Slashdot Mirror
Modern LaTeX Replacement?
javierzinho writes "For many years I have been using LaTeX to compose scientific documents, but truly I am getting tired of its complexity. You have to install new packages for new features, compatibility issues are everywhere, you need to know commands for everything, table composition is torture, image insertion is an odyssey if you don't have the 'right' format, and you need to be a LaTeX Jedi master to create a new document class. I'm looking for a document processor (not a word processor) that is a viable replacement for LaTeX, possessing all of its advantages — consistency between text and math text, automated cross references, direct PDF creation, etc. — but that is not stuck in the 1980s with the compiler metaphor and weird font technology. An application with visual interface and so on. I've tried Scientific Word and Lyx but both are front-ends for LaTeX. Publicon only produces PDF files by exporting to LaTeX and subsequently using pdflatex. Add-ons for MS-Word are a joke, and webEq is intended for web publishing, not for PDF production. Does anybody know of a decent, scientific-structured document processor that is a modern application?"
I can't help but question the complaints on the complexity. I generally have a repertoire of packages that I use frequently like the ams packages, pstricks for image drawing, beamer for powerpoint-like presentations, and the external program image magic to make pictures the correct format.
Using other packages periodically tends to not have too many conflicts, except when trying to conform to required document classes of certain journals. But the workarounds generally don't take too much time.
I have yet to find something as robust as LaTeX, yet relatively user-friendly. Then again, I've never tried to create my own document class, merely modified what is already there. That always seemed to be the domain of the nuts-and-bolts programmers rather than the people who just want a typsetting language. So my idea of "user-friendly" may be a little skewed.
Having used LaTeX to typeset my dissertation, I share these concerns about LaTeX. The documents it produces are beautifully typeset and look great -- especially for math. The notion that the writer is agnostic of the typesetting procedure and methods with LaTeX is a complete lie. I've never had to worry about ratios, measurements, indentations, word-per-line, empty pages and other problems as I have in LaTex. LaTeX submissions to journals are becoming less and less available -- in physical chemistry and chemistry journals at least.
There is a large and important market for high-quality typesetting software with excellent math functionality. More importantly, something which interfaces with bibliographic software well, and produces high quality PDFs. (Bibtex does a decent enough job, but I find that it's plagued by the same problems as LaTeX.)
I've searched for an alternative as well, and I'm quite sure that none exist. I haven't seen other type setting documentation formats for journal submissions, which I think is an important hint.
I understand your qualms with LaTeX as a long time user, but given the alternatives I find it better (though word processors are easier to use, LaTeX makes things much prettier).
A word processor front end (let's pick Open Office Writer as an example) with a LaTeX backend would be a good mix, but also give you the downside of WPs, namely constant layout fiddling instead of focussing on content.
I don't quite understand your complaint about the way LaTeX is structured wrt packages. It's pretty much the same thing you see with Firefox where you have a core program with lots of useful plug-ins for added functionality, and as such it's the same argument as it has.
We do not live in the 21st century. We live in the 20 second century.
The problem is that all the things with decent interface have crappy quality of output. Truth is, latex (tex really) have far FAR better output than anything else. Nothing comes close in terms of typesetting text and math correctly. I can spot a word document once it's printed. Not by the font, but by text layout. Reading something written in a gui word processor like word (or openoffice) hurts your eyes and your brain.
Plus, your problem was the interface. So why not consider something that outputs latex? It needs to be a front end that handles all the dirty work and uses latex for what it does best. Just like you don't care that most of your operating is written in C which is just as old technology.
Plus, most places that want mathematics documents, really want you to submit latex. You're better off with something that can output it natively.
Writing something that does the same thing is stupid if what is wrong is an interface. If a good interface is written, you might never know you are using latex (or tex) in the background.
I've taken a couple of stabs at LaTeX through the years. I have no real need for a proper type-setting platform like LaTeX because I am not in the world of academia that demands it, so I was never able to get past the learning curve imposed by LaTeX.
