Your opinion: latex or context for me?
casioc...@gmail.com
I'm learning on windows right now but in the long term I plan to
transition to linux. I won't be doing any math or graphics, but mostly
text, ranging from short letters to book projects.
I had planned to just learn latex since it was the most widespread but
have since heard good things about context. So I decided I'll learn
latex and either know of context or it probably won't harm if I ignored
it altogether.
But then I downloaded context and it seemed pretty well-done. Now I'm
wondering if I should just use context and 'know of' latex in case I
come across a situation where context won't be available to me (though
in likelihood nor will latex!), though seeing how the download I got of
context loaded and ran without requiring an installation I'm now
thinking that perhaps if need be I could carry a CD of it with me, or
just edit the plaintext till I later on compile it - I guess
portability should not be a concern. After all, there's that tex live
cd though I have not looked into it.
I'm sure some would say use what I'm comfortable with, but what's your
opinion in terms of what's a good discipline for me to adhere to?
Should I learn and use context and just know of latex, or should I just
use latex and just know of context?
Thanks
Will Robertson
Well, obviously it all depends on who you ask. The advantage to LaTeX
is that there's a large (relatively) community behind it and a large
number of extensions by way of classes and packages. This has the
downside that it may be possible to do something, but (a) you might not
know that the extension you want exists, and (b) it might clash with
something else. (Generally, though, these aren't problems often
encountered.)
On the other hand, ConTeXt will do probably everything you need out of
the box, so you've less places to look when you get stuck (this being a
good thing).
Not being a ConTeXt user myself, I can't give you a definitive answer,
but I have heard very good things about it. If you're willing to start
fresh and you don't need to use LaTeX for things like journal
submissions (and it sounds like you don't), I'd certainly give ConTeXt
the first look.
You can always switch over to LaTeX at a later date if you feel you're
missing out on something!
Will Robertson
John Culleton
You will get opinions that reflect the preferences of individuals. I have
used LaTeX in the past, and even consulted on some LaTeX jobs with e.g.,
The Federal Reserve Bank of Cincinnati. But now I work exclusively in
either pdftex or Context. I find Latex to verbose, too confining, and too
dependent on a multiplicity of packages, each with its own documentation.
Context OTOH is kept under pretty tight control by its original author, Hans
Hagen, who is still very active in its development and support. It is also
a newer product and hence IMO more in tune with today's needs.
I have no doubt others will have very different opinions. This newsgroup
consists of about 95% LaTeX users. But FWIW I come down on the side of
Context and pdftex. And I charge money for my services.
--
John Culleton
Able Indexers and Typesetters
ivo...@gmail.com
I am struggling with the same question. Worse, I already know latex as
an end user (not as a programmer) quite well, and rely on many hacks
that I have accumulated over the years. So, here is my perspective.
>From what I know, latex has many more books (especially the LaTeX
companion, which I find indispensable) and many more users (esp. on
comp.text.tex), which can be helpful if you run into trouble. latex
may also have many more specialty packages, though I have no explored
this yet. context lacks the equivalent of the Addison-Wesley series of
books on latex.
For all practical purposes, latex seems to have stopped development at
an ordinary speed. The latex3 project appears dead. (Progress appears
mostly in pdflatex.) A latex document is also practically
unparseable, and latex itself has no built-in feature that allows one
to interface with other languages. (For many practical purposes,
perltex is too slow, and other solutions tend to be weird or complex
hacks [no insult intended] that are not relatively universally
installed with latex distributions. I have begged repeatedly for a
fast \system{} call, like the \write18 call.)
context still seems to be alive, but I am afraid of what would happen
if Hans were pulled into another direction. I definitely hope context
will take over from latex, but until then, I am cautious. I have
begged Hans to commission someone to write a latex->context converter,
which may fail on more complex code, but which would make the
transitioning from latex for people like me (with many latex documents)
a lot simpler. the name of "context" also makes it very difficult to
search in google for related documents. context documents appear to be
reasonable parseable, and there are even XML interfaces built-in.
hope this helps the agony...
/iaw
PS: of course, this gets me to my usual complaint---pdf* usually are
programs going from pdf to something else, like pdf2ps or pdfcrop.
only pdflatex/pdftex has this reverse. it should have been called
latexpdf/texpdf. at least, the texi2dvi stuff is logically named.
