View Single Post
Old 01-08-2008, 03:25 AM   #24
nairbv
Connoisseur
nairbv began at the beginning.
 
Posts: 88
Karma: 15
Join Date: Nov 2007
Device: still looking for an ebook reader device
regarding itunes, ... it's obviously not going to be an exact copy of itunes, but itunes has a hell of a lot of good ideas and nice features. I'm also curious what in particular you don't like about it though.

Being able to edit the author of a selected list of books for example, is something that itunes can do (but with songs of course), and the most ebook software currently out there can't do. Most software expects you to select each book individually. This is lame. If I have a directory with 37 plays of Shakespeare, I should select be able to select the 37 files, right click, hit "get info," change the author, and be done with all 37 books. Sure that doesn't enable me to edit 37 titles in one go, but at least I can make *some* progress in organizing files. I don't understand why so many of the programs I've tried didn't implement this basic feature.

Photogallery likewise is another piece of software with lots of good ideas to borrow from, like the way it handles tagging of photos. It also has a very usable ways of finding your pictures... like all the "folders" (not all of them actual folders, just buttons that dynamically get a bunch of photos based on tags or other search criteria) for grabbing a set of photos by tag, date, recently imported, etc etc.

Also, if you copy your files to another computer or try a different piece of cataloging software, in both of these programs you don't have to start over on a data entry process like with some of the ebook software out there. I mean, sure it's a lot more difficult with varying ebook formats, and maybe some of these cataloging programs have some way to export these databases but... still far from ideal.

This is what I like about ekitaab. His solution isn't perfect, but at least it addresses this problem. meta-data updates would be better if possible.

having all three options (cached in a database, updating metadata, and renaming of files that don't support metadata, as well as the searching functions for getting new data like in ekitaab and my ebook library), and then allow the user to override a default configuration. That way the user can decide if he wants to allow the software to change filenames, update metadata, or just rely on the database, ... that would be the best solution. Maybe not built that way initially, but the code could be written in a way to allow adding all three options and configurability later.

@acemccloudxx:

For the sake of re-usability, and since you're focused on java, maybe a good start would be to work on java libraries for managing content, ... for example, a start on a library for performing file conversions, or at least fetching/updating metadata in various ebook file formats.

You'll have to write this code anyways for the sake of importing metadata etc into your own database, but if you're careful about writing it well, and writing it in an extendable reusable way, the same code could also fill a gap in the ekitaab software at the same time. You'd be bitting of a smaller first chunk, actually getting something useful finished quickly, and still making progress on what you want to eventually build.

If I start writing some code, that's probably what I would focus on. Just writing little libraries that will be useful in a bigger project, and that add functionality that I want.

Likewise, even if you don't work on the other java project, you might be able to pull out some code like for however he does his amazon lookup for fetching ISBN numbers from file names.

This is supposed to be much of why we write OO code anyways right? If enough of the necessary libraries for managing ebook content get written, then anyone can slap together exactly-what-they-want in no time.

I wonder how re-usable the code in libprs500 is.... even for someone does want to make a custom exactly-what-they-want ebook app, ... if libprs is written in a way that it has a bunch of reusable libraries, it's still a big argument for working on python book management code. I mean, he's already got convert-damn-near-anything-to-lrf built right? Even if you don't care about lrf in particular, that at least means that he has a way of extracting data from damn near every file type, which is a big piece to be able to borrow, even if writing a new program.

Of course I'm assuming all this software is licensed in ways that the code reusable too...
nairbv is offline   Reply With Quote