Monday, June 21, 2010

\title{LaTeX}

\documentclass{whiny blog post}

\section{Introduction}

I spent some time this weekend thinking about the costs and benefits of switching to LaTeX. It seemed to me that it would make a better post if I got some empirical data concerning LaTeX, and so I downloaded MacTex and attempted to convert a paper. I followed Charles Tanksley's instructions, which were helpful.

\section{Pros}

It's nerdy.

The finished product looks super awesome. And you don't have to think much about how it's going to look; you just write the content and LaTeX automatically makes it look awesome. This is supposed to save you time and stress.

It is alleged to render equations and other symbols nicely. Perhaps I chose my conversion project poorly, but the paper I chose to convert to LaTeX did not contain many weirdo symbols or any equations at all.

Apparently it is easier to manage bibliographies and references. There's a heavy up-front investment, though. Also, question: suppose you have to reformat your bibliography to conform to some journal's idiosyncratic style. Does LaTeX make this easier or harder?

You can make 1 simple formatting change that instantly applies to the whole document.

You are not subject to the whims of the MicroSoft Corporation. Their decisions about the viability of the .doc format are not going to fuck up your shit. Nor is their decision to totally stupidify the user interface for Word.

Edit: In comments, Kevin Klement mentions several other pros:

But there are a lot of other reasons to prefer LaTeX you don't discuss. The maximum portability of plain text formats. The non-proprietaryness. Heck, even the fact that plain text formats are much less likely to be corrupted than encrypted binaries ought to be mentioned.


End Edit

\section{Cons.}

You have to have Word (or something like it) anyways, because your students have word. You have to have something that will read their documents. Plain text editors won't do the job.

Frankly, I just don't see the point about how much time you'll save if you don't have to think about formatting. For one thing, I don't spend that much time thinking about formatting. Titles; paragraph breaks; blockquotes; offset definitions and principles; section headings; I spend almost no time thinking about these things. For another thing, you have to think about them in LaTeX, too, because you have to give a '\section{x}' command in order to get LaTeX to know to put a section heading in. You still have to write the code correctly.

Now, if you have to reformat something for a journal, say, you might save time by applying 1 simple formatting change that instantly applies to the whole document. But you can do that in Word, too, if you've been using your headings right (which is no harder than doing it in LaTex--anyone who could figure out LaTeX could figure out Word). And I bet I could just reformat the section headings by hand one at a time in less time than it would take me to identify the 1 simple change I have to make by doing a google search, finding the relevant information, applying the relevant procedure, testing it to make sure it worked, and then fixing any bugs that might have come up. (For example, I attempted to create subscripted text. I had to find instructions on google, and then implement the code I learned, and then identify and repair several coding errors--I have trouble remembering to hit curly brackets rather than parentheses, though I'm sure this will go away with familiarity. Then, when it still wasn't working, I had to do more googling in order to learn that the original set of instructions left out that I had to be in Math Mode in order for the subscript command to work properly, and that in particular I had to be out of math mode in order for the text after the subscript to look right. In Word, this would have taken at most 1 second, since I have a subscript button on my toolbar and I know the shortcut keystroke that results in subscripted text.)

It seems to me to be easier in Word to make 1 simple formatting change that instantly applies to exactly as much or as little of the document as you want it to.

Although I admit that equations are a problem with Word, regular logical symbols and notations are not. Although it could be easier to navigate, there's an insert symbols menu, and you can program shortcut keys for the ones you use the most.

It is easier to annotate a word document. This makes it easier for someone to comment on a draft, and also for you to connect the comment to the appropriate location in the paper.

And if your friend doesn't annotate your document, the page numbers will still be meaningful if she's looking at your Word document. But there seems to be no inherent connection between the pages in the typeset PDF file that LaTeX creates and the text in the .tex file. In fact, the document you print out does not look at all similar to the document you edit. All of its formatting and pagination characteristics are completely different. This makes it difficult to find and edit individual paragraphs and sentences. Much more difficult than simply going to page 12, paragraph 2, sentence 3, or whatever.

Word's GUI is (somewhat) intuitive and doesn't require you to memorize a bunch of codes. However, the new version of word is substantially crappier. I don't want to be the guy who defends MicroSoft Word or anything, because I have spent a huge amount of time so angry with MicroSoft Word that I wanted to kill the balls off of it. But a lot of the problems with Word that I see people saying contributed to their abandoning it in favor of LaTeX are easily solved and seemed to me to stem from not knowing how to use it.