Will Robertson
ivo...@gmail.com wrote:
> I am struggling with the same question. Worse, I already know latex as
> an end user (not as a programmer) quite well, and rely on many hacks
> that I have accumulated over the years. So, here is my perspective.
There appear to be some misconceptions in what you say.
> For all practical purposes, latex seems to have stopped development at
> an ordinary speed. The latex3 project appears dead. (Progress appears
> mostly in pdflatex.)
You may wish to look in the LaTeX3 code repository and sort by date to
see how "dead" LaTeX3 is:
<http://www.latex-project.org/cgi-bin/cvsweb/experimental/expl3/>
Sure, progress is slow, but there ain't many people working on it.
> A latex document is also practically unparseable,
How so, exactly? Write your document specification well and it'll be as
parseable as XML. Look to gellmu for an example of XML that looks just
like LaTeX.
> and latex itself has no built-in feature that allows one
> to interface with other languages. (For many practical purposes,
> perltex is too slow, and other solutions tend to be weird or complex
> hacks [no insult intended] that are not relatively universally
> installed with latex distributions. I have begged repeatedly for a
> fast \system{} call, like the \write18 call.)
This is due to TeX, not due to LaTeX. If you wish to enable
--shell-escape on your machine, there's nothing stopping you. What more
do you need?
> context still seems to be alive, but I am afraid of what would happen
> if Hans were pulled into another direction. I definitely hope context
> will take over from latex, but until then, I am cautious.
What if LaTeX ceased being developed? Just before you were talking
about how new LaTeX developments don't appear to be happening (in your
point of view). I don't see how Hans going away would suddenly render
all your documents incomprehensible.
> I have
> begged Hans to commission someone to write a latex->context converter,
> which may fail on more complex code, but which would make the
> transitioning from latex for people like me (with many latex documents)
> a lot simpler.
Begged him to commission? I don't think he is financially rewarded with
more people using ConTeXt; on the contrary, his support costs would
increase. Why should he be interested in spending money to solve other
people's problems?
Besides, Brooks Moses is currently creating such a "converter", as far
as I've been lead to believe due to his postings to this newsgroup.
Actually, it is a module for ConTeXt to interpret LaTeX syntax, which
might not be exactly what you're looking for. You might wish to
correspond with him if you're interested in the project.
Will
ivo...@gmail.com
hi will---not too surprising that I do have some misconceptions.
the latex4 project began in 1994. this was the time of windows 95.
;-). so, to me, it seems as if latex3 has crawled almost to a halt.
this is not to diss the developers, who produced latex2e and many other
important features, and do all of the work for free. to me, it seems
more like evidence that latex is so obscure, painful, and complex in
its underbelly that it is slowly reaching the end of its lifecycle as
an actively advancing typesetting engine. I don't believe we will ever
see a latex4. fortunately, this does not apply everywhere. latex
stuff on top (packages) continue being developed, as is stuff below it
(pdftex).
you misunderstood my system{} request. of course there should be a
--shell-escape for any \write18 like functionality. but it is
impossible to write a general latex macro which continues latex
processing on the text that the invoked external shell script writes to
stdout. there are some kludges that work sometimes, but that's about
it. (we have had discussions about this earlier on this newsgroup.)
i think and hope that hans hagen will profit handsomely if conTeXt were
to take over the world. the key is "taking over big-time, though." if
money won't come through reputation and his company, it could come
through book royalties. so, he has a lot to gain from ConTeXt's
acceptance IMHO, and not just moral satisfaction.
hans is at the center of ConTeXt, which is both good and bad. if Hans
were to go off and work at google on other things, ConTeXt would
probably lose steam and thereby lose its appeal for many to switch
over. secretly, I am hoping that most of the latex3 team wizards (and
latex companion writers and others) were to endorse conTeXt and help
develop it.
I hope Brooks' work will make switching to conTeXt easier. but it is
not exactly what I and other potential switchers would also benefit
from. [there was a reason why MS Word included Wordstar and
Wordperfect to Word converters long ago before it had market dominance.
it made switching easier.]
regards,
/iaw
Will Robertson
I've replied to your email out of order, and snipped the rest of the
ConTeXt stuff; there's nothing for me to comment on there.
ivo...@gmail.com wrote:
> you misunderstood my system{} request. of course there should be a
> --shell-escape for any \write18 like functionality. but it is
> impossible to write a general latex macro which continues latex
> processing on the text that the invoked external shell script writes to
> stdout. there are some kludges that work sometimes, but that's about
> it. (we have had discussions about this earlier on this newsgroup.)
