed with tiredness gets to
survey all, made easy for the human.
(there's a pretty-print of the inputs to the beam calculation earlier,
as a simple but representative example. Simplicity is virtue here)
Plus those two crucial numbers, the minimum force and maximum force
you need to key-in at the testing machine's console.
Again, I'm saying - yes pursue the finer points of what the computing
and information technology way ahead might be - but - do remember how
well-serving are these "sub-optimal" methods which are never-the-less
extremely rational and effective ways of working.
They will get the best out of the human, doing whatever it is that
their endeavour is.
So I want to encourage you.
Both to go on, and to appreciate how much these existing computational
tools you work with are valuable to those of us in our abstruse
specialisations.
With best wishes,
Rich Smith
Steve G
> Technically neglible knowledge and understand.
> I'm a user as it serves Trade welding, engineering and science work.
Richard Smith
(Leslie Lamport is creator of "LaTeX" (which rides on Knuth's TeX
typesetter(?)))