To Tiff or not to Tiff ...

[Diane Madrigal]

On Thu, 19 Jun 1997 Carolyn Gonzalez wrote (in part):

> The most intriguing comment I've come across in my search thus far came from a
> scanning vendor (let's just call him the TIFF-Man).  He said, "Suppose you've
> got a platoon leader sitting out in the desert with a lap top who needs to
> download and review two or three pages of a document before he goes into battle.
> If the document is PDF'd, he's going to have to sit and wait for that whole
> file to download before he can review those few pages he needs--and that might
> take too long.  With a TIFF, that platoon leader could download just what he
> needs and then go about his business."
> 
> Is this technologically possible?  If a document is being held as a multi-paged
> TIFF, could a library user simply download a few specific pages of the document
> rather than getting the whole thing?

Carolyn: 

At the New York State Library we recently began scanning State documents
as TIFF images. We don't do "multi-paged TIFF" images, however; each page
of a document is a separate TIFF image. I suspect the vendor was referring
to this type of setup when he talked about downloading just two or three
specific pages. Of course, you have to know exactly which two or three
pages you want - that's the tricky part! 

We've been creating an HTML page for each document, to provide links to
the multiple images that make up the document, and also to make it easier
to index and search for documents. We don't link to every page - some
documents are far too long for that - but we do provide a link to, at a
minimum, the cover, the table of contents (if there is one) and the
first page of text. Even if there is a table of contents, though, our
'electronic page numbers' don't match the printed page numbers because we
scan all pages (covers, blank pages, introductions with roman numerals,
etc.) plus we add some pages with scanning information. (I said this was
the tricky part.) At the moment, the best we can do is tell viewers how
much the pages are off by, a number which is usually consistent within
a document (i.e. you can find the electronic page number of any page in
document X by adding 8 to the page number listed in the table of
contents).

When we upgrade to version 8 of SIRSI, the catalog entries for scanned
documents will contain a link to the HTML page (which in turn links to the
images). There will also be links from our web site - we're still working
on that aspect, though, so the project hasn't gone public yet. (I can
direct you to some of our test pages, though, if you want to take a look.)

Another reason we haven't gone public yet is that we don't have a good,
free TIFF viewer for Mac users. If you or others on the list have
suggestions, we'd appreciate them!

Feel free to contact me off the list if you'd like more info. about this
project. (And I'd definitely like to know how your project progresses if
you go with the TIFF images.)

Diane Madrigal
dmadriga at unix2.nysed.gov
New York State Library
State Education Department
Albany, NY