Message-Id: <mailto:199407142104.QAA28384@library.wustl.edu> Date: Thu, 14 Jul 1994 12:27:20 -0700 From: Robert MacKimmie <mailto:rm@CALIFHISTSOC.ORG> Subject: Re: Group III TIFF images (Introduction--yet another visual To: Multiple recipients of list IMAGELIB <mailto:IMAGELIB@ARIZVM1.BITNET>
Here at the California Historical Society, we have been pushing a "Visual Imaging/Automation Project" for the past 2-1/2 years with great success. When I was getting the whole project off the ground three years ago, everyone said, "visual imaging, what's that?," so obviously the world has gotten much, much better. Computers are powerful and affordable enough to be able to do something substantial, and now that the "Info Hwy" is a household word, people tend to "get" what it is that we are doing and "why".
We are a research library with very high caliber material. Half a million photographs are well represented by Carleton Watkins, Eadweard Muybridge, Arnold Genthe, Minor White and even Ansel's early work. We also have maps, broadsides, rare books, ephemera, etc.
We are dumping it all into electronic form, with the main objective of setting standards for full and comprehensive intellectual access via a system that is powerful, easy to use, and with access methods that meet the state-of-the-art standards for libraries, museums, etc. This includes high quality digital images.
Just for perpective, I have been using Macintosh since the release in 1984, but the platform ran out of power and vision, so I have migrated on to NeXT Computers which seems to embody the highest standards in the computer industry today, being even friendlier than a Mac to use, but it is a client-server UNIX based workstation, so it is REALLY POWERFUL and a pleasure to use. I have had to have great confidence as NeXT stopped producing hardware, concentrating solely on Software. It is such a good system to use that it is worth enduring all of the "I thought they went out of business" comments. The NeXT operating system, NeXTSTEP, is now running on 486, Pentium machines, the HP "Gecko" 712 RISC machine, Canon's new Object.Station 486 (100 mhz) machine, SUN Computers (by the end of the year), and so on. NeXTSTEP is the first widespread system to be a fully developed object-oriented system software, shipping in version 3.2, while Taligent (from Apple-IBM) and NT/Chicago/Cairo (from Microsoft) are both vaporware with slipping arrival dates. The object-oriented software is not as simple to produce as it sounds, but NeXT is fantastic and is proven. <Diatribe off...>
What we have done with it is: We are scanning in various types of materials, and doing very full and complete cataloging via full USMARC cataloging directly into the RLIN (Research Libraries Information Network) international database. MARC is a standard cataloging format that is universal like ASCII or TIFFs.
We are really trying to push the boundaries by having VERY FULL subject descriptions using all of the standard subject thesauri: LCSH (Library of Congress Subject Headings), LCTGM (Library of Congress Thesaurus for Graphic Materials) and AAT (Getty's Art and Architecture Thesauri). Few institutions use all three, and we realize it is very indulgent, but all three sources for subject terms are complete or robost enough to use as a sole source. All three combined actually provide the type of subject/image retrieval that the field actually needs. We also run the MARC database on an in-house OPAC (Online Public Access Catalog). We happen to use VTLS' InfoStation [Dr. Vinod Chachra's post is what prompted this] to demonstrate our automation project because it is such a fine interface, and VTLS has such high standards as a bibliographic utility. InfoStation is the multimedia front end to a very robust bibliographic database. They are very fine and responsive, so five stars from me regarding being a quality vendor in a field that needs the finest tools.
For the imaging aspects, we scan as high as possible dpi. We do 300 dpi because that is the limit of our scanners without interpolation. 600 dpi would be much preferred, as the images could then be used for electronic publication [kills many birds with one stone]. Obviously this generates huge electronic file sizes, so we create an online copy at a lower "screen resolution" 72 dpi, and put the "high rez" file on a back up "optical disk" which holds 250 megs each. The opticals fill up quickly also, so when I have 10 optical disks filled up, I dump those onto a 2 gigabyte DAT (Digital Tape) backup. These tapes are $20 each, so are a good storage bargain. I don't consider these to be a long-term archiving technique, but in five years when new technology brings online storage down in cost and up in capability, things can go back on line. I am anxious for the "terabyte" drives at a modest cost, and it won't be long.
We are scanning photographs, broadsides, maps, rare bood illustrations, etc. We are also OCR scanning printed materials so that descriptive guides, finding aides and publications can go online in electronic format and thus be very useful.
With a client-server network, fax-modems, printers, scanners, cd players, etc, are all shared and quite reasonable in terms of creating a great system.
The results are astounding and we tend to amaze whoever stops by for one of our "dog and pony" shows. Our adherance to the top standards in the research library/museum field has prompted many collaborative and progressive efforts and we truly feel that we are helping to invent the future by dreaming about what we want to have in ten years, instead of asking what we can buy today. Pushing the cutting edge of technology is a lonely thing, but having a really robust, powerful and respectable system is a big point of pride.
It may be of little suprise that we have essentially built a rocket to go to Mars, and our board of trustees haven't looked out the back window to see what we are doing, or they have, but don't understand the significance. My whole point is that while a major university would require two years of committee work to decide if there coffee should be served at the meetings, we have been able to deploy a rather amazing and successful $1/4 million visual automation project based on the highest standards and utilizing modestly priced, but powerful client-server computer networks. All we lack is the direct internet connection which will allow all of our work to be shared with all of you.
My standard offer: if anyone is in the Bay Area, or visiting, and would like to come by and see what we have put together, please let me know. We always are anxious to share and compare notes.
Also, anyone with access to the RLIN database, our institution code is CHSV and I can provide printouts of a few scanned images if you want to compare image with cataloging record.
Happy automation,
Robert MacKimmie Curatorial Director of Photography California Historical Society, San Francisco 415-567-1848 mailto:rm@califhistsoc.org