A Deep-dive Database of Local History, Attitudes, and Ideas

Recently I purchased a small group of mid-Hudson Valley material that I found useful as examples of what would logically be included in a deep-dive experimental database for the New York State counties mid-way between New York City and Albany. A what? A deep-dive-database is a full text searchable database, something like what Google does in its Books section. Whether it is a d3 or a FTSD or something else remains to be seen, but it is the future.

Databases of the printed word have generally been confined to brief descriptions and details of books and printed documents. To see an actual copy, for example if you are using the OCLC, you are provided locations where such copies, physical and electronic, are found. On RBH we focus on auction records and dealer descriptions to illuminate the emerging understanding of an example’s importance and value. Such databases are potentially very large as ours is, more than 9 million full text records.

But what is now emerging are full text databases. That is, they capture the complete contents of a document in word searchable form, not only as a scan but as a word document. Some efforts currently look for references in text but they have generally been dull instruments, in some cases because the references need to be dug out and in others because they are behind paywalls. This will change and with this change there will be full text readable versions searchable online – and in many cases, searchable for free.

This experimental free database for the mid-Hudson Valley will include the standard reference materials, town and county histories, maps that convey changes, appropriate books by local authors, broadsides, pamphlets and ephemera – all in full searchable text.

The search will be different because the most common form posted will be ephemera that will outnumber books and pamphlets somewhere between a thousand and ten thousand to one.

Books usually include the title, author, publisher/printer, place and date printed. When even one of these facts is missing it can complicate searches. For ephemera you might be lucky to have three of these factors. The others will require associated factors such as “they are among a group of letters in the same hand”. Here’s an example. A collection of letters from A.M. to B. R. dated by day and month but not by year. However, one envelope is dated 1863 and the events mentioned suggest the Battle at Chancellorsville. Can this be figured out? Probably. As this example suggests, judgments will be made.

Here are some of the fields needed to identify and contextualize such letters.

Date or date range stated or implied

Names implied or known

Subject[s] such as events and places

Regimental references and information including cross-references

In addition, other fields will sometimes play a part:

Watermarks

Context of the document [among a group of similar items or with other related materials]

References gleaned from genealogical sites

References from online searches on Google and others

Altogether it will often, but not always, be possible to contextualize material, thus creating a deeper perspective – a perspective I believe that will change our understanding of the past.

Here are some other examples: Ulster Mine at Ellenville, Ulster County, New York, a series of 5 printed documents, many with illustrations, that relate to this mine from 1852 to 1855 that include:

A 16 page report dated July 1^st, 1852

An abbreviated broadside version dated July 1^st, 1852

A 12 page report dated December 10^th, 1852

A broadside, brief financial statement dated 15^th December, 1852

A 16 page report dated January 3, 1854 titled Official Reports of the Ulster Company for the year 1853

This mine was located a short distance from the Delaware & Hudson Canal and was opened in 1852 during a period when Americans were looking everywhere for gold because of the stories emerging about the gold strikes in California. In Ellenville they found lead while in Kingston some 20 miles away they believed they found gold that, when assayed, turned out to be pyrite or fool’s gold. Such documents are so much more interesting than a title, date, author and print date.

Among the other documents I purchased is a stock receipt for the Hobart Branch Railroad Company signed by Thomas Cornell, who was a man of wealth whose steam boats coursed the Hudson River in the latter half of the 19^th century. He was based in Rondout but his influence reached in every direction.

Another is a menu for the Hotel Kaaterskill at Catskill for Thursday August 24, 1899. Tastes have changed!

A small one is an 1857 7.625” x 5” broadside circular calling on teachers in Orange County to participate in a quarterly meeting to be instructed on new teaching approaches. The teachers were expected to pay their own way but a handwritten note suggests the costs may be shared.

These are a few of the many documents that will contribute to an understanding of what life was like and altogether convey the changing assumptions and understanding people generally had. Life has never been a paved highway and in the mid-Hudson Valley it seems more like a gravel path; every spec of gravel evidence of unique personal history.

An intensely focused, full text searchable database will bring these details to light.

Images of some of the examples are included with this article.