Quote:
Originally Posted by HarryT
Could you elaborate?
Both LIT and MobiPocket are based on the Open eBook (OEB) standard. Both are essentially HTML within a custom "wrapper"; that's why it's so easy to convert from LIT to Mobi, or vice versa. To the best of my knowledge, they have pretty much the same capabilities.
In what way do you consider LIT to be better?
|
I’m not sure of the exact history, but I believe Mobipocket has been around since before the OEBPS. They’re part of the IDPF and had a say in the standard, but that would mean Mobipocket’s format informed the OEBPS rather than vice versa. So it’s certainly possible to produce a Mobipocket book from an OEBPS book, but it’s a destructive transformation – the original OEBPS book cannot be fully recovered from the Mobipocket book. MSLIT conversion on the other hand appears to be a fully reversible operation.
Three classes of information are lost in Mobipocket books:
- The exact contents of the OEBPS OPF file. Many Mobipocket books sometimes preserve the contents of the ‘metadata’ and/or ‘guide’ elements, but this does not appear to be guaranteed.
- Directory structure, file names, and file separation. Mobipocket generation merges all HTML files into a single unnamed stream. All image filenames are lost, reduced to their PDB record numbers. Any other files which were part of a source OEBPS book are lost.
- Semantic markup and CSS. The Mobipocket HTML renderer is not CSS-based, so Mobipocket conversion destructively translates the OEBPS’s semantic markup plus CSS into purely presentational markup with some proprietary features. All semantic classes are lost. The source CSS is lost. All named anchors are lost, converted to file positions.
Is this actually a problem? I don’t know, but I suspect that it will be easier to convert with cleaner results MSLIT books to future formats than it will be to convert Mobipocket books.