Now, let me say... I get it. I understand how invaluable it is to submit a paper in a format so less time can be wasted "making it pretty" and more can be spent on the meat of the work. That fact doesn't elude me.
What I never figured out was how to download a stinking template from IEEE and start writing a document. I never figured out how to compose my own document type so I could use it to empower the written arts that I am interested in. I never got past the hurdle, so to this day I still use OpenOffice Writer as my word processor and haven't been able to "transcend" to a proper type-setting program so make all the boring formatting tasks easy.
I even read the LaTeX Wikibook a number of months ago and this didn't even get me over the hump on my way to publication.
So, I echo the sentiments of the article submitter. LaTeX is hard, and either better documentation or a better alternative is needed to make it accessible to the rest of us.
Support the 30 Hour Work Week!!!
What is it about LaTeX that makes it so special? Can't scientific documents be laid out correctly in a word processor? I ask out of ignorance, not rhetoric.
Admit it. You post strawman arguments as AC so you get modded Insightful for refuting them, rather than Troll
So tell the AAAS that... http://www.sciencemag.org/about/authors/prep/docx.dtl "Because of changes Microsoft has made in its recent Word release that are incompatible with our internal workflow, which was built around previous versions of the software, Science cannot at present accept any files in the new .docx format produced through Microsoft Word 2007, either for initial submission or for revision. Users of this release of Word should convert these files to a format compatible with Word 2003 or Word for Macintosh 2004 (or, for initial submission, to a PDF file) before submitting to Science.
Users of Word 2007 should also be aware that equations created with the default equation editor included in Microsoft Word 2007 will be unacceptable in revision, even if the file is converted to a format compatible with earlier versions of Word; this is because conversion will render equations as graphics and prevent electronic printing of equations. Regrettably, we will be forced to return any revised manuscript created with the Word 2007 default equation editor to authors for re-editing. To get around this, please use the MathType equation editor or the legacy equation editor included in previous versions of Microsoft Word, which can be accessed from "Insert Object" from the "Insert" ribbon in Word 2007."
Um... I don't think Word is an answer
Hi, I recently finished a PhD in math; while I probably haven't Texed as many pages as you, I have plunked out my share over the years. I've found the unix/linux front end program Kile to be extremely labor saving, particularly its newest version. It has forward/backward search, automatic completion for \ref commands, and a built in library of click-to-use symbols (and for these you are automatically advised of what packages are needed to use them.) I am agnostic on the issue of whether something "better" than LaTex is possible, or whether with great power always comes irritating details, but for what's out there, I think Kile can greatly improve the experience.
I thought that math was actually pretty reasonable in OO. I usually use it to write short equations, but I found that it was convenient to have the functionality to program a math function rather than deal with some wissywig editor. If I needed to write any moderately complex equations though, I'd probably go with LaTeX hands down. Learning it is a pain, but you have a bit more control when dealing with various markups and macros.
You should think of TeX as a slightly high level description language for your document, eg if PDF (say) takes the role of machine languague, then in this analogy TeX would be C and LaTeX would be C++, and LyX would be like Visual Studio. With this analogy, we can see the flaw in your question: there's nothing wrong with these tools, other than the fact that you're no longer willing to use them, because you want something even higher level.
You really have two choices depending on your temperament: If you like to have control of all the layout details, then you should learn the tools properly and start taking advantage of the features to simplify your workload dramatically (you obviously don't know the tools well enough or you wouldn't complain about document classes, table composition, etc.) I suggest you learn how to use macros, and maybe read the TeXbook. In this way, you will be able to grow your own high level interface to LaTeX which will suit you extremely well. Since you've used LaTeX for years already, this is a good investment.
If however you're happy to delegate the fine tuning of your documents to the software, then your other choice is to give the LyX developers some feedback on what you'd like to see, or wait for a better front end to come out, which hides the complexity even more than LyX. Those things happen every once in a while, but they invariably introduce complications that make life more difficult when working on a joint paper together with other people. Try TeXmacs if that's what you want.