Although, you can make 1 simple formatting change that instantly applies to the whole document, it is often not particularly easy to know how to make that change. Suppose I don't want my paragraphs to indent, and I want to demarcate paragraph breaks by adding an extra space between them. How?

Footnotes in LaTex are kind of a pain. And although I'm no Kripke, I use a lot of footnotes.

Quotation marks are kind of a pain. And because I'm no Kripke, I use a lot of quotation marks. (Zing!) (Just kidding.)

Word count. As I was writing this, I realized that Word has a pretty sophisticated word count utility. My LaTeX editor has a "statistics" utility that counts words. I hope it counts only the words in what you intend to be the text, and not all the extra formatting junk. I know that Word's does this, and Word's utility allows you to exclude footnotes easily. Does LaTeX's? Also, Word's word count allows you to select blocks of text and get stats on the selection. My LaTex editor doesn't seem to. The word count utility on MS Word seems to be pretty superior.

In conclusion, although there are some cool features here, I don't think there is enough to compel me to switch.

--Mr. Zero

\end{document}

34 comments:

Anonymous said...

There must be a LaTeX meme going around as I've also been playing around with it lately. My test subject is my CV since that seems to be the place where pretty matters the most.

According to colleagues in the sciences, most journals have LaTeX style sheets that you can download and more or less instantly reformat your essay. Not sure this is true for philosophy.

\usepackage[parfill]{parskip} % begin paragraphs with an empty line rather than an indent

Anonymous said...

no wai!!!!

Kevin Klement said...

There are lot of mistakes in reasoning here. You attribute certain things to being easy, when you clearly mean, "it's not what I'm used to." And some of the things you had to spend time Googling clearly you'd never have to do again, so you're miscalculating the amount of time it takes. After a period of time, you won't be doing things like that anymore.

Yes, there is a learning curve, but even the most cursory introduction would have covered subscripts, superscripts and footnotes. In what way are footnotes more difficult than in Word? It isn't. Substantiate your claims.

It is not true that you need to own a copy of Word. I don't. Heck, Word won't even run on my Operating System. I do keep a copy of either OpenOffice or Abiword installed, but they're free. However, I rarely use them. When a student submits a paper to me, I convert it to LaTeX, which standardizes the font size and margins for me, so students can no longer play tricks. This is all done by a script, applied to every student paper at once.

If you're not now thinking in terms of sections and subsections, and forcing you to write "\section{...}" forces you into thinking like that, that's a GOOD thing, since you SHOULD be thinking in terms of sections and subsections when you write. If you're not thinking about that now, you're not doing your job as an author.

It's also true that you can duplicate the differentiation of form from content using styles in a word processor to some extent. Though their use is not encouraged, and setting things up to use them rather than direct formatting commands standardly is not that easy in Word--it just isn't. Word has a quick-key for entering italics. It does not have a quick-key for entering emphasis, or anything else semantic. Yes, Word can do this, but it takes more work.

And even if you do use styles, it's not true that you can use Word to quickly "match" the final output of some journal, however, because no professional journal typesets using Word. (Word doesn't have a sophisticated layout program that correctly handles hyphenation, kerning, whitespace usage, to the same degree or quality as LaTeX does.) So if you use Word, it will always be the case that the typesetters have to redo what you've already done. That's a money drain for the profession. On the other hand, if you submit LaTeX code to a journal that uses LaTeX to typeset, they can typeset your article in a manner of minutes. I've seen it in my own experience. If I submit a Word doc to a journal, it always takes a month or more to get page proofs back. If I submit LaTeX code to a journal that actually uses LaTeX to typeset, I often get page proofs back that week, or even the same day.

LaTeX does not count words. That's a feature of your editors. Some editors have good ones. Some don't. You didn't even mention which one you were using. But certainly you can get a word counter for LaTeX to exclude footnotes. The TeXcount perl script, as implemented e.g., here separates these. Don't blame LaTeX for the fault of your editor.

But there are a lot of other reasons to prefer LaTeX you don't discuss. The maximum portability of plain text formats. The non-proprietaryness. Heck, even the fact that plain text formats are much less likely to be corrupted than encrypted binaries ought to be mentioned.

Anonymous said...

After a lot of haranguing from several friends in my department, I tried LaTeX for about, oh, three months. I then went back to a WYSIWYG editor. (Mellel is my word processor of choice. And if I weren't using Mellel I'd use Pages. I have a deep and perhaps irrational hatred for Word.)

