View Single Post
Old 02-03-2013, 10:10 PM   #44
Freeshadow
temp. out of service
Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.Freeshadow ought to be getting tired of karma fortunes by now.
 
Posts: 2,787
Karma: 24285242
Join Date: May 2010
Location: Duisburg (DE)
Device: PB 623
Quote:
Originally Posted by fjtorres View Post
eBook typography rules and controls belong in the reading app, not in the file format, simply because the file knows nothing of the rendering device (Screen size, resolution, aspect, etc).

A good reading app *should* have embedded algorithmic rules for white space, line spacing, widows/orphans, and even formatting style (Via templates). But that stuff is *hard* to code. So, at best, we get hyphenation and user-set margins and spacing and at worse none of the above or one-size fits all typography.

Going back to the days of the now-deprecated but still mourned MS Reader app we can find better embedded typographic rules than in most modern reading apps.

Now, that is just the theoretical typographical purist critique.
The real world reality is that mainstream consumers don't care about typographical issues as long as they can set reasonable font sizes and the available font families aren't aggresively ugly.

For the ebook market of today, the formats are mre than adequate and the reading apps acceptable. There might be an opportunity for one of the payers to gain a minor competitive advantage by offering quality dynamic typography in their reading app but impact is likely so small there are higher-priorities at Amazon and Apple.

What we have isn't perfect but it's good enough for the market.
The typographical shortcomings are evenly distributed between lacks in the reading software AND ebook format specs. Examples for both following.

Quote:
Originally Posted by taustin View Post
To me, "enriched content" means "it's not a book any more." It's a web page. A multimedia web page. I have zero interest in a book with animated GIFs and blinking text, and music that plays automatically. Especially there is no chance that those will not be used predominately for advertising.

Turning ebooks in to multimedia web pages will only amount to abandoning the book market. At which point independent publishers (mostly self-published authors, I suspect) will rush to fill. Poorly, for the most part, but hey, Sturgeon was an optomist.
Exactly. I think it was and still is a major mistake in epub spec definition to look at "how web does it." Instead of looking at TeX: both are:
  • WYSIWYM
  • Separating styles and content
  • Intended for digital 'print products'

Apart from that I fully share your opinion about animation etc. in books:
It ruins legibility and concentration on the textflow. The only way to animate a book and still making it to be a book is to throw it.

Quote:
Originally Posted by JSWolf View Post
You cannot blame the format for typographic techniques being ignored. You have to blame the software used for reading.
Sorry. With "format deciders" I meant "people responsible for a formats specs". Now examples for both format and software
Poor handling of whitespace in images and therein resulting poor (or none) text wrapping around them is a software weakness.

The epub specs state OpenType as supported font encoding method. Nevertheless there is no way to adress and use alternative glyphsets of a font such as monospace numbers, Dropcap capitals etc. TeX has a method to adress alt. characters.

Quote:
Originally Posted by tompe View Post
Well, the format need to encode enough information for the typesetting algorithms and the current format do not do that from what I know.
Quote:
Originally Posted by JSWolf View Post
The reading software needs to handle things like kerning and hyphenation. Kerning is one thing reading software does ignore.
One of the major differences between True- and OpenType is, the latter having (more) kerning information embedded. The data is there - ignored.

Quote:
Originally Posted by Elfwreck View Post
There are people who claim that correct spelling and use of commas are just too much of a bother. The fact that a lot of people want to ignore production standards doesn't mean it's reasonable to accept their laziness as the default.

Currently, the problem is that ereaders don't consistently support soft hyphens. Whether or not it would be "easy" to add the Tex hyphenation routine to readers, authors and publishers should be able to establish specific hyphenations for unique words--character and place names, scientific or magical terms, neologisms that haven't been added to the protocols yet, and so on.
For epub it is mandatory to have the contents' language specified.
The most logical solution would be the following ruleset:

1. use hyphenation dictionary equal to books lang metadata
2. Differing fragments wrapped in [ lang=langcode ] and a closing tag invoke use of different dictionary.
3. Soft hyphens neologisms, names and similar only.
Freeshadow is offline   Reply With Quote