It seems your needs were more complicated than I anticipated. Something
like this
\documentclass{article}
\newcommand\typesetshell[1]{\immediate\write18{#1 >
\jobname.tmp}\input\jobname.tmp}
\begin{document}
This is the time and date: \typesetshell{date}
\end{document}
does indeed work for simple cases, but I'm guessing you tried this and
came up short. In this case, it looks like you should be writing your
whole TeX document with the shell script, and then sending *that* to be
typeset, rather than the other way 'round.
> the latex4 project began in 1994. this was the time of windows 95.
> ;-). so, to me, it seems as if latex3 has crawled almost to a halt.
> this is not to diss the developers, who produced latex2e and many other
> important features, and do all of the work for free. to me, it seems
> more like evidence that latex is so obscure, painful, and complex in
> its underbelly that it is slowly reaching the end of its lifecycle as
> an actively advancing typesetting engine. I don't believe we will ever
> see a latex4. fortunately, this does not apply everywhere. latex
> stuff on top (packages) continue being developed, as is stuff below it
> (pdftex).
There are too many errors within this paragraph to critique.
With all due respect, you clearly don't really know anything about the
current LaTeX3 project. I suggest actually going to the code repository
and seeing what they're up to. The fact that they've been a little
quiet over the last few years has no correlation with their current
progress.
Will
Jonathan Fine
ivo...@gmail.com wrote:
> the latex4 project began in 1994. this was the time of windows 95.
> ;-). so, to me, it seems as if latex3 has crawled almost to a halt.
> this is not to diss the developers, who produced latex2e and many other
> important features, and do all of the work for free. to me, it seems
> more like evidence that latex is so obscure, painful, and complex in
> its underbelly that it is slowly reaching the end of its lifecycle as
> an actively advancing typesetting engine. I don't believe we will ever
> see a latex4. fortunately, this does not apply everywhere. latex
> stuff on top (packages) continue being developed, as is stuff below it
> (pdftex).
Not many people know about LaTeX4. Here's a source.
Leslie Lamport wrote:
<http://lists.w3.org/Archives/Public/www-html/1994Sep/0000.html>
===
Building a new formatting language is also a formidable task.
However, there is no need to start from scratch. There already is a
widely used, reasonably adequate language: TeX. The easiest approach
is to use standard TeX plus some \special conventions for dealing
with the things that TeX doesn't handle--links, video, etc.
The problem with standard TeX is that it's not incremental. You can't
run TeX just on the page you want to display next. One can place
restrictions on how TeX is used that would make this possible--for
example, all global declarations must appear in a special place. Such
restrictions would be unnatural to a Plain TeX user, but would be
quite natural to a LaTeX user. If this route is chosen, I recommend
that the problem of enhancing HTML be merged with the LaTeX3 project.
A more ambitious plan is to design a new language in such a way that
most of TeX's typesetting engine can be used to display the output,
but in which the input has more of a markup flavor. A major goal of
this plan would be to integrate the viewer and the document editor, so
the user would have something more "WYSIWYG" when creating a document.
This would fit in with what I call LaTeX4, a long-term successor to
conventional TeX/LaTeX.
===
I stumbled on it only by chance a few weeks ago.
Let me highlight two parts of Lamport's vision
> the problem of enhancing HTML be merged with the LaTeX3 project
> A major goal of this plan would be to integrate the viewer
> and the document editor, so the user would have something more
> "WYSIWYG" when creating a document.
These statements were made in 1994. I believe that we have
missed a very valuable opportunity. However, it is still
not to late to do valuable and useful work in these general
directions.
Two key proposals I suggest for today are
a) Roundtripping LaTeX to and from XML
b) interactive use of TeX the program (the TeX daemon)
--
Jonathan
Jonathan Fine
<snip>
> ivo...@gmail.com wrote:
>
>>the latex4 project began in 1994. this was the time of windows 95.
>>;-). so, to me, it seems as if latex3 has crawled almost to a halt.