Why did I give up on it? Because I began to realize that formatting helps me think and structure my thoughts. I want my footnotes to be on the same page I'm writing but to be clearly separate from the main paragraph. I want to be able to clearly see headings (because they're bolded!) and sub-headings. I like being able to see the paragraph breaks and indentations and because it helps me see the structure of my argument. Same with blockquotes.

By losing the formatting, I was just staring at what looked like a jumble of words (and code. Ugh.) without any structure. And I didn't like it.

So it's back to WYSIWYG for me.

Anonymous said...

Nobody's mentioned LyX yet? It seems to me that it has most of the benefits of LaTeX, but also many of the benefits of Word.

What do the hardcore LaTex-heads think of LyX?

Anonymous said...

I also tried LaTeX over the weekend (again, having tried it two or three times previously over the past couple years). In the end, I decided that it wasn't worth the time, and went back to my word processors (I waffle between using Nisus and Mellel).

Most of the supposed advantages of LaTeX don't strike me as very compelling. The word processors were already working fine for me, and there were no problems that LaTeX was solving (as for the whole "the final product looks better" point, I am left cold. Word-processed papers don't look that bad). And I don't want to have to learn, or look at, the code.

Another problem I have with LaTeX--it is weird the way that some of its users defend it with a sort of cult-memberesque ferocity. Live and let word process, I say.

Anonymous said...

To the bibliography question: this is actually one of the distinct advantages of LaTeX. The bibliographic database is stored in separate fields, and there is a separate style file that determines how references are represented in the bibliographic section. In at least some areas of the sciences, journals will supply a bibliographic style file. So the work in formatting the bibliography can be limited to just specifying the appropriate file. Submit somewhere else, just change the style and you're done.

Kevin Klement said...

LyX is OK for a fairly simple document. I like to define a lot of my own commands, especially for a logic-heavy document, and with that LyX gets tricky.

Of course, I don't have trouble seeing my document. I always keep the preview panel open. Some LaTeX editors (like gummi or the whizzytex plugin for emacs) actually provide a live-updating preview, and even those that don't, refreshing the preview is just a button away.

And as for the source mark-up (--calling it "code" is an overreaction; tell a programmer that writing HTML is writing in "code" and watch them laugh--) you can even set apart or boldify or whatever the section headers there too, at least with a decent editor.

Am I passionate about LaTeX? Yes. This is in part because of the vast potential savings it could have for our libraries, and how it could shorten publication times.

But I'm also someone who is thinking ahead. WYSIWYG only makes sense if your document is only going to displayed in one way. In the next century, though, people aren't just going to want to print and read. They're going to want to read documents on their portable devices, or have them read out loud, you name it. It is really only possible to have a flexible document if it is *semantically* encoded, and not designed within the confines of a single viewpoint.

In some of my own projects, I'm distributing up to 10 different versions of the same document (PDFs designed for four or five different sized pages or screens, ePubs for portable devices, HTML for the web, etc.). It is a nightmare to reformat documents that were designed as WYSIWYG to look good on other displays, because, after all, the author was only focused on what it looked like on the WYSIWYG display. But if the document contains semantic mark-up, adapting it to a different screen or page size is trivial.

It IS kind of possible to have similar results if you use styles consistently in a word processor, but who does?

Anonymous said...

I think it's easy to understate the "whims of Microsoft" issue. There is a large set of open source tools that add functionality slowly without substantially changing their interface (with "\section{}" being a sort of LaTeX interface, for the sake of this discussion). So while these tools require a higher initial investment to use them, that investment pays off over a much larger period of time.

Commercial software tends to change more in interface, partly to justify repurchase. Much of repurchasing is driven by OS upgrades that interfere with compatibility, but if the new version were just the same old version for a new OS, people would be angrier about having to replay. Thus you download the latest version of Word and it's doing some funky new toolbar thing, and you have to invest a bunch more time figuring out what's going on. Someone with a LaTeX and, say Xfig toolchain (some ffields have a strange phenomenon called "diagrams") may have faced such problems only once or twice in 20 years, such as the switch to LaTeX2e that changed some of the class loading directives. This differnece becomes more important when it comes to the more advanced stuff you can do in Word, which has support for automatic indexing and structural markup and the like. The way you do that stuff changes too, so the more stuff you work with the harder it is to deal with the changes they've made.

