Andrew,
Since it's all JavaScript, all the customizations are open for 
inspection by viewing source or using a debugger.  While I didn't change 
any BookReader core code in hopes of making upgrades less painful, our 
customizations are a mess, very institution-specific, and I don't plan 
to put any work into packaging them up in a way that can be shared 
easily.  This was implemented about three years ago and I'm looking to 
start fresh.

I think it depends on what you mean when you say "get Djatoka working 
with the BookReader".  If you simply want to deliver images to the 
BookReader via Djatoka, that's relatively easy.  Start with the example 
demo [1], and edit the "getPageURI" function in the 
'BookReaderJSSimple.js' file by basically changing the "url" var to 
Djatoka's OpenURL format.  You will also need to tell it how many pages 
are in the book (which you can also set in that page), but that should 
be it.  For starters, you should simply size your dynamic JPGs to the 
sizes the BookReader expects from the IA architecture.

If you want to implement deep-zoom without hacking the BookReader's core 
code, that's going to require quite a bit of work.  Unless I really had 
to, I wouldn't go down that route at this point for various reasons.  I 
can elaborate, but I wouldn't want to bore anyone.

Since development on the BookReader has largely stopped, and it's 
perfectly fine at what it does, I've been more interested in developing 
a new object "viewer" that would support a number of different object 
types -- something easy to drop into any site without needing to know a 
lot of JS.  I have some strong opinions on what that should look like, 
but it's still early and much discussion around specifications should 
happen, probably on a different list.

[1] 
https://github.com/openlibrary/bookreader/blob/master/BookReaderDemo/BookReaderJSSimple.js

-Shaun

On 2/25/13 1:14 PM, Andrew Hankinson wrote:
> I would be interested in seeing your customizations. I've tried getting BookReader installed a couple times, and each time I got fed up with the install instructions, since it seemed specially tailored to the IA infrastructure. They mention that "others" have managed to get Djatoka working with BookReader, but I've scoured the Google and couldn't seem to find anyone who would share their code to get this working.
>
>
> On 2013-02-25, at 9:01 AM, Shaun Ellis <[log in to unmask]> wrote:
>
>> Kyle,
>> We have lots of old books too, and use the Open Library BookReader [1] for viewing.  It's been designed with the iPad and other tablets in mind.  I have customized it to work with Djatoka, allowing us "deep zoom" and other niceties of using JPEG2000 . However, out of the box, you can follow the Internet Archive's recipe [3] of zipping up a variety of derivative sizes, which works nicely as well.  It's pretty easy to set up.
>>
>> I should mention that I met a number of folks at the conference who are using the BookReader and interested in extending/adapting it in a sustainable and cooperative way, with recent projects like the IIIF Image API and OpenAnnotation integration in mind.  Let us know if anyone else is interested in being part of that discussion and development.  We haven't put together a separate mailing list or anything yet, but probably will get one together soon.
>>
>> [1] http://openlibrary.org/dev/docs/bookreader
>> [2] http://pudl.princeton.edu/objects/ms35t871w
>> [3] http://raj.blog.archive.org/2011/03/17/how-to-serve-ia-style-books-from-your-own-cluster/
>>
>> -Shaun
>>
>> On 2/22/13 7:50 PM, Kyle Banerjee wrote:
>>> We have a few digitized books, (some of them are old -- we're talking 500
>>> years). Sizes are all over the place but the big ones are easily the size
>>> of a large briefcase.
>>>
>>> We want to make these works more accessible/usable and there's some demand
>>> to make them available for tablets. What experience do people have with
>>> stuff like that, and what software/services/methods do you recommend?
>>>
>>> Source files are 600 dpi uncompressed tiffs so they're pretty big and
>>> there's nothing special about a book being over 10GB in size. Thanks,
>>>
>>> kyle