INDIANA CENSUS PROJECT. There's been a lot of interest among participants in 
the National Genealogy Echomail Conference concerning the possibility of 
computer genealogists cooperating in a project to index census records. In 
spite of the magnitude of the job and the obvious logistical roadblocks, these 
folks - like a lot of us - dream of the day when they will be able to press a 
few keys on a computer keyboard and instantly locate an ancestor in a given 
U.S. census record. 

We're not there yet, but at least one such project has been completed and it 
provides a roadmap for others and a glimpse at the enormous benefits the 
computerization of census data can provide genealogists. 

The Indiana Historical Society, in cooperation with Indiana University-Purdue 
University Indianapolis (IUPUI), has completed indexing the 1860 census of 
Indiana and the results are now available to researchers. 

The indexing project began in May of 1985 and was completed in April of 1987. 
Entering data was accomplished at the home of Ruth Dorrel, an employee of the 
Indiana Historical Society, in St. Paul, IN. ISPUI designated her place as a 
"micro site" and furnished three microcomputers (two IBMs and a Zenith). Also 
furnished were microfilm of the census purchased from the National Archives 
and three microfilm readers. Data entry was done by Ms. Dorrel and two others 
hired for the task. 

Computer specialists at ISPUI wrote a C language program for data entry. The 
program screen closely approximated the page layout of the 1860 census, 
including the information at the top of each page. Among the features of the 
softwarare: Header information was automatically carried forward and needed
changing only when an item of information changed, such as the name of the 
township or census area. Household/dwelling numbers, page numbers and other 
elements were automatically incremented. Since a large majority of the folks 
were born in Indiana, that state was automatically entered for the state of 
birth field; the operator had to make an entry only when the birthplace was 
elsewhere. The microfilm reader screens were set at eyelevel in front of the 
computer keyboard with the computer monitor off to the side where it could be 
checked as necessary. 

Even with a "streamlined" process like that, looking at some of the figures 
tells you why it took three people almost two years to enter all the data. The 
database contains 1,354,617 records (there are not quite that many individuals 
since some enumerators recorded vacant dwellings and these were made part of 
the database). In all it took 653 360-K 5-1/4-inch floppy disks to hold all the 
entries. All these records are now stored on the mainframe computer at IUPUI. 
Along with necessary indexes and other controls it takes up about 314 megabytes 
of storage (in mainframe parlance, 6,637 tracks of 47,476 bytes each or 440 
"cylinders"). 

During a phone conversation, Sally Fadely, ISPUI computer specialist who 
oversees the database, provided me with a glimpse of potential benefits for 
genealogical researchers. 

Could I, for instance, get the records for all those with the surname Pence and 
Pentz on a floppy disk? Piece of cake. She asked for the soundex codes. As 
quickly as I could figure them out (P520 and P532), she reported that there 
were 803 and 427 individuals, respectively, with those soundex codes. She said 
the database has been indexed on the soundex code, but there is the additional 
ability to select only certain names within that code (e.g., Pence and not 
Pines). Downloading all those with the above codes would take only seconds and 
one floppy would hold all the records. 

What would it cost? Only a guess, Ms. Fadely said. No way it would be as much 
as $100; maybe as low as $20. Mine would be the first request for data by 
surname, so she couldn't be more specific. 

Ms. Fadely said orders for data from the census should be made on forms 
provided by IHS. This form includes a use agreement which must be completed. 
(Understandably, after such a long and costly project, the Indiana folks want 
to make sure someone doesn't "appropriate" their work.) After the form is 
received, you will be given a cost estimate for getting the requested 
information. Address of the Indiana Historical Society is 315 West Ohio Street, 
Indianapolis, IN 46202. Ask for forms for ordering data from the 1860 Indiana 
census. 

Information on floppies is provided in ASCII (standard text) format and each 
user must provide his or her own software to manipulate the data. Any database 
manager which can accept an ASCII file (dBase III, for instance) should do the 
job. Ms. Fadely says that the information can then be sorted in a variety of 
ways. A sort on dwelling number, for example, would group people by households 
and indicate relationships as well as proximity. 

The 1860 census is available on floppy disk through IUPUI by individual county. The cost is $200 for each county, plus $1.00 for each floppy disk. Computer 
tapes for a mainframe are also available. These likely will be used for such 
purposes as statistical analyses as well as for genealogical research. 

The IHS is selling microfiche copies of the 1860 census. This version of the 
census is a straight alphabetical listing by surname. The entire set costs $300 
but individual sheets are $5.00; a single surname such as Pence likely will be 
found on one sheet. Depending on where the page splits come, others might take 
two or more sheets. IHS shortly will have available microfiche listings by 
county; the price for these hasn't been determined. 

The computer in the reading room of IHS reading room has dial-up access to the 
IUPUI mainframe. Users must first have an account number with IUPUI for billing 
purposes. On-line costs are $15 a minute of central processing unit time (only 
the time the mainframe is actually working on your request - you won't be 
charged for the time you spend fiddling with the keyboard or looking into space 
while you wonder what to do next!) Someone familiar with the process likely 
would be charged for only a few seconds of computer time for a specific 
request. 

Ms. Fadely said CD-ROM (compact disk-read only memory) is being investigated as 
an additional means of distributing data. Costs and lack of standards for both 
CD-ROM hardware and software are two current drawbacks. 

Those who keep records on a specific surname, such as for a family association, 
can quickly grasp the value of getting all the data on that surname for one 
state for one census year on floppy disks. Even at $50 or $100 the price might 
not be too steep, considering the number of hours it would take someone to 
extract the same information by hand. Even with an index, it wouldn't be 
unusual to spend two or three days copying one surname for one state. Or 
suppose you know the name of a great grandfather who was a child in 1860, but 
haven't found his father because you don't know his given name. Find the child 
in the census, relate it to others with the same dwelling number and you home. 
Harried genealogists who have difficulty finding time for research or can't 
travel to where the research facilities are will understand the implications. 
                                                               
I'm waiting for the order form so I can get my data request in!