That said, TeX has always been a bit of a mess (if you get into a situation where you need to work with variables, there are several varieties with different ways of setting them), and LaTeX is only a partial improvement. Doing useful things like supplying margin changebars requires outside support from scripts, and is therefore difficult to get working outside of a unixesque installation, and the pdf switch broke support for that anyway. It's a mixed bag.

Hanuman said...

Thanks for taking the time to write this up, Zero -- and for offering your rebuttal, Kevin. I haven't actually tried LaTeX, but now I think I will.

Anonymous said...

In his first post Kevin Klement says that he doesn't need to have Word on his machine, because he can convert his students' papers to LaTeX.

If this is the case, then most of his other arguments in favor of LaTeX seem like potentially good arguments in favor of LaTeX existing in general (and thus some people being able to use it), but not necessarily in favor of any of us in particular using it (because our word-processed papers could be converted in the way that his student's papers can).

Anonymous said...

Somewhere Marshall McLuhan is smiling because of this thread:

http://en.wikipedia.org/wiki/Tetrad_of_media_effects

Obviously the X in La Tex is the tetrad.

Something about this whole conversation makes me want to write in longhand. With my Smartpen, of course.

Laura Cooper said...

I will \textbf{never} use Word again. Taking a 5-week class on it made me hate it \emph{and} its intricacies (intricacies I was blissfully unaware of before).
Why can't your students submit PDFs?

Mr. Zero said...

Prof. Klement,

There are lot of mistakes in reasoning here. You attribute certain things to being easy, when you clearly mean, "it's not what I'm used to."

I may have made some mistakes in reasoning here. I don't think that this was one of them, though. Sometimes people say that things are hard to do in Word, when they seem to me to be easy. Tanksley, for example, mentions Word's automatic formation of numbered lists and problems with footnote formatting. I wasn't there when they happened, but when things that I would describe that way happen to me, I find them to be pretty easy to fix.

Whereas, when I wanted to change the behavior of my paragraph breaks in LaTeX, I couldn't see how to do it, and a google search wasn't super helpful. Anon 12:23 solved that problem, and so now I'll never have to worry about it again--thanks Anon 12:23!

Point being, as you say, there is a learning curve. It will take time to gain the proficiency with LaTex that I currently enjoy with Word. Maybe I'm wrong, but my initial investigation did not convince me that it was worth it, given what I need from the program I use to write papers with.

The things you say in your second comment about reading papers on different kinds of devices makes a lot of sense, though.

In what way are footnotes more difficult than in Word? It isn't. Substantiate your claims.

I quote Tanksley: " It can be difficult to figure out where a footnote begins and ends when it is embedded in the paragraph, and it is difficult to change footnote 23 when the notes in the .tex file lack numbers."

In Word, I just hit the footnote button I installed on my toolbar. The footnote is then numbered and at the foot of the page.

When a student submits a paper to me, I convert it to LaTeX, which standardizes the font size and margins for me, so students can no longer play tricks. This is all done by a script, applied to every student paper at once.

That sounds pretty awesome. How do you do it?

If you're not now thinking in terms of sections and subsections... you're not doing your job as an author.

I mean, I am doing that. My point was, you can do that just as easily in Word.

LaTeX does not count words. That's a feature of your editors. Some editors have good ones. Some don't. You didn't even mention which one you were using.

I used TexShop, following Charles Tanksley's suggestion to download MacTex.

I didn't intend to blame LaTeX for the sins of the text editor. But if I gotta hunt around the web for a text editor for use with LaTeX that counts words the way I want, that's going on the "con" list.

And even if you do use styles, it's not true that you can use Word to quickly "match" the final output of some journal, however, because no professional journal typesets using Word.

That's true. You cannot typeset your article using Word. That's not what I was getting at, though. Usually when you send a final draft to a journal, they want you to format the bibliography in their particular way. This is a pain in the ass in Word. My question was, is it easier in LaTeX?

I was about to ask why semantic formatting is important, but you say why in a later comment. Thanks.

Am I passionate about LaTeX? Yes. This is in part because of the vast potential savings it could have for our libraries, and how it could shorten publication times.

I think these are worthwhile goals.

Honestly, I'm finding your reasons fairly persuasive. Much more so than the "You don't have to think about formatting" stuff I often hear.

zolover said...

Mr. Zero,

For better bibliographying, get (your near-eponym) Zotero.

