Ebook Formating Tool Wishlist

[moz]

Originally Posted by kovidgoyal

umm if you're reading an ebook, the computer will count paragraphs for you.

Firefox doesn't do that for me, neither does IE. Is there a way to get the PRS505 to display that, or are you talking about using a PC and some kind of program to do it? Does anything in LibPRS allow me to search for text and get the reference you're talking about?

Even if so, as a general solution it's not very elegant. At least the title+text approach works for existing documents and software without requiring the user to find a program that does tricks. For many references you can use the section title to narrow it down a great deal, and ideally they'll use hierarchical numbering to make even more compact, but for the general case we still something.

[kovidgoyal]

Originally Posted by moz

Firefox doesn't do that for me, neither does IE. Is there a way to get the PRS505 to display that, or are you talking about using a PC and some kind of program to do it? Does anything in LibPRS allow me to search for text and get the reference you're talking about?

Even if so, as a general solution it's not very elegant. At least the title+text approach works for existing documents and software without requiring the user to find a program that does tricks. For many references you can use the section title to narrow it down a great deal, and ideally they'll use hierarchical numbering to make even more compact, but for the general case we still something.

From a programming perspective adding the ability to do that is trivial. If you're going to advocate title+text, you may as well drop the title and leave just the text (since with search, you don't need the title at all), but then your reference becomes more like an extract.

[Jeff Duntemann]

Originally Posted by kovidgoyal

I thought we'd decided eons ago that the correct way to do referencing is by section and paragraph numbers?

Sigh. I certainly didn't decide that, nor do I see any significant number of publishers that have. Not all books can be divided into paragraphs. In fact, most of the print books I've published in the last fifteen years or so have typographic or design elements that don't map well onto paragraphs at all.

I'm interested here in how to allow ebooks and pbooks to coexist for the next 20-30 years or so, without breaking current research models. I don't see why a paginated master view can't satisfy all parties, assuming a compiler that can spit out all the major formats. The consumer can ignore the page markers if he/she chooses, but there's no reason for them not to be there, and available to the reader software.

[kovidgoyal]

Originally Posted by Jeff Duntemann

Sigh. I certainly didn't decide that, nor do I see any significant number of publishers that have. Not all books can be divided into paragraphs. In fact, most of the print books I've published in the last fifteen years or so have typographic or design elements that don't map well onto paragraphs at all.

I'm interested here in how to allow ebooks and pbooks to coexist for the next 20-30 years or so, without breaking current research models. I don't see why a paginated master view can't satisfy all parties, assuming a compiler that can spit out all the major formats. The consumer can ignore the page markers if he/she chooses, but there's no reason for them not to be there, and available to the reader software.

Because pages are a completely unnecessary artifact of the limitations of paper. There is absolutely no reason to stick with pages now that we are moving on from paper. You can have ebooks co-exists with pbooks by simply using the ebook version rather than the pbook version in new references and using the pbook version for old references.

[aapezzuto]

Good God!
I create a new thread and leave it to percolate for 24 hours, while I take care of sick babies and a sleep deprived med-student wife, and i come back to a 3 page or 34 section argument about how to site references... a topic that will never come to a consensus, as the current system is truly just the evolution of a most popular approach based on what was available.

So I will put my 0.02$us in on this, and then I would take it as a kindness if people would let this topic drop... I will take it under full advisement that people would like to find things in their books...

The point of reference is to allow other people to get close to the information you are basing something on. The big issues are

• granularity (can I give them directions that get them close enough that they can find it in an arbitrarily decided "reasonable" amount of time)
• consistency (can everyone, despite different resources use these directions)
• longevity (will they still be able to find this in xyz large units of time in the future)

No system will be full proof. A reasonable compromise is to allow paragraphs to belong to a page number (whatever page they start on in a given print version) and allow a cross reference between the two. As ebooks become the standard, section and subsection are a nice way to reference down to a reasonable level of granularity (no one advocates chapter\paragraph\character range count, without being laughed at), but again a version number of some sort is of the utmost importance.

It is 99% likely that there will be a simple toggle to insert paragraph numbers into the xml of the book, 99% (dependent) chance that you will also have the option of having those numbers shown based on your output. Other than time, there is no reason why page number references could not also be added, with a slight loss to resolution (I am not a fan of adding page numbers in the middle of a small subsection whenever avoidable)

So with this set of solutions, if there is a 10 "page" smallest level of section... eather accept that it is meant to be only consumed as a single idea and a reference should reference the whole thing, or format it into smaller sections from the start. poems into stanzas, big-ass paragraphs into... only god knows....