>>this is not to diss the developers, who produced latex2e and many other
>>important features, and do all of the work for free. to me, it seems
>>more like evidence that latex is so obscure, painful, and complex in
>>its underbelly that it is slowly reaching the end of its lifecycle as
>>an actively advancing typesetting engine. I don't believe we will ever
>>see a latex4. fortunately, this does not apply everywhere. latex
>>stuff on top (packages) continue being developed, as is stuff below it
>>(pdftex).
I have been following the LaTeX 3 project since almost its very
beginning. I'm even a one-time volunteer. (At the Prague EuroTeX
meeting in 1992(?) I volunteered to do some timing and performance
tests. Which I did, and learnt a great deal from.)
I agree completely with Ivo, and am very impressed that he knows
about the LaTeX 4 project.
> There are too many errors within this paragraph to critique.
> With all due respect, you clearly don't really know anything about the
> current LaTeX3 project.
Will, I don't see any errors in Ivo's paragraph. Perhaps you would
provide some examples.
> I suggest actually going to the code repository
> and seeing what they're up to. The fact that they've been a little
> quiet over the last few years has no correlation with their current
> progress.
The LaTeX 3 project as still coding in TeX macros. The future, in
my opinion, is to integrate TeX into a modern scripting language,
such as Python, as a loadable module. (By the way, the same
criticism applies to ConTeXt.)
It is over 10 years since the LaTeX 3 project started. Much has
changed since then. As far as I know, the goals of the LaTeX 3
project have not changed, to reflect the new situation.
Almost by definition, a project that delivers to goals that are
over 10 years old will deliver obsolete software.
--
Jonathan
Will Robertson
> I agree completely with Ivo, and am very impressed that he knows
> about the LaTeX 4 project.
>
> provide some examples.
Okay, I admit I exaggerated somewhat. I should have said "disagreements
in our thinking" rather than "errors". In fact, I erred when I thought
"LaTeX4" was something that didn't exist (but I still maintain that it
was probably a typo). Lamport's use of the term clearly is unrelated to
our discussion, which is centred now around LaTeX_3_.
As for LaTeX being "so obscure, painful, and complex in
its underbelly that it is slowly reaching the end of its lifecycle as
an actively advancing typesetting engine" -- yes, everyone agrees with
this to an extent. And that is why LaTeX3 provides powerful new
programming concepts to overcome this problem.
The details are described in
<http://www.latex-project.org/cgi-bin/cvsweb/experimental/expl3/source3.tex>
I haven't heard commentary on these ideas besides from (one or two or
three of the) members of the LaTeX3 team itself; I would be interested
on hearing the reactions of LaTeX2e programmers and ConTeXt programmers
on the new design. Clearly, there will be a somewhat steep learning in
the beginning.
But you next lament programming with TeX macros; the ideas in LaTeX3
add a layer of abstraction to make things different, hopefully for the
better.
> The LaTeX 3 project as still coding in TeX macros. The future, in
> my opinion, is to integrate TeX into a modern scripting language,
> such as Python, as a loadable module. (By the way, the same
> criticism applies to ConTeXt.)
>
> It is over 10 years since the LaTeX 3 project started. Much has
> changed since then. As far as I know, the goals of the LaTeX 3
> project have not changed, to reflect the new situation.
The appeal of TeX is an integrated environment. Obviously, it would
certainly be advantageous to try out new ideas within scripting
languages, but the 20 year accumulation of TeX knowledge is a very big
hump to overcome.
Perhaps my thoughts on the matter can be explained by something of a
thought experiment: TeX's advantage over other systems is that it has
good H&J, and is programmable. But the H&J stuff isn't really that
complex; the algorithms are well-known and easily implement-able these
days. The programmability is nice, but nothing that couldn't be
provided (and exceeded) by a scripting language, and TeX itself is
deficient in some areas, e.g., insertions (and output routines are
fiendishly complicated, so a "better way" would be, well, better).
Access to fonts, as XeTeX has shown, is probably better done these days
by hooking into the operating system. So: TeX with no fonts, no
programming, nothing necessarily outstanding for paragraph composition
is ... nothing at all?
My argument is that due to the long years of accumulated knowledge in
the community, TeX is more than a sum of its parts, and if it's broken
up, there's no point to still using the resultant pieces. That is, you
may as well start from scratch.
