When it's about archiving and doc portability, it's all about PDF. I heard about DjVu somes years ago, and it seems to be now mature enough for serious usages. The benefits seems to be a small size format and a fast open / read experience.
But I have absolutely no feedback on how good / bad it is in the real world :
- Is it technically hard to implement in traditional information management tools ?
- Is is worth learning / implementing solution to generate / parse it when you now PDF ?
- Is the final user feedback good when it comes to day to day use ?
- How do you manage exchanges with the external world (the one with a PDF only state of mind) ?
- As a programmer, what are the pro and cons ?
- And what would you use to convince your boss to (or not to) use DjVU ?
- And globally, what gain did you noticed after including DjVu in your workflow ?
Bonus question : do you know some good Python libs to hack some quick and dirty scripts as a begining ?
EDIT : doing some research, I ended up getting that Wikimedia use it to internally store its book collection but can't find any feedback about it. Anybody involved in that project around here ?
I've found DjVu to be ideal for image-intensive documents. I used to sell books of highly details maps, and those were always in DjVu. PDF however works really well; it's a standard, and -everybody- will be able to open it without installing additional software.
There's more info at: http://print-driver.com/news/pdf-vs-djvu-i1909.html
Personally, I'd say until its graphic-rich documents, just stick to PDF.