FBReader now reads the OEB format. While many e-book formats are based on OEB, only two previous e-readers* could read the OEB format directly. FBReader is the open-source e-reader that runs on Linux desktops and the Nokia 770 (among other Linux-based carryaround computers).
Five thousand books from Project Gutenberg are available in the OEB format at GlobalMentor. Non-DRMed books in the Microsoft Reader .lit format can be decompiled with several openly available but potentially illicit tools, and the result can be read directly in FBReader. Since OEB facilitates use of separate source documents, FBReader permits all the files to be dropped into a zip archive and will read the compressed text directly.
FBreader isn’t yet available for Windows or Macintosh computers, but the source code merely awaits a developer ready to port it to those platforms.
Support for RTF was also announced with the release of version 0.7.4 of FBReader.
* As far as I know, GlobalMentor’s Mentoract Reader was never officially released, though downloads have been available since 2000. ION Systems’ Java-based eMonocle Reader seems to have been released without fanfare or subsequent development.
Added later: For the record, FBReader will read e-books in FB2 or FictionBook2 format, HTML, plain text, OEB, RTF, Plucker PDB, PalmDoc (aportis), zTxt (Weasel), and TCT (psion text), even inside tar, zip, gzip and bzip2 archives. Multiple titles can be read within one of those archives too. FBReader’s developer, Nikolay Pultsin, has indicated that the program will support OpenReader, non-DRMed Mobipocket and CHM formats.
Also for the record, FBReader does not support author-supplied style sheets nor can it yet process tables, a feature which is still forthcoming. Both aspects would apply to all the file formats mentioned, including OEB.
Funny, I was dealing with FBReader just today and was playing around with it.
Users have two different needs: building ebooks of their own content and taking web content by others and converting it to some ebook format. Also, they need to have some way to define optimal file size and memory usage of self-created ebooks.
I created a pdb of a single webpage with links to about 75 stories, and I’m finding the binary file is over 3mb long (which taxes FBreader and nokia both). Something is clearly not right. We need more transparency about this.
the other issue is formatting original content. It’s fine to support text standards, but but not having even basic css support or even graphics really makes it impossible to create content optimized for ebook binaries.
for the record, the latest package of fbreader on my nokia 770 did not even load when I installed it. I had to uninstall it and go back to a previous version. Still the release notes sounds promising.
for the record, the large size of the pdb was probably the culprit. After I changed the path for ebooks, it loaded fine. However, I’m still having a problem adding pdb files to the library.
First of all — what program are you using to make the pdb? Plucker? Sunrise?
I’m not especially familiar with those programs, except to say that it’s easy to get super large files by promiscuous setting of how far links should be followed and images included. If you include a link to my page and my page has links to three other pages, you may be getting 4 pages from that single link. If there are no links in your stories, of course, this wouldn’t seem to apply.
Second, FBReader was originally written to deal with titles in the FictionBook format, whose conception explicitly deprecates publisher formatting. Reading titles in FBReader, you are expected to pick your settings, not the publisher. (Of course, with fiction this has a certain clarity.)
However, that would apply only within FBReader. If you want to include styling information for others to use when reading in a different program, there’s nothing that prevents your doing that.
Third, you may want to try alternate formats — FB2, OEB, html, say — and drop these files into a zip archive and see how that works.
Fourth, an older version of FBReader did not recognize pdb files if they weren’t in a specific directory, but that has been corrected. How far back did you go? Also, I haven’t had any problems with 0.7.4. I urge anyone to bring their problems to the Google group for FBReader so that any bugs in the program can be identified and corrected (as I see you have).
So in general, what’s the easiest way to roll your own ebook for FBREADER? What about sucking up a webpage or multiple text files?
How do you read multiple files in a zip directory (say of html files with an html TOC)? I tried putting html pages in a single zip file, and nothing appears hyperlinked. I can only view only HTML file at a time.
BTW, here’s the only English-language documentation about FB2 .
I have experimented with making FB2 files and expect to add some material to fill the dearth of documentation on it. Not unexpectedly, I’ve gotten the best results using FB2.
For websites I have tried Plucker and Sunrise; I just haven’t used them enough to say I’ve mastered them.
I think OEB promises the best solution for a multiple-file scenario, with the OPF replacing a TOC file. Since practically speaking, FBReader is the first to implement OEB as a distribution/reading format (and not just for interchange), I haven’t located a full range of files to test this out in FBReader.
Robert,
With FBReader, Plucker is a pretty reasonable multi-file format for an ebook, and it’s designed to “pluck” websites. I’ve used it to package up entire sites like the Stanford Encyclopedia of Philosophy.
It supports embedded and linked images, and limited text styling (like bold and italics). When I last worked with it some five years ago, the biggest limitation was that sections (chapters) had to be short; long ones were stitched together with hyperlinks at the beginning and end of the sections. I don’t know if they ever got around to doing something about that. Text sections are compressed with the standard zlib “deflate” compression protocol. Hyperlinks both within the document, and to external documents, are supported (though the external links depend on your platform’s capabilities).
I am happy to report that the pdb file bug has been solved with today’s version. BTW, I just posted something related, how big should an ebook file be?