Just thought I might plug some software we're developing to solve the book image navigation "misery" that Kyle mentions.

and a demo:

We developed it because we were frustrated with the "image gallery" paradigm for book image viewing, and wanted something more like Google Books' viewer, but with access to the highest resolution possible. We also were frustrated with having to download large PDFs to just view a couple pages.

Diva uses IIP on the back-end to serve out image tiles, so you're only ever downloading the part of the image that's viewable -- the rest is auto-loaded as the user scrolls. 

We've used it to display a manuscript that's ~80GB (total), with each image around 200MB.

It's also got a couple other neat features, like in-browser brightness/contrast/rotation adjustments via canvas. (Click the little gear icon in the top left of each page image).


On 2013-11-08, at 4:22 PM, Kyle Banerjee <[log in to unmask]> wrote:

>> It is sad to me that converting to PDF for viewing off the Web seems like
>> the answer. Isnít there a tiling viewer (like Leaflet) that could be used
>> to render jpeg derivatives of the original tif files in Omeka?
> This should be pretty easy. But the issue with tiling is that the nav
> process is miserable for all but the shortest books. Most of the people who
> want to download want are looking for jpegs rather than source tiffs and
> one pdf instead of a bunch of tiffs (which is good since each one is
> typically over 100MB). Of course there are people who want the real deal,
> but that's actually a much less common use case.
> As Karen observes, downloading and viewing serve different use cases so of
> course we will provide both. IIP Image Server looks intriguing. But most of
> our users who want the full res stuff really just want to download the
> source tiffs which will be made available.
> kyle