dorsey film conversion questions and estimate

[Pat Warren]

On 11 Jan 2004, Mark-Awakuni Swetland wrote:

> Bottom line, I'd recommend shooting everything to keep it together.
Somebody
> might find a use for what we are currently called "duplication".

There's a very important thing to notice when dealing with unpaginated or
multple-original-page-per-frame microform. I consider in this case, and
really with any microform manuscript sources, that in digitization, you're
actually working with two different sources. An original source and a
reproduction of it.

First, you want the content of the original (the writing) with its physical
structure (the separation of writing by layout and paging). So you want
here to have each card as 1 image. Along with this you need to write a good
description of the structure of the original with what information you have
via the film. If you have access to the original documents you can do an
even more thorough description of the collection.

Second, you have the film itself which may serve as the only (reasonably)
available identifier of the overall organization of the pages (or cards),
so in this case you want 1 image per frame of film. You here need to do a
good description of the film itself: sequence of frames, contents of
multiple reels, info about the filming, etc. I wasn't convinced of this
until AFTER scanning Iapi Oaye, so soon I'll have to get both the film and
fiche versions again and do a real good description of them. But learning
by trying's the only way to make it sink in. For me anyway. So from now
on...

> While my particular reader screen shows 1-1/2 frames that may not be the
> case universally. I have to center the upper and lower frames for best
focus
> (with tri-focals and a kinked neck)

The readers here at the U of MN have handled every fiche and film I've
tried. The Minolta readers have the three different lenses and can display
the full frame of any film from 9x-50x. I don't know of any film that's
outside this range in common use. Yet.

> Ideally, a single card per image would allow for maximum flexibility in
> sorting, deleting, etc. However, I would not impose that obsessive
> compulsive approach on someone other than myself. So the efficient
> comporomise that I would recommend is that we adopt a single frame per
image
> approach.

I don't mind sharing work, so I wouldn't mind just doing the scanning,
cropping, and straightening of the 1 frame per scan images. Then I can send
along the images to anyone else to separate, crop, and straighten as 1 card
per image. But I'll send along software to try too. The stuff I use is much
better for these functions than any imaging program I've tried, Adobe or
otherwise. My whole project is focused on compartmentalized processing
steps so each little step you take on the digital path produces something
useful, and each step could be done by a different person. The cooperative
approach is the way it should be.

In response to John K. mentioning concern about asking me to dedicate
months to this project... I'm offering. This kind of work IS my work now.
Not my paid work, but MY work, what I like spending my time doing and feel
very morally motivated to do. In addition, I think that making the
conversion to a digital research environment should become normal for
people in their own fields, not for specialized industries. Academia got
hooked on commerical publishing too much, rather than including the cost of
publishing in the budget of DOING academic work, it became something
external to academia and made it dependent on a commericalized industry for
a voice. That problem is reaching an extreme with the international
publishing giants and the academic journal racket. To spend $1,000 or more
now on digitization of one source probably to the exlusion of others, and
getting comfortable with always hiring out would be a sad direction to go,
I think. For now there's not enough people capable of doing the full
digitization themselves, but this is how you can set precedents. One main
goal in doing this work for the last two years is to set up a replicable
process that anyone anywhere studying anything can be trained on:
standards, processing steps, equipment, data encoding, interface
design...at the lowest cost possible in time and money. But beyond that
there's also the great american fear and suspicion of anyone who offers to
help and doesn't want anything material or monetary in exchange. I've
already had many people tell me that nobody will want the results of my
work if I'm not charging anything for it. There's some truth to it, but
people seem very capable of converting it into productive guilt. I'm not
saying these are your thoughts, John, but it's an important issue to bring
up.

As far as an obsessive compulsive behaviors in linguistics, I think that
might be par for the course. And for squirrels.

Pat