It plugs into Firefox. It plugs into Word. In Firefox, it slurps up bibliographical information from a web page (Philosophers' Index, Amazon, Google books, very likely your university library's catalog pages...). It makes a library for you with an interface like iTunes' interface.

In Word, you can just drag a reference from the library into the paper. And you can use one of Zotero's many pre-made styles, or write your own style.

Zotero will also save and catalog the article itself, if you ask it to.

It's awesome.

I'm almost certain I first heard about it here at The Philosophy Smoker. But hmm, no, a search on 'zotero' yields no Smoker posts, so I guess I'm wrong about that. Anyway, thankyouthankyouthankyou to whomever introduced me to Zotero.

Word Verification: sples

Kevin Klement said...

Word isn't hard, except for some things. Frege's logical notation was a big one for me, as someone who writes about Frege. But also things like batch find-and-replace in multiple files and similar.

The turning point for me was the introduction of the .docx format. Free converters exist now, but a few years ago only Word 2007 read that format; I didn't have the budget for it at the time, and I resented having to spend the money. Planned obsolescence is the worst.

…a google search wasn't super helpful. Anon 12:23 solved that problem…

This is a problem generally with open source software. Since there's no one who is paid to answer your questions, it falls to you get those questions answered, which can sometimes be easy, sometimes hard. I'm surprised Google wasn't more helpful, but when in doubt, consider asking at LaTeX Community (or comp.tex.tex). A question like yours about paragraph breaks would be answered very quickly. Other questions might take longer. Charlie Tanksley and the rest of us at the PhilTeX blog are hoping of starting up a forum aimed specifically for philosophers and philosophy students using LaTeX.

If it's not there yet, the cumulative effect of information being gathered on blogs, forums, etc., however, will make it easier and easier to get help.

Personally, I set apart footnotes in my source document with comment markers.%
%
\footnote{Like this.} %
%
That sets them apart visually without putting in any extra spaces.

There shouldn't a problem finding the footnotes to edit. You just need to make sure your PDF program and TeX editor support SyncTeX foward and inverse search. I shift-click on a spot in my PDF and jump directly to the part of my LaTeX document that corresponds to it. This works fine with footnotes.

A good LaTeX editor can be configured with a footnote toolbar button that does what I suggested above. Some of them come with them. Or better, a quick-key combo. (Using the mouse is slower.)

That sounds pretty awesome. How do you do it?

I mentioned in a previous post that I have AbiWord installed. AbiWord is an free, OpenSource and cross-platform Word Processor. I rarely use it as a word processor, but I do use some of the command-line tools that come bundled with it. In particular, AbiWord will convert between any two formats it understands from the command line. It reads .doc, .docx, .rtf, .odt, and quite a lot more. (In fact, it even supports .pdf as an input format, though its PDF conversion is far from perfect.)

If you install the additional export plugins, it will convert to LaTeX. The conversion tool can be used inside a BASH script. I wrote one to convert every file in a folder, and do some post processing to get the same fonts and margins, and so on.

There's also things like pandoc or the unoconv tool (for OpenOffice). I don't have much experience with these. The program rtf2latex2e works decently well too.

Since it's a plain text format, you can process LaTeX files with any tools out there for processing text files, like sed, awk, perl regular expressions, and so on. (Though I don't imagine many philosophers have much experience with these.)

Anonymous said...

Zolover,

For what it's worth, you did hear about Zotero here, actually. I recall it coming up in the comments to an entry re: Endnote and similar programs, when someone recommended it. Can't find it right now, but I'm 98% sure it was here.

Verification: nakyd

Euthyphronics said...

Zero: Yes, reformatting bibliographies is (after an initial learning curve) much easier in LaTeX. If you use BibTeX, you jut need different bib style files for different journals/formats. Just replace one of these for the other, and presto: bibliography is reformatted.

One advantage about LaTeX that nobody else has mentioned yet is its cross-referencing feature. Maybe Word has one of these -- I never figured out how to do it, so if I wanted to have "see section 4.2" or "as mentioned in note 11", I just typed that in. If my section numbering changed or I added another footnote, I had to go back and find these and change the section or note number. Now, I just drop an anchor "\label{coolstuff}" at section 4.2 or note 11, type "see section \ref{coolstuff}", and LaTeX keeps track of the indices for me.

lyxer said...

For the average philosophy paper (I'm talking low on equations etc) Lyx is definitely the way forward—it combines a graphical interface reasonably similar to word but allows you to output pdfs in Latex formatting (as well as tex or plain text files).

