Ebook Formating Tool Wishlist

[delphidb96]

Originally Posted by HarryT

You can't sensibly quote page numbers at all on machines like the Gen3 which allow the user to load any font they wish onto the system. The pagination changes with the font.

Pagination changes with the font family *and* the font size. IOW, page numbers make no sense whatsoever for an electronic book - unless one insists upon using a one-size-fits-none approach similar to PDF. It would behoove authors and publishers to think in terms of citations and footnotes being gathered by chapter, part or section, not by page. After all, there's no real reason to not build in a hyperlink to the relevant note right into the text.

Ah well, perhaps the publishers will figure out - some time within the next 10,000 years - that putting an axle between two wheels will allow for a platform to ride on the wheels...

Derek

[aapezzuto]

Originally Posted by Olympus

I apologize. I was tempted to start a new thread with this topic, however this project has come up because someone wants to program for 200-500 hours and has come up to spend the precious time to re-factor BookDesigner.
I just wanted to make a point that other e-book related "urgent" matters could be addressed: Ebook Management.

This project is at the top of my list for lots of reasons. If it makes you feel any better, I feel that the obvious next steps once this project is done are to get an ebook organizer, a light weight ebook converter(using the same plugins as this project, but with no markup interface), and an auto rss feed converter all working well together... and if I can get a few major online publications to pre-markup the documents, well I win!

also, the page reference thread is elsewhere... Please let is stay there.

but rely, what do you want as features for a program you would use to format or write ebooks with?

[kovidgoyal]

Umm all the conversion, rss fetching, ebook management is already done in libprs500, but if you want to re-invent the wheel, who am I to stop you.

[llasram]

Originally Posted by kovidgoyal

Umm all the conversion, rss fetching, ebook management is already done in libprs500, but if you want to re-invent the wheel, who am I to stop you.

Maybe it's time for libprs500 to change to a less PRS-500-centric, more "consumer-friendly" name. :-)

Originally Posted by aapezzuto

but rely, what do you want as features for a program you would use to format or write ebooks with?

Something like LyX's WYSIWYM (What You See Is What You Mean) interface. Make it easy to edit books in terms of semantic entities and difficult/impossible to edit the formatting directly.

[aapezzuto]

Originally Posted by kovidgoyal

Umm all the conversion, rss fetching, ebook management is already done in libprs500, but if you want to re-invent the wheel, who am I to stop you.

There is a reason why those parts are lower on my list...

Originally Posted by llasram

Something like LyX's WYSIWYM (What You See Is What You Mean) interface. Make it easy to edit books in terms of semantic entities and difficult/impossible to edit the formatting directly.

This is a great suggestion, and Tex and Fb2 are two of the formats I get a lot of excitement about.
How would you denote in the editor that a section had markup applied. Some things are easy, such as bold italic etc... but would you come up with a way to tell that those words are a chapter heading and not the title of a poem??? what about color coding?

[JSWolf]

How well does your friend know LRF, IMP, Mobi, FB2, and LIT formats?

[aapezzuto]

he has no problem learning something from a specification. Last summer he learned how to program in 2 non standard languages, and wrote webservices to help control automated product flow... It will just be a matter of finding the documentation.

[DaleDe]

Originally Posted by aapezzuto

he has no problem learning something from a specification. Last summer he learned how to program in 2 non standard languages, and wrote webservices to help control automated product flow... It will just be a matter of finding the documentation.

the wiki is a good starting point for documentation. If it is missing anything critical then fix it!

Dale

[kovidgoyal]

Originally Posted by llasram

Maybe it's time for libprs500 to change to a less PRS-500-centric, more "consumer-friendly" name. :-)

It's on my TODO list

[Olympus]

Originally Posted by aapezzuto

but rely, what do you want as features for a program you would use to format or write ebooks with?

Personally as fiction reader I do not require a full EBook Writer. I do need an import from various formats, a toolbox to fix-up ebooks (a combination of E-Tidy to clean up the source and a simple paragraph tagger as provided by BookDesigner (Title, author, header and paragraph) and to finalize the good work an export.

As former DTP-er I have experience reformatting input documents into proper books and I experienced that most novels only need some basic text-tags and cleaning:
BookTitle, BookSubtitle, Author, Header, Footer, New page, Chapter Title, Chapter Subtitle, Chapter sub-sub title, paragraph, indented paragraph

Cleaning:
Transform all beginning spaces into tabs,
remove all tabs when preceded by a CR
Quote marks -cleanup

thus most fiction books are fairly simple.

[aapezzuto]

ok then, as a slightly more technical question, would regular expressions + a find replace type statement be about the extent of scripting needs inside a program such as this?