[disclaimer: I am the main author of LilyPond, a system that you could easiest describe as "LaTeX" for music notation]
The problem is not with TeX. Knuth is mostly as brilliant as people say he is. The problem is that
1. the extension infrastructure of TeX is very outdated (WTF, a macro expansion language?)
2. the development ecosystem around TeX is filled with souls that are of lesser stature than Knuth. They're mostly people that need to write mathematics (physicists. mathematicians), as opposed to people that know how write software.
LilyPond back in the day used TeX as a backend engine, and I vividly recall all of the brokenness I encountered in the support-tools that surround TeX (dvips, xdvi, etc. etc.). Things have gotten a lot better now that we have pdflatex - it cuts a whole truckload of crappy tools out of the document pipeline.
Font handling remains atrocious. In case you're wondering: someone was bright enough to base parts of the fontsystem on the DOS 8.3 restriction, so URWGothicL-Demi is and will be called uagd8a forever inside TeX -and worse- if you have to add a modern (OTF, TTF) font, you have run scripts to make LaTeX's font subsystem understand these files in terms of the ridiculous naming scheme.
People get hung up over TeX's beautiful formatting algorithms, but they are not actually that complicated, and by todays' standards TeX is a small program: tex.web is just 25k lines, and that includes its ample comments. LilyPond has page layouting and line breaking that is far more complex.
The real problem with typography, whether for music or documents, is that it's full of traditions that predate automatic processing, and are not specially suited to computerizing. For example, in some language words change their spelling/typography when they get hyphenated (eg. the German eszet letter which hyphenates to s-s).
IMO The challenge is designing the software such that these idiosyncrasies can be captured effectively without hardcoding them, so people can create their own idiosyncrasies.
As for the original poster's question, the system that looked the most convincing to me is Lout, but I have never tried it out.
Han-Wen Nienhuys -- LilyPond
How many times can one person +5 for saying the same thing repeatedly in the same topic?
I don't know if this is the case for you, but I find most people who find LaTeX hard are using it wrongly. Specifically, they are trying to precisely control the formatting, placement, etc. etc. of everything in their document. This is, pretty much, how you use today's WYSIWYG word processors. It's very cumbersome and arcane to do the same in LaTeX, and the results don't usually look very good in either case.
The right way to use LaTeX is to basically enter the semantic structure of your document, let LaTeX do all the typesetting, and then tweak it a bit as necessary. Realizing this was the point where I stopped fighting with LaTeX and started letting it work for me. I've been getting compliments on how beautiful my documents are. There's a lot of typesetting knowledge encoded in LaTeX, and, really, it probably does a better job than most of us can hope to do. One particular example I like to share is that, when I took my thesis to the printer, he remarked how glad he was that, finally, someone had thought about making the margins large enough that the text would be readable once printed and bound. I hadn't. But LaTeX had.
Incidentally, the above is also why I don't see a lot of value in WYSIWYG editors for LaTeX. On the one hand, being able to see what your final document will look like while you are creating it is good. On the other hand, it makes it very easy to fall into the trap of spending all of your time correcting this or that perceived layout error, instead of getting your actualy work done while letting LaTeX do the typesetting. I am not even sure WYSIWYG can be made to work right; a lot of algorithms in LaTeX are simply slow, and changing even one letter can cause your text to jump around, which is very annoying while editing.
Then, of course, there is the matter of commands. I recognize that having to type in commands is a significant hurdle for many people. Being a programmer and having a lot of experience with HTML, this isn't the case for me - I am used to using commands. As a programmer, I actually see LaTeX as having an advantage here: by defining new commands, you can automate repeating tasks and increase the maintainability of your code...err...document. I don't actually do this a lot, but it's very nice to have that ability for when it's useful.
