[frabjous]

Originally Posted by nick_

The art of typography is still valid for ... typography!

It is irrelevant for the new media.

An e-Ink screen looks almost exactly like a piece of paper. And even a computer screen doesn't look so different from a piece of paper that everything we learned will somehow become irrelevant. Why would you think that?

There are always those who try to impose the old standards to a new media. This is a mistake that has been repeated many times in history. Are you sad for the disappearance of calligraphy, an art of 3000 years, because of the "vulgar" typography?

I'm actually quite fond of calligraphy, but I agree that the art of text needs to update with changes to the media... but what about the switch from paper to device screens in any way is relevant to whether or not the advantages of the things I cited remain advantageous? I can't think of any reason to think there's a relevant difference with regard to these things whether you're looking at paper or an electronic medium. It's sort of like suggesting that the rules of typography are different depending on what species of tree your paper comes from.

Why are they different?

Originally Posted by nick_

I will give you three:

1) Separation of content from presentation (XHTML+CSS).

LaTeX's mark-up is also a semantic mark-up language which separates form and content. It is exactly like HTML in that respect. If you've never used LaTeX (which obviously you haven't), why do you think you know what it is? Changing the fonts used, or the page size globally is done in almost precisely the same way it's done with HTML--it's just a matter of setting some global definitions or adding packages at the beginning, leaving the content body alone, just like you'd change things in HTML by changing the CSS and leaving the tag structure alone.

In fact, separation of presentation from content is the usual selling point of LaTeX. The idea was that an academic would only need to change one or two lines at the beginning of the document to add the package that loads the style for a given journal in order to format the article for that journal's paper size and fonts, citation and bibliography style, etc. In fact, this is how most math journal articles are published.

This separation is set to become even stronger in LaTeX3.

Do you really think it matters whether we flag emphasized text like this:

This <em>word</em> is emphasized. (HTML)

Or like this:

This \emph{word} is emphasized. (LaTeX)

?

Do you really think it matters whether we signify paragraph breaks with <p>...</p> or LaTeX's way with two linefeeds? (Except perhaps that the latter is slightly easier to read?) Or whether we write

<h2>Chapter Name</h2> (HTML)

or

\chapter{Chapter Name} (LaTeX)

to signify where a chapter starts?

Again, the differences in the mark-up languages are unimportant.

2) Human eyes have a wide field of view, that's why we have wide LCD screens (and movie theaters) and no A4 screens.

Firstly, LaTeX can be used to output in a wide format just as easily as HTML, so this is not argument in favor of one over the other.

Secondly, as for wide fields of view, there is tremendous amount of research to suggest that reading overly long lines of text causes eye strain, which is why magazines and newspapers organize material into columns. This is not going to change on a computer or other device screen, so having a renderer that can handle such things is (And currently, LaTeX's renderer does columns well... I don't know offhand how to do columns in HTML without manually fixing the column breaks, or whether it's even possible.) The renderer for LaTeX can even handle "visually" aligned margins where, e.g., punctuation can extend slightly into the margin to give a more uniform look, to eliminate reading distractions.

3) Flexibility.

LaTeX is not less flexible than HTML.

Take a look here (this is my webpage). I have six different sized PDFs made with LaTeX. These are all made with precisely the same LaTeX souce code. Changing font or font-size, etc., is not any harder with LaTeX than it is with HTML. You're confusing the LaTeX source with the PDFs it generates. It's like confusing HTML with a screenshot of someone's web-browser with particular settings.

LaTeX is most often used (these days) to output PDF, but don't confuse PDF with LaTeX. LaTeX can be used to output DVI or PS or even HTML instead. Explain how it's possible that LaTeX can be less flexible than HTML when it can be used to generate HTML?

There are people out there with 30 inch screens, with 3 inch screens, people with a variety of eye handicaps etc.

And this is relevant how? I can use the same LaTeX source to create an A4 sized document and an iPhone-sized document . In fact, I have. (See link above.)

Originally Posted by nick_

No. The discussion is (XHTML+CSS) versus PDF or LATEX. These are the formats.

I said nothing about the rendering engines.

Then you've changed the topic. I've stated multiple times that I don't think the differences between LaTeX's and HTML's mark-up are very interesting. In fact, I've explicitly called them trivial. My only stated preference for LaTeX is that there currently exists typographically rich renderers for LaTeX source, and there isn't for HTML source, though conversion in either direction is theoretically not difficult.

I don't even disagree with the statement that HTML/CSS is the future of mark-up, but that has nothing to do with the future of rendering or the obsolence of typography.

It's like this: we already have excellent way to automate rich typography. LaTeX loaded onto our actual readers, whether it's reading ePub source or LaTeX source or whatever, or any other, could be used to reflow on the fly, but do it well. Since that technology already exists, why do we put up with the much worse looking ebook reading experience we currently get instead?