I will make separate posts responding to the other 5-6 posts that have not gone down this path.

Thank you for the huge response all ready, keep it up... just not about this particular topic anymore

[moz]

Come on, we can't even agree on spelling, what chance we can agree on sitation standards? You're all fulls!

I am going to make my tagline here "homonyms are not synonyms".

[aapezzuto]

ok, now on to everyone/everything else:

Originally Posted by kovidgoyal

Consider doing it in python rather than C++, much faster lead times and much easier long term maintenance.

I will not argue the merits of different programming languages, I will push to have python be considered as the plugin language of choice, or at least an interface for it to be made. I share your affection for python, as me and c++ have never gotten along for anything other than image manipulation. However when a person says "here, let me do a full time job for you for 12 weeks, and you dont have to pay me... what do you want?" you don't argue when they say they want to do it in something that they can compile for different platforms rather than something that can be interpreted on different platforms. And yes I know that everyone who has written code since the dawn of interpreted languages has a strong view on this one way or another... But to this dialog its just not significant.

Originally Posted by Gudy

the internal format should probably be closer to something like FB2, except perhaps richer.

The appeal of the epub standard is that it will allow for something like fb2, but with significantly more flavor and texture. We have been talking about defining an epub lite (choosing how much flavor we want to support) and also the idea of allowing people to define interface elements that will create new flavors as needed. The slippery slope here is that for every new and strange case you add, though epub compliant, you make it MUCH more difficult for the average programmer to write a good converter to output to the format of your choice... Thus, the less standard the file, the less useful it will be to others.

Originally Posted by HarryT

When you say "replace Book Designer", do you mean a GUI tool which will, like BD, allow numerous formats (text, HTML, RTF, PalmDoc, etc) to be input, and the creation of numerous output formats (Sony LRF, MobiPocket, IMP, etc)? If it's to replace BD, it's obviously vital that it be able to read BD's ".html0" storage format so that books can be transferred to it from BD.

YES! This is exactly what we are after. BD is the default because its all that there is for us non-programmers (yes i consider myself object minded, but sense i would delegate ANY coding task to others.. i am no programmer). It will natively (out of the box) only support epub files, however... as a fundamental feature, we will be developing a standard interface for load/save add-ons. This feature is so significant that it will supersede many of the interface elements as far as priority. At the end of core development, all extra energy will go to developing as many different high quality import/export plugins as possible, utilizing this interface.

Originally Posted by HarryT

Forgive my ignorance, but the main thing that distinguishes BD from the other tools is its GUI book editor. Can one write GUI apps in Python?

my general rule of thumb is not to ask this exact question... I have found that people will argue the most extreme cases when asked if something is possible, perhaps a question of ideal or a "good idea" is a better one.
Python has great capabilities, and is my language of choice for all sorts of tasks. If i don't want to do it by hand, I want a program to do it.. if i don't want to run this program on a hand held device I do it in python or java. both of those languages have the possibility to write the entire front end gui of an entire OS, control robots, and even do FEA, However I will leave it to people willing to do the work to decide what tools they want to use. If I cant trust them in that respect, I wont work with them.

Originally Posted by Jeff Duntemann

What I really need is probably a little more than a 500-hour project, fersure (especially if executed in C++) but here's the gist: I'd like to have a way to generate a fully paginated pbook print image (basically, a PDF) as well as all the major reflowable ebook file formats without losing pbook page equivalence.

Though I don't think this will fill your every desire, if the project meets my goals, it will serve as a good platform to extend to yours. The real design problem here is how to have the output of one file feed back into the original source... and that is not difficult, its just time consuming to decide the best course. Kinda like solving a suduku puzzle... it takes 45 sec to write down 65-70 missing single digit numbers... it takes longer to decide how to write them down.

Originally Posted by moz

Come on, we can't even agree on spelling, what chance we can agree on sitation standards? You're all fulls!

I am going to make my tagline here "homonyms are not synonyms".

This is either a great observation, or a low blow to my monumental lack of spelling grace. Probably both. I entered college with a 5th grade spelling level, finished with a 7th grade level.. all things considered, I'm glad that you can read what I type and extract most of my meaning... My wife is currently trying to get my brain to tell the deference between sense and since... it took me more than one try to get spell check to move me to each of those words, so she still has a ways to go!

[kovidgoyal]

Is your friend planning to maintain this project after the 12 weeks are up?