Or something along those lines. I'm still young, so my mind can't seem
to bring itself to conclusive thoughts most of the time...I didn't even
mean to involve myself in *this* particular discussion.
Will
casioc...@gmail.com
Jonathan Fine wrote:
> Will Robertson wrote:
>
> It is over 10 years since the LaTeX 3 project started. Much has
> changed since then. As far as I know, the goals of the LaTeX 3
> project have not changed, to reflect the new situation.
>
> Almost by definition, a project that delivers to goals that are
> over 10 years old will deliver obsolete software.
>
> --
> Jonathan
Hi Jonathan. My question is, why should it matter? As far as I can
tell, for me, it is a feature that latex is mature and stable. I think
it's full-featured enough, and if nothing changes for the next 50 years
I'll perhaps even be happier. This is one thing that I like about Unix,
mature technologies. I'm switching to latex in part because I don't
like that Microsoft Office changes every two or three years.
As regard to WYSIWYG, I don't like that. I like that I can edit a latex
document on any text editor, however small and simple the editor may
be. For this reason I have avoided using Lyx or Texmacs.
I should also state that I'm no fan of XML. I like the plaintext
terseness and simplicity of latex.
Had I wanted a WYSIWYG and XML I would perhaprs be using
OpenOffice.org.
Jonathan Fine
> Jonathan Fine wrote:
>
>>It is over 10 years since the LaTeX 3 project started. Much has
>>changed since then. As far as I know, the goals of the LaTeX 3
>>project have not changed, to reflect the new situation.
>>
>>Almost by definition, a project that delivers to goals that are
>>over 10 years old will deliver obsolete software.
> Hi Jonathan. My question is, why should it matter? As far as I can
> tell, for me, it is a feature that latex is mature and stable.
In some respects it is. However, look at hyperref - that seems to
be changing. Frequently, on this newsgroup, we learn about
incompatabilities with packages.
I'm all in favour of maintaining backward compatibility, where
possible. Let me ask a question. Do we have any tools that
tell us if new versions of packages will change the typesetting
of existing documents? I don't think we do.
If we had such tools, then I think people such as yourself would
be better protected against such change, and more willing to
move forward as part of a community.
> I think
> it's full-featured enough, and if nothing changes for the next 50 years
> I'll perhaps even be happier.
Many people felt the same way about LaTeX 2.09. And some people
still use it.
> This is one thing that I like about Unix,
> mature technologies. I'm switching to latex in part because I don't
> like that Microsoft Office changes every two or three years.
This is a good point. The stability of LaTeX is founded on the
stability of TeX. I am concerned, now that the LaTeX project
have decided that TeX is not sufficient to meet their programming
needs, and are using pdfetex.
> As regard to WYSIWYG, I don't like that. I like that I can edit a latex
> document on any text editor, however small and simple the editor may
> be. For this reason I have avoided using Lyx or Texmacs.
This is your personal preference. Other prefer visual editing tools.
I think we want a system that unites the two approaches.
> I should also state that I'm no fan of XML. I like the plaintext
> terseness and simplicity of latex.
I'm a fan of both XML and plain text, for example wiki markup.
> Had I wanted a WYSIWYG and XML I would perhaprs be using
> OpenOffice.org.
I'd like OpenOffice to be using TeX as its typesetting engine.
One final point. If the inputs and outputs of our typeset
documents are properly specified, then we may be able to
ensure backwards compatibility. This, I think, is a very
important point.
--
Jonathan
Jonathan Fine
<snip>
> As for LaTeX being "so obscure, painful, and complex in
> its underbelly that it is slowly reaching the end of its lifecycle as
> an actively advancing typesetting engine" -- yes, everyone agrees with
> this to an extent. And that is why LaTeX3 provides powerful new
> programming concepts to overcome this problem.
>
> The details are described in
>
> <http://www.latex-project.org/cgi-bin/cvsweb/experimental/expl3/source3.tex>
>
> I haven't heard commentary on these ideas besides from (one or two or
> three of the) members of the LaTeX3 team itself; I would be interested
> on hearing the reactions of LaTeX2e programmers and ConTeXt programmers
> on the new design. Clearly, there will be a somewhat steep learning in
> the beginning.
