docbook vs. latex vs. html

danmc · 01-13-2011, 03:17 PM

I know this question ultimately has a lot of personal preference in the answer, so I'll try to inject some of my preferences into the question.

I am interested in a good work flow for creating books that target both devices like kindle, kindle for android, as well as PDF and hardcopy output.

Sometimes I may be starting with OCR text that needs a lot of manual fix up, sometimes I may be starting totally from scratch. For my immediate project at hand, there will be quite a large number of hyperlinks that I really want to work right.

In general I am a bit LaTeX fan for typesetting documents. However, I tend to be less than thrilled with the output of latex2html for a web page and it seems that the output may be substandard when viewed on a kindle.

So maybe I should be looking at some other source. I am *not* a fan of microsoft word (don't even have it) or openoffice.

One thought is maybe I should be looking at docbook or even just a home brew xml based thing. I hate to re-invent the wheel though on a home brew approach.

It seems that docbook may be better suited to producing html from the xml sources and also it clearly has been used to produce hard copy books.

My other option is to forget the hardcopy desire and just mark up my text in straight up html and view kindle and a web browser as my primary targets and relegate hardcopy to 2nd class.

I prefer staying with NetBSD, linux, solaris, etc as a computing platform and I generally prefer text based and command line based tools. I do have quite a bit of experience with various programming languages and some with xml and could learn more if that is the right path.

Comments, thoughts?

Most of what I've found so far seems to be either a microsoft word -> html -> manual edits flow or <something> to PDF flow.

Thanks
-Dan

Sweetpea · 01-14-2011, 01:55 AM

What is your problem with the latex2html? I've never tried it, so I've no idea how it would look...

Jellby · 01-14-2011, 04:22 AM

I would probably create separate LaTeX and HTML versions, they have enough differences to make automated conversions problematic. You could make a first version in either of the formats and convert it to the other, and then format each of them separately.

Or, you could create the PDF directly from the HTML using Prince, for instance.

frabjous · 01-14-2011, 04:15 PM

I would just stuck with whatever you're more comfortable with and go from there. Converting between the various XML based formats is probably fairly trivial. Going between LaTeX and XML is a little trickier, but there are lots of tools to look at besides latex2html. There's also TeX4ht, TtH, Hevea, tex2page, etc. At least some of those are very customizable so if you don't like how something is handled by default, you can change it.

For my projects, I ended up writing my own sed scripts for converting the LaTeX source to (X)HTML. They're very project specific, though, and take awhile to perfect, but they still beat having to work with two independent sources in case I need to make small changes.

I think in addition to Prince, however, there are some ways now of using ConTeXt macros with either the TeX or LuaTeX engines to typeset various species of XML directly. I've been meaning to learn about that, but can never find the time.

rplantz · 09-15-2011, 02:15 PM

This thread is a little old, but it addresses exactly my problem. I have written a 500+ page highly technical textbook with many code listings, circuit diagrams, etc., in LaTeX. I would like to make it more accessible by creating an ePub version. Current converters crash around chapter 2.

I currently have two sets of page layouts that produce two pdf versions, one for double-sided 8-1/2 x 11 printing, the other for ereading (uniform, very small margins). I have managed to keep pagination synced between the two versions. But I believe that ePub is a better ereader solution.

So I am trying to determine the best toolchain for producing both an ePub and a print version of the book. I know that O-Reilly does this with many of their books. Their toolchain is discussed by Adam Witwer at http://answers.oreilly.com/topic/262...077&#entry5077.

I'm beginning to think that I should convert the primary source of my book to docbook. I think that may provide the best chances for producing both an ePub and a print version that are very similar in appearance.

BTW, I wrote the original versions of the book in html in 1999, thinking that students would like to read it online. They used up the printing resources in our labs, printing copies of the web pages. Now most of them prefer ereading on their mobile devices. I shoulda left it in html!