What's more, it has a very comprehensive set of documentation built in so it is generally quick and easy to find answers to formatting issues. If that fails, you can search the mailing list online where you question has probably been asked before.

Hook it up to a bibtex bibliography (any bibliography software can generate a bibtex file as far as I know) and the power is awesome, especially for very long documents (thesis anyone). For instance there is a dropdown menu that keeps track of all the section and subsection headings that you have made and allows you to navigate straight to them with a single click.

Anyways, give it a try—it is the middle ground that most philosophers have been looking for!

Daniel said...

I had a similar experience with LaTex -- and ended up with Mellel, which I think combines the positive elements of LaTex ("1 simple formatting change") with a very nice WYSIWYG editor.

Anonymous said...

@Kevin Klement 3:48: calling it "code" is an overreaction; tell a programmer that writing HTML is writing in "code" and watch them laugh

TeX is code; unlike HTML, it's a Turing-complete language. You can create a universal Turing machine in it.

Anonymous said...

The whole movement scares me. I can only just use microsoft word, email and navigate a few philosophy webpages. Can a complete technophobe ever be converted to such a thing?

Anonymous said...

About the "student submit Word documents so you need a word processor" point: why not require them to submit their papers as plain text or html (i.e., if you think formatting is necessary)? Word can export into many formats, including those.

Additionally, you can allow your students to submit Word documents and then convert them into text yourself with a utility like antiword (or one of the docx-to-text converters).

FWIW: I keep a word processor handy for the work of colleagues and memos from administration. Since I have to have a word processor anyway, I require students to submit work in a non-proprietary format like RTF.

Kevin Klement said...

Anon@11.51. You really shouldn't worry so much. LyX is just as easy--probably easier--than Word for technophobes.

Anon@5:04. I'm afraid RTF is a proprietary format, and is owned by Microsoft. (Though as plain text based format, they obviously can't keep its secrets hidden!)

HTML is a better choice these days as a inter-conversion format, since it is non-proprietary, not to mention much richer than RTF.


Anon@5:19 Yes there are converters, but using them unfortunately does away with many of the advantages of LaTeX, since the resulting LaTeX document is often messy and shows non-semantic elements inserted because of the WYSIWYG creation process of the original. (In fact, this is a very good way to reveal what a mess Word Processor documents are underneath.) It's OK for my student papers, but would be far from ideal when it comes to things aimed at publication.

@Anon 4:59. You can make Turing machine with some markers and a few robotic parts. Doesn't make coloring with markers writing code. The issue isn't what you can do it, but what you typically would with it, which for the purposes of this thread, is write philosophy papers. If I remember right, Word has its own modified version of Visual Basic you can script things with in it. But how many Word users use that? The point is, writing a philosophy paper in LaTeX is typically not anything like the process of writing program code.

Corey said...

Word lost me when they changed their macro language from the simple to use word basic to visual basic. I understand why they did it, but I liked the old one and it was more appropriate for word processing.

I moved first to FrameMaker which is a great technical typesetting program I used while working in industry.

When I arrived at Waterloo to do my PhD I noticed that over half the grad students used LaTeX, likely because we a Phil of Sci, Logic, and Cog Sci heavy dept, and a lot of Grad students originate from these field's cognate disciplines .

Finally I settled on LaTeX. It is simpler, not simpler to use maybe, but simpler to fix things when there are problems. I like to know why the formatting looks like it does. When I use Word or even Open Office there are occasionally little glitches the cause of which always seem opaque to me (-- Visual Basic Developer once tried to explain the object model of a Word file to me, it is crazy complicated.)

That said I use A LOT of notation. I write in LaTeX math notation now and do not even need to see it converted to know what it looks like.

I can see why someone may use a word processor, if all they do is write papers in prose: no examples, no equations, no diagrams, no logical notation, not much in the way of a bibliography. But even then the portability, the light weight file sizes and the clean look of LaTeX would make me prefer it.

There is one final reason to use LaTeX that nobody seems willing to admit. It is FUN to use, especially when you discover how to do some new thing with it. It has been expanded so much, with so many packages that you can waste days making everything just so. I think it appeals to philosophers because philosophers like logical games, and programing is a logical game, even when the language output is a document instead of some other sort of data. (If you don't think LaTeX is a programming language, you are not using it right!)

Anonymous said...