I have been going around in circles about the idea of exposing an object model of the document, and the problem is... as markup changes, the object model can be DRAMATICALLY changed, rapidly... leading to large levels of computing power being needed to update the model after a change. Some of this can be avoided by proper modeling / engineering... some of it can't. Because of this, we have been tempted to only build a stream in/out type interface for outside scripting during editing, leaving the robust stuff for input and output builds.

I find it reassuring to see such a small list of markup types as the core needs of the project. I know that in my editing all I user are those core 8-10 things.

also, since the epub format all ready supports a container to keep everything packaged together, would people like to see it keep your export preferences inside a specially marked e-pub file? Or the more pointed question... do you want to distribute a formated file with your convention preferences intact?

[Gudy]

Originally Posted by aapezzuto

ok then, as a slightly more technical question, would regular expressions + a find replace type statement be about the extent of scripting needs inside a program such as this?

Scripting + regexp + find/replace? This sounds like the BookCleaner functionality in BD. Is this what you had in mind? Anyway, yes, this is probably the most useful scripting application for such a tool. While I am sure that many here can think of dozens of other things they would like to be able to script (ISBN lookup or other net-oriented things, e.g.), I wouldn't consider those core functions of the tool, while something like BookCleaner probably is.

Originally Posted by aapezzuto

also, since the epub format all ready supports a container to keep everything packaged together, would people like to see it keep your export preferences inside a specially marked e-pub file? Or the more pointed question... do you want to distribute a formated file with your convention preferences intact?

I personally don't, no.

[Olympus]

Originally Posted by aapezzuto

ok then, as a slightly more technical question, would regular expressions + a find replace type statement be about the extent of scripting needs inside a program such as this?

Mostly the cleaning up is difficult as most "authors"-documents have specific markup errors that seem to be specific for each source document.
extended - find & replace and/or a simple RegFindReplace is needed to cleanup. After that only a very-small-font preview is required to quickly identify Chapter Titles and Tag these and a zoom-in (larger preview font) to set paragraph-specific-tags (indent, quote etc)

Originally Posted by aapezzuto

... Because of this, we have been tempted to only build a stream in/out type interface for outside scripting during editing, leaving the robust stuff for input and output builds.

That could be the main engine and provide a "plug-in" to perform the cleaning/tagging of the document. (But then question is: what is the difference between your tool and BD and LibPrs500 and EBookPublisher etc etc??)

Originally Posted by aapezzuto

I find it reassuring to see such a small list of markup types as the core needs of the project. I know that in my editing all I user are those core 8-10 things.

I did NOT provide a final list of required markup tags. I always keep in perspective that an E-Book can never display the result of a full DTP-ed product. Should formatting be such an important issue than use a PDF-file and viewer or even the paper-version (a.k.a. p-book)

Originally Posted by aapezzuto

also, since the epub format all ready supports a container to keep everything packaged together, would people like to see it keep your export preferences inside a specially marked e-pub file? Or the more pointed question... do you want to distribute a formated file with your convention preferences intact?

It would be nice to have a export-settings file for each document that is exported, so one is able to re-create the export or tweak the export. I have no knowledge of the EPUB format.

[HarryT]

Originally Posted by Gudy

Scripting + regexp + find/replace? This sounds like the BookCleaner functionality in BD. Is this what you had in mind? Anyway, yes, this is probably the most useful scripting application for such a tool. While I am sure that many here can think of dozens of other things they would like to be able to script (ISBN lookup or other net-oriented things, e.g.), I wouldn't consider those core functions of the tool, while something like BookCleaner probably is.

Find/Replace with Regular Expressions is one of THE most useful features of BD - it's something I use all the time.

One simple example: many books you get from PG have page number of the form [Pg xxx] in them. These can trivially be eliminated in BD by searching for "\[Pg [0-9]*\]", setting the "Replace" string to an empty string, and checking the "RegEx" box.

Similarly many PG files use either underscores surrounding a word, or have a word in capital letters to indicate italics. These too can easily be located using a RegEx search.

It really is an essential feature of the program.

[aapezzuto]

Originally Posted by Olympus

That could be the main engine and provide a "plug-in" to perform the cleaning/tagging of the document. (But then question is: what is the difference between your tool and BD and LibPrs500 and EBookPublisher etc etc??)

I love book designer, but it gas always left me wanting a bit more. I have much the same feelings about fb2. All the tools I have used have had good points, but none of them have had quite the level of friendliness and refining that I was looking for. The other thing that I want from this is a central defacto standard for the base document. Right now, with the readers and formats available, I think lots of people would rather have a highly tuned, device optimized pdf...

The biggest things that I hope this project offers:

• an easy interface
• robust feature set (serve 99%+ of the needs of the target market)
• near infinite expandability as far as output format options.
• a sustainable, open code base

I have an ungodly number of ebooks, and about 1/10 of them are in fb2 format... the rest are basically (under/un)formatted. I want a more ideal tool for my use... however my use is not exclusive, and if I can incorporate other peoples ideals, than all the better.