01-13-2011, 03:17 PM	#1
danmc Junior Member Posts: 1 Karma: 10 Join Date: Jan 2011 Device: kindle	docbook vs. latex vs. html I know this question ultimately has a lot of personal preference in the answer, so I'll try to inject some of my preferences into the question. I am interested in a good work flow for creating books that target both devices like kindle, kindle for android, as well as PDF and hardcopy output. Sometimes I may be starting with OCR text that needs a lot of manual fix up, sometimes I may be starting totally from scratch. For my immediate project at hand, there will be quite a large number of hyperlinks that I really want to work right. In general I am a bit LaTeX fan for typesetting documents. However, I tend to be less than thrilled with the output of latex2html for a web page and it seems that the output may be substandard when viewed on a kindle. So maybe I should be looking at some other source. I am not a fan of microsoft word (don't even have it) or openoffice. One thought is maybe I should be looking at docbook or even just a home brew xml based thing. I hate to re-invent the wheel though on a home brew approach. It seems that docbook may be better suited to producing html from the xml sources and also it clearly has been used to produce hard copy books. My other option is to forget the hardcopy desire and just mark up my text in straight up html and view kindle and a web browser as my primary targets and relegate hardcopy to 2nd class. I prefer staying with NetBSD, linux, solaris, etc as a computing platform and I generally prefer text based and command line based tools. I do have quite a bit of experience with various programming languages and some with xml and could learn more if that is the right path. Comments, thoughts? Most of what I've found so far seems to be either a microsoft word -> html -> manual edits flow or <something> to PDF flow. Thanks -Dan

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Converting DocBook XML source book with calibre!	baccilus	Calibre	1	12-04-2010 12:32 AM
HTML to Mobi conversions (DocBook XSL, and content.opf?)	AndrewLB	Calibre	3	09-04-2010 09:02 PM
Docbook to ePub	IncarnadinePress	ePub	2	04-07-2010 11:25 AM
Chaîne complète d'édition : HTML, LaTeX, PDF, EPUB à venir.	Randy11	Software	2	04-05-2009 11:42 AM
DocBook XSL 1.74.0 adds ePub support!	Alexander Turcic	News	1	06-14-2008 07:06 AM

01-14-2011, 01:55 AM	#2
Sweetpea Grand Sorcerer Posts: 9,707 Karma: 32763414 Join Date: Dec 2008 Location: Krewerd Device: Pocketbook Inkpad 4 Color; Samsung Galaxy Tab S6	What is your problem with the latex2html? I've never tried it, so I've no idea how it would look...

01-14-2011, 04:22 AM	#3
Jellby frumious Bandersnatch Posts: 7,515 Karma: 18512745 Join Date: Jan 2008 Location: Spaniard in Sweden Device: Cybook Orizon, Kobo Aura	I would probably create separate LaTeX and HTML versions, they have enough differences to make automated conversions problematic. You could make a first version in either of the formats and convert it to the other, and then format each of them separately. Or, you could create the PDF directly from the HTML using Prince, for instance.

01-14-2011, 04:15 PM	#4
frabjous Wizard Posts: 1,213 Karma: 12890 Join Date: Feb 2009 Location: Amherst, Massachusetts, USA Device: Sony PRS-505	I would just stuck with whatever you're more comfortable with and go from there. Converting between the various XML based formats is probably fairly trivial. Going between LaTeX and XML is a little trickier, but there are lots of tools to look at besides latex2html. There's also TeX4ht, TtH, Hevea, tex2page, etc. At least some of those are very customizable so if you don't like how something is handled by default, you can change it. For my projects, I ended up writing my own sed scripts for converting the LaTeX source to (X)HTML. They're very project specific, though, and take awhile to perfect, but they still beat having to work with two independent sources in case I need to make small changes. I think in addition to Prince, however, there are some ways now of using ConTeXt macros with either the TeX or LuaTeX engines to typeset various species of XML directly. I've been meaning to learn about that, but can never find the time.

09-15-2011, 02:15 PM	#5
rplantz Member Posts: 21 Karma: 493474 Join Date: Mar 2011 Location: Northern California Device: Acer A500	This thread is a little old, but it addresses exactly my problem. I have written a 500+ page highly technical textbook with many code listings, circuit diagrams, etc., in LaTeX. I would like to make it more accessible by creating an ePub version. Current converters crash around chapter 2. I currently have two sets of page layouts that produce two pdf versions, one for double-sided 8-1/2 x 11 printing, the other for ereading (uniform, very small margins). I have managed to keep pagination synced between the two versions. But I believe that ePub is a better ereader solution. So I am trying to determine the best toolchain for producing both an ePub and a print version of the book. I know that O-Reilly does this with many of their books. Their toolchain is discussed by Adam Witwer at http://answers.oreilly.com/topic/262...077&#entry5077. I'm beginning to think that I should convert the primary source of my book to docbook. I think that may provide the best chances for producing both an ePub and a print version that are very similar in appearance. BTW, I wrote the original versions of the book in html in 1999, thinking that students would like to read it online. They used up the printing resources in our labs, printing copies of the web pages. Now most of them prefer ereading on their mobile devices. I shoulda left it in html!

Advert

Advert