All in all, I won't deny that LaTeX is hard. I know it is. On the other hand, I am not actually sure it is harder than Microsoft Word, which, in my experience, is its main competitor. Although Word is probably easier to get started with, learning the basic LaTeX necessary for creating a simple document is really not that much work, and the documents you produce will look a lot better than what Word produces. When you get to more complex documents, I find Word has a tendency to screw up - it will crash and/or eat parts of the formatting or content of your document. Granted, that's bugginess, not something inherent in WYSIWYG word processing, but it still ends up causing you a lot of frustration and losing you a lot of time. I've never seen LaTeX do this, and, even if it did, you would still have the source code of your document - at the very least, all your content is still there.
So, there you have it. My opinion, my experience, with input from quite a few others - LaTeX users, non-LaTeX users, and "I tried LaTeX but couldn't figure it out" users. In the end, my conclusion is that LaTeX is far from perfect, but it's still the best.
Please correct me if I got my facts wrong.
Actually it was the exact same thing for me. I did my masters in MS-Word in 98 and i hated it (open the document and all images shifted position.. .*ARGH*).
When I started on my PhD, a colegue told me about LaTeX and gave me the 'Not so short guide' and I started writing using a simple template as starting point. It worked like a charm. Sure I have had my times where I had to fight a bit to get what i wanted (especially when I had to install the institute style class in order to finish my thesis).
Also I have done papers in MS-word and LaTeX, and anyday I would choose Latex for an article, as getting the margins etc right for a MS-word publication is REALLY a pain whereas the journals supporting LaTeX have done it extremely simple by publishing style-classes. This makes submitting articles for review (where the layout must be different than the finished articel, e.g double linespacing etc.) MUCH less of a hasle than in Word/OO-Writer.
So if you are serious about using a typesetting program, use LaTeX!
Yours Yazeran
Plan: To go to Mars one day with a hammer.
I'm not submitter, but I share some of his frustration.
First, I've used LaTeX since around 1987 (that was on the Amiga, btw) and I'm still convinced that for most of the documents I write LaTeX is clearly superior to other systems I've tried.
That's not to say there aren't some obvious points for improvement. Font support is archaic and really not something from this decade (nor the previous). I understand that XeTeX has some improvements in this area, I'll check it out. I need/want to use the Gentium font but after jumping through all the hoops that are indicated in the sparse documentation, it no worky.
But, for me, the number one frustration is the sheer impossibility to create a new base class. I write software for a living and have used many obfuscated languages. However, I just don't "get" the intricacies of programming for LaTeX.
I would love to have a letter.cls that doesn't look like it's an afterthougth. There is dinbrief.cls, but that has all sorts of problems of its own. I've tried several times to create a letter class, for A4 paper, that looks professional to use in a business. No such luck, I just couldn't do it. And the lack of alternatives on the 'Net seem to indicate that there aren't many others that could do it either.
So, to summarize, LaTeX a wonderful tool for typesetting reports and articles. Especially if those are heavy on math. But for other correspondence it isn't so great, or at least, it hasn't kept up with modern developments in font technology and document design.
Can it match the styles advocated by the various journals?
Actually, it wasn't so much made for European languages. It was made for typesetting in general. In fact, look to the KOMA-Scripts package, which was designed because European publishing wasn't originally well accounted for in TeX.
There are a number of other benefits which are, perhaps, why it looks more "bookish". Kerning, ligatures, finer control over hyphenation, glyph variants, real fonts, support for semetic languages, support for asian languages, etc. Take a look at the index of Knuth's Art of Programming. Arabic, Chinese, everything beautifully typeset. The Index seamlessly generated with appropriate sorting.
I suppose it's possible to look at professional typesetting and say, so what? In the end, though, that's the benefit. TeX is a typesetting system and a lot of people seem to want a Word Processor. These are different things.
There are things that are possible in TeX that aren't possible in Word. You have more control over the document (although Word certainly gives an appearance of control). TeX can make type that looks GOOD. However, some people think of text and publishing as commodities in the online world. They decide that they can live without these things.
That's fine with me. Personally, I enjoy seeing something done really well. I use TeX as it suits me. Someday, maybe someone will make a good WYSIWYG typesetting system. Until then, we have TeX.
I think Mauve has the most RAM. --PHB (Dilbert Comic)