[aapezzuto]

Originally Posted by kovidgoyal

Is your friend planning to maintain this project after the 12 weeks are up?

as time permits, for a few years at least

as a side note, the 12 weeks came from approximating 500 hours... not from an arbitrary time commitment or artificial deadline.

If the code is well formed and proves useful, and the project attracts enough attention, I'm sure it will get maintained, even if it is written in fortran, and the math statements use a special reverse polish notation mod.



I feel like a strange thing happens here... Like if I ask about what people want in a new car, i get 2 pages on the speed gauge being metric vs empirical, and a few posts on why the engine block should be made out of a specific steal.. not to mention that the passenger door should definitely be 42.5kg.

feel free to:

• state what you assume it would come with, just to be on the safe side (4 wheels, seats etc...)
• state things you want it to have (power steering, power windows, gps, cd changer)
• state things you want to do with it (carry a giant stroller, move furniture, go camping, carry up to 6 people)

Please avoid:

• imposing new design constraints (it must be made out of 100% recycled soy plastic)
• try to hijack design (and it should be 18.5m long, and use a turbine engine)
• solve for problems outside of the domain of the tool (and world hunger is bad)

Sorry to take the car analogy so far, but I feel like you all are hording the good ideas while giving me all this other stuff. I love the ideas here... and I know you all have the answers I'm looking for... But some of you don't make this easy on me... as I tell my girls though, most of the things in the world that are helpful or useful are not comfortable or easy... they don't understand yet.. but they are only 6 months old, so they have time.

[kovidgoyal]

See https://www.mobileread.com/forums/sho...91&postcount=8

EDIT: You should represent the document internally in some sort of object model and make that object model available to plugins

[Olympus]

Originally Posted by aapezzuto

I have been approached by a friend who is soon to graduate with a CS degree to outline him a reasonable 200-500 hour project. What we have come up with is an ebook formating program, written in C++, using QT, and thinking in epub. Its main goal is to be able to replace book designer as a more plug-in friendly opensource project. So as we are building the project outline and specification, and coming up with use cases, what would the community like to see in such a tool?

I am well aware of the talent in both programing and formating we have here, so I am excited to see what we come up with.

Although some might say that BookDesigner needs improvement, it's performing it's task....
I would suggest an alternative project (that has been tinkering in my own mind for some weeks now): a working e-book (file) manager.

Here is my problem: I have some 1000's e-books in various formats.
1: help to organize the files on my file system: author-folder\author - series/title - title - version
2: batch file rename with some help from external sources: author / title-database
3: identify duplicate formats (allow margin for ie. txt files)
4: create an XML-formatted - database: /author/series/series-readingorder/title/read or not/genre/cover-art/publish-year @story-type(short story/novel) @fiction @version @reader-format (txt/doc/prb/imp/html ...) @filename @file-path
(when @filename is missing then this becomes a "whish" ?)
5: import into database using all information that is stored in the file-path and file-name crossmatching author name variations, title variations, part of a series etc.)
5a: identify duplicates
6: Create/Manage Authors-Database with: genre - personal preference ranking
7: bulk update database with genre info using a separate Author database
8: Bulk - retrieve description/cover-art based on title & author from external sources like Amazon or ISFDB or Barnes & Nobles etc ...
9: "random" suggestion based on read/books read from same author/reading priority/genre

.....

I have use vBookSorter by Vangelis Dimou that handles some file management, but I haven't found ANY e-book manager that is able to extract all the information stored in the filename itself (like a proper MP3 library does)
My EbookManager is in itself very stupid as it stores all filenames as title's and it has no proper way to extract the author from the folder-name or the file-name....

These are just my requirements for a proper e-book management programme and I am eager to learn what people use now - or have other wishes for such a tool.

[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.

[HarryT]

Originally Posted by Olympus

Although some might say that BookDesigner needs improvement, it's performing it's task....
I would suggest an alternative project (that has been tinkering in my own mind for some weeks now): a working e-book (file) manager.

Can we not get side-tracked onto talking about other projects here, please. Would you mind starting a new thread to discuss this? Let's keep this thread for talking about the original subject.

[Olympus]

Originally Posted by HarryT

Can we not get side-tracked onto talking about other projects here, please. Would you mind starting a new thread to discuss this? Let's keep this thread for talking about the original subject.

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.

If the moderater shares your point then I would like to invite him/her to create a new poll and/or move my response to a new topic.

[Dave Berk]

Harry is a moderator