Indexing Project Instructions and Practices Need to be Revised
There are a number of issues involving indexing projects that need to be addressed. The following are just four that have come to my attention, either through my personal use of the indexed records (once they find their way onto FamilySearch and Family Tree) or through reading comments in the Indexing category of Community.
The problems would seem to relate to unsound practices, as well as inconsistency between “general” and ”specific” project instructions.
1. Marriage licence and banns records continue to be indexed as if these records related to the marriage event itself. Fortunately, many projects do now appear to be indexed so their data (date and placename) is indexed under a “Marriage Notice” rather than a “Marriage” heading. However, it appears the old practice - of not differentiating between a marriage licence, or banns record, and that for the marriage event itself - still continues.
The problem this causes lies in misleading FamilySearch / Family Tree users into believing the related record represents the correct information regarding the marriage ceremony. If indexed as a Marriage, the Family Tree source linking process often forces accurate marriage information off the Person page, as the associated (licence / banns) data will include an earlier date, which will take precedence over the (later) ceremony date.
There is also the position whereby these records are being indexed as marriages, yet a marriage (in spite of banns or licence records) might never have taken place. So, quite plausibly, a sealing event might be undertaken for a couple who were never married, nor possibly had any close relationship – purely on the basis of it being (reasonably) assumed the “Marriage” data in the indexed record was literally that.
2. In some cases, instructions appear to be issued to index any legible name or event, even if it has been crossed through. Recent reports have shown the reason for details being crossed through has simply related to them being found to have been recorded incorrectly. For example, a baptism or marriage being recorded, but did not actually take place. Or, a name of an individual being added to a list of family members and it being subsequently found that individual belonged to another family (of similar identity).
3. Where it is unclear whether a year recorded as, say, “83” refers to “1773” or “1883”, some instructions say that the two-digit format should be recorded. Apart from this of being of limited use, it leads to the year being recorded as “0083” once in the FamilySearch database, causing confusion and making it impossible to locate when undertaking a search including a “date range” input. It must be quite rare not to be able to establish the correct date / century from an original document, but if this cannot be ascertained, surely it is better to leave the date field blank in such cases?
4. There are a number of ongoing projects where it has been reported there are records included that do not match the collection title. An example is the “Northumberland Non-Conformist” collection, which, in fact, includes many records of established (Anglican) church(es) of neighbouring County Durham. The instructions to index material not relevant to the collection, instead of it being ignored (and left to be indexed, say, within an appropriately titled project at a later date) leads to researchers having great difficulty in finding these records once they go online. Logically, a FamilySearch user would input “Durham” for the county of their relative’s christening, but this would never locate the record when it has been indexed (knowingly in many cases) under the (wrong) county – of “Northumberland”.
It is my hope that these examples of poor indexing practices (illustrated as such by the examples provided above) might be read by, or passed to, the overall manager / team leader of the “Indexing section” for their kind consideration - in evaluating necessary changes to improve the quality of the data being indexed, and subsequently published on the FamilySearch website.
Unfortunately, the current (general and project) instructions appear to be very inconsistent and do not appear to be fit for purpose, in many cases. In some instances, the indexed material does not even act as a useful finding aid. It must be taken into account that there is often no link to the relating images, which themselves might even be difficult to view elsewhere (e.g., if located on a website with a paywall, or at a record repository a great distance from the researcher’s home).