djvu: an alternative to scanned pdfs?

[urlwolf]

http://www.lizardtec...cts/doc/overview.php

This seems like a very reasonable thing to have around.
Not sure if anyone has calculated average space savings...

[Lashiec]

As always, Wikipedia sheds some light on the matter. For your PhD papers, doesn't seem to bring much advantage, but for magazines it could be great, and considering the size of some digital magazines, I'll give it my vote. The problem is that, although the format is a free implementation, the available options for a free encoder are not exactly "glamorous" (command-line based tools) and they produce worst results than the commercial options.

Now my dream is being capable of digitalize my entire collection, so I can save some space in the shelves

[Armando]

This seems like a very reasonable thing to have around.

-urlwolf (July 14, 2007, 06:01 PM)

Yes it is...

But I still have ambiguous "feelings" about it.

When I first wanted to use it, I wondered about its future... the pdf format is almost universal and very well supported, but not djvu. And if you're going to archive things, it's an important factor.

Also, I didn't really like the available viewers : I couldn't insert comments on the pages, I found that reading stuff compressed in that format was a bit hard on the eyes. It might be OK for images though. Your experience may vary...

Anyway, when after some tests, I didn't find that it was that worth worrying about using it for my own archiving.

[app103]

This format is only good for image files...scans.

Seems like it would be best for all those out of print public domain works. It would save much time from the furrent method of OCR and saving just the text, by eliminating the need for proof-reading. And the small drawings in many of them wouldn't end up lost as they do with the current method of saving just the text.

I was just thinking about old sheet music too. And antique childrens books.

[Armando]

Seems like it would be best for all those out of print public domain works.

-app103 (July 15, 2007, 07:19 PM)

I completely agree.

But I have a couple books here (scanned and compressed in that format), and I find them a bit heart to read.

I guess it depends what you want to do with the files.

[mikiem]

Well, looking at the testimonials page, the 1st thing I saw was an assortment of typos... For archiving purposes I'd want the providers to have an overwhelming obsession with detail, rather than a "good enough" mind set. 

The second thing I saw is that it was developed by ATT, which is cool I guess from a tech viewpoint, but a bit worrisome when it comes to legal. Are viewers going to be available years from now, or the victim of lawsuits like the Lucent MS mp3 stuff?

The Wikipedia article was interesting, but misleading... PDF docs are made by a lot of different software since Adobe opened it up a bit. PDFs are text when text is available, often raster otherwise. In fact, AFAIK there's still no universally great trace software to convert scanned, raster images to vector. In that sense the comparison is faulty, as are stated size differences.

Common thought with video seems to be to retain the original, simply because the delivery format is subject to change over years. It makes sense to me to take the same approach with documents -- archive the original scans, save the OCR if performed, and use a compressed format for viewing.