Well, you've prompted me to download and build this document.
And just to prove I done it, there's and error in the compile
===
! Undefined control sequence.
l.282 \def_new
:Npn \iow_newline: {^^J}
?
===
Maybe I'll post a review later in the week.
--
Jonathan n
Jonathan Fine
> In some respects it is. However, look at hyperref - that seems to
> be changing. Frequently, on this newsgroup, we learn about
> incompatabilities with packages.
Sorry - serious typo.
I meant to say "incompatibilities _between_ packages"
I did not mean to give hyperref a special place here.
--
Jonathan
Will Robertson
Jonathan Fine wrote:
> Well, you've prompted me to download and build this document.
>
> And just to prove I done it, there's and error in the compile
> ===
> ! Undefined control sequence.
> l.282 \def_new
> :Npn \iow_newline: {^^J}
> ?
> ===
I neglected to mention that it requires LaTeX3 itself.
Sorry about that...
Will
Jonathan Fine
Well, if it it compiles scarcely without error on LaTeX2e
then either
* that's not much new in LaTeX3
or
* it's got pretty good backwards compatibility
(smile)
--
Jonathan
casioc...@gmail.com
Jonathan Fine wrote:
> casioc...@gmail.com wrote:
> > As regard to WYSIWYG, I don't like that. I like that I can edit a latex
> > document on any text editor, however small and simple the editor may
> > be. For this reason I have avoided using Lyx or Texmacs.
>
> This is your personal preference. Other prefer visual editing tools.
> I think we want a system that unites the two approaches.
>
> --
> Jonathan
Thanks Jonathan for the reply. I must say, had Latex been a WYSIWYG
visual system I wouldn't have become interested in it. I find WYSIWYG
too limiting and distracting. I thought the point of Latex was that I
just concentrate on content and not worry about how it looks, the
compiler will take care of how it looks later on, and I don't find it
tedious to put a "this is a chapter" tag every chapter or so (though I
would've probably found it tedious had it been xml). I like that I can
edit a latex document on anything even a palmtop, have it as a
plaintext file, and compile it whenever. I like that the plaintext file
is recoverable, less prone to corruption, will be readable anytime in
the future, and is portable across platforms.
Right now I'm falling in love with the command line interface and the
plaintext data format. I like that I don't have to dig through menues,
switch between keyboard and mouse, and worry that something I want is
not in the menu or not ergonomically situated. I would much prefer a
CLI and a reference book, and in a little time I'll learn those
functions I need that I won't need the reference book. And I like that
I can use the same little editor and CLI for my text
editing+typesetting, for my statistics work (R project with it's huge
reference book), and for textfile databases. I feel I have replaced MS
Office with something far simpler yet far more versatile, powerful,
portable, and permanent.
I would understand though that WYSIWYG would be useful for something
like mathematics.
Robin Fairbairns
>PS: of course, this gets me to my usual complaint---pdf* usually are
>programs going from pdf to something else, like pdf2ps or pdfcrop.
>only pdflatex/pdftex has this reverse. it should have been called
>latexpdf/texpdf. at least, the texi2dvi stuff is logically named.
iirc, the original was called tex2pdf. i don't believe it was ever
released under that name.
i'm sure you'll get over this minor blip in nomenclature, given time.
--
Robin (http://www.tex.ac.uk/faq) Fairbairns, Cambridge
Dan
No it doesn't. I, too, was prompted to download it and I, too, got the
same error message, but I simply corrected it.
Open l3io.dtx and change lines 281 and 283; from
% \begin{texnote}
\def_new:Npn \iow_newline: {^^J}
% \end{texnote}
to
% \begin{macrocode}
\def_new:Npn \iow_newline: {^^J}
% \end{macrocode}
taking care that there are exactly 4 spaces between "%" and
"\end{macrocode}".
Dan
Morten Høgholm
> Open l3io.dtx and change lines 281 and 283; from
> % \begin{texnote}
> \def_new:Npn \iow_newline: {^^J}
> % \end{texnote}
> to
> % \begin{macrocode}
> \def_new:Npn \iow_newline: {^^J}
> % \end{macrocode}
> taking care that there are exactly 4 spaces between "%" and
> "\end{macrocode}".
Fixed in source, thanks.
--
Morten