Several people have commented on how much better LaTex looks. Is there anywhere on the web I can see specific examples of this?

Mr. Zero said...

Anon 7:03,

This has some pretty good examples.

I've been fiddling around with it all week, and I'm starting to come around.

Kevin Klement said...

Comparing the looks of LaTeX output to that of Word isn’t fair. LaTeX needs to be compared to professional typesetting software like InDesign and Pagemaker, and even then often comes out favorably.

A great place to look for nice examples of what (La)TeX can do is the TeX showcase. There’s some really neat stuff in there.

If I can use this as an excuse, you could also look at the ten different versions of Russell’s Introduction to Mathematical Philosophy I have on my website. Compare the PDF versions (made with LaTeX) to, e.g., the HTML version. No comparison really. Here are some screenshots I took of the various ebook versions for comparison. (The bottom one is a LaTeX-PDF.)

A lot of journals online are typeset with LaTeX, including some Philosophy ones like Philosopher’s Imprint. (If you have access to JSTOR, you can look at just about any math or logic journal, like the Journal of Symbol Logic.)

If you have access to Philosophy Compass, compare the “professional typesetting” of this article of mine, done presumably with Adobe InDesign or similar super-expensive professional typesetting software, with my “amateur” typesetting made for free with XeLaTeX. I made mine after theirs first came out since there were a bunch of things I disliked about how they handled it, and I wanted to show them how I thought it should look. I’m self-consciously mimicking their overall style, though, which I don’t even like that much. Afterwards they fixed most of the things I disliked, but there are still some subtle things I like better about mine, which I'll point out if you like.

Anonymous said...

So, maybe I am crazy or "see weirdly" or something, but I think Kevin Klement's HTML version of the Russell text looks a lot better than the LaTeX pdf. On the basis of his last post, I actually opened the first pdf on his site and the html version in two windows side by side, and looked them over for quite a while--and the upshot was that I really preferred looking at the html.

This fits with something I have been thinking for a while--LaTeX products often look odd to me. I can't always put my finger on it, but here is one small concrete example about what looks odd in Klement's Russell text:

In various words, the tops of the letters don't line up in the LaTeX file, which makes things look "wavy." For an example, find the word "infinity" in the preface and look at the "ini." Then find the same word in the html version and see how the letters line up across the top.

I am guessing that this is not an issue for LaTeX per se, but just the way Klement set things up. Or perhaps the way the LaTeX file looks is actually proper typesetting, and I am some sort of typesetting philistine. In any event, I find the look of those LaTeX files off-putting.

Kevin Klement said...

I'd guess either this is a problem with the anti-aliasing of your PDF reading software relative to your screen resolution, or you just prefer one font over the other. The HTML version uses Georgia by default which was designed as a screen font, not a print font, so may look better on screen. Try printing a page of each. Or switch the HTML version's font to Palatino, which is more similar and see if your preferences change.

Anonymous said...

I don't know...it sounds to me like Anon 8:24 either sees weirdly or is a typesetting philistine.

Anonymous said...

I am using LaTeX for the first time to write a report and i am LOVING it. To start with I was really against the idea because of all the arguments above, but there's something about it which is just more relaxing than word - it's so very...logical. even when there are things you have to fix it's like 'well I can see why the program tried to do it that way'. Not like word which just seems to randomly change stuff in the formatting.

Also, we only really find word easy because we've been using it for 10 years. If we'd been using LaTeX for 10 years it would probably seem even easier and all our documents would look amazing.

Having said that I'll still be using Powerpoint for presentations. I think doing presentations in latex is going a bit far - hard to make them look good.

Kevin Klement said...

Hard to make presentations made with LaTeX look good? Have you tried Beamer?

I think Beamer presentations look great.

I also can't stand PowerPoint, and the OpenOffice presentation software is just about unuseable.

David said...

Markdown is what you want. Dead simple and easy to learn. Only the semantic markup you *need*. Easy conversion to HTML, RTF, LaTeX, PDF, even OpenOffice's ODT.

Specifically, for academic papers, you want multimarkdown or pandoc, since they extend markdown to include support for footnotes.^[like this one] Multimarkdown is easier to install, but its support for conversion to RTF is a bit flaky (no real footnotes, for example).

Pandoc is harder to install, but it will convert directly to ODT, which can then be saved as RTF or DOC for journals that will only accept those formats. And it has experimental support for citations using the same style files that Zotero uses for export. And it can convert various formats into markdown, which is a nice trick...