FIND MY PAST SEARCHING - Warning

[Hardwork]

Recently I had cause to contact FMP because a search from the link on their home page to Baptisms, Marriages and Burials by parish resulted in a list of names returned that had no associated image from the parish register, yet I hade viewed the image before and was merely checking something. Going back to last year, I highlighted to them that searching for events in Norwich, Norfolk brought up events for the surname throughout all Norfolk, hence not just Norwich events. I was told that the search facility was working normally according to their technicians. As you can imagine, I was very disgruntled with that response.

It now turns out, after some too-ing and fro-ing with the help desk that the reason the image link didn't show was because it was in a different dataset. Apparently, the Baptim, Marriages and Burials search facility, though refined by location to town or parish level, isn't what one might expect it to be but is linked merely to the Family Search dataset. There is no mention of this that I can see anywhere on the home page in in a description of the search. This goes someway to explaining why I couldn't isolate Norwich previously. Obviously when I queried that last year FMP were searching totally differently straight from their county datasets, which worked fine. No one checked or even indicated the that the search facility I was using was a completely separate one.
Perhaps naively, I assumed a search from the home page would include information from all datasets in the absence of stating anything to the contrary on the search page, so be warned. What still isn't explained is how I found the image previously (2019) when I had never searched any countywide dataset before, never seeing the need.

[Mick Loney]

It explains my experiences lately.

I have being going through all my trancribed marriages/baptisms, to see if I could get an actual image of a parish register entry (I never trust transcriptions, until I’ve confirmed them myself )
I had previosly found trancriptions in various Findmypast datasets, but was surprised that now, FMP found nothing, neither transcription nor image.

[Hardwork]

Ditto Mick, about seeing an image from the actual register. It was a bit of a shock after 9 years of searching FMP for a one-name study to learn that "Search Baptisms, Marriages and Deaths" wasn't all-encompassing of the relevant material FMP holds. Were you using the individual county search facility or the one from the home page? As I had already viewed the particular register page on FMP I'm tempted to conclude that they have actually unlinked register images, at least in some cases, as I cannot see any other explanation.

[Mick Loney]

I start using search ‘Birth, marriage & deaths’ from home page
Insert full name of groom, year of marriage, and county. I use the exact year as I already have GRO reference or a transcription.
When I get no result for the Groom, I then double check by searching for Bride. Often this throws up a result, even though it show same Groom, so it should have been found on first search.

What I do find frustrating is when I find two transcriptions,which show two different places for same marriage! Then when I do find an image, discover that place of marriage is different from one of the transcriptions! These often come from Familysearch datasets. Normally one gives just the Parish, the other, a church within the Parish but in a different town/village.

[paulr1949]

I don’t know if it makes any difference, but when searching on FMP I always use the “search all record sets” button. I have a lot of ancestors from Norfolk (mainly rural) so I enter the county which then allows me to search baptisms, marriages, burials, etc, BT’s or browse the registers.

[meekhcs]

Like Paul I use the same method on FMP and Ancestry believing the sites are so large now that better results are achieved by looking at a definitive set where possible.

I can't remember which site I was using, or which method I was using, but a while ago I found myself unable to locate records I had previously found.

Perhaps another result of the site being tweaked by technicians, without a thought to the user?

[Hardwork]

I think the real point is that a search button shouldn't be created stating a search for something like "Search, BMDs" unless it is made clear and obvious that it isn't actually searching all BMDs. It is completely misleading for experienced and inexperienced researchers alike. The search criteria from the home page allows a parish and county to be added along with names, so has the ability to refine the search but to what avail if it doesn't link to all datasets containing that information?

Thanks for highlighting on how you all search. Searching by parish in a particular dataset probably works well, but I'm not not sure that searching all records helps for a one-name study, particularly where my method of working is looking for a surname parish by parish, county by county, and only in parish registers, but I think we can agree, each person will search in a particular fashion suited to their requirements. It is knowing exactly what data the search engine is actually looking through that is key.

[Mick Loney]

To add to this debate, I have just found an PR image for a Marriage in Easington, ERY, which clearly shows the date as 21 Dec 1879.

Yet Ancestry, in their England, Select Marriages, 1538-1973 dataset (from Familysearch), have a transcription which shows a date of 16 Dec 1879. One can only assume this was extracted from the Marriage Banns rather than the Marriage itself, but it is not made clear where the transciption derived from!

[meekhcs]

Mick

The difference between Bann dates and actual marriage dates is something that constantly crops up in transcripts of indexes as opposed to images of the original.

As is constantly mentioned the only thing you can 99% trust is to look at an original document, but even those can contain errors as I have found to my cost!

Hardwork
I totally agree when pressing a search button you should be made thoroughly aware of what the search will produce.

[Mick Loney]

Just found another quirk in FMP search criteria

Whilst searching for Marriage of Drusilla Eliza Holt in 1913, I carelessly left off final a of her first name, and entered Drusill, yet despite ticking use name variants for forename, it failed to find her! I had to include full name to get a result. One would think that Drusill was close enough to Drusilla to meet their name variant criteria, for FMP to find the marriage, apparently not!

I erroneously thought that, like GRO, entering a partial names was as good as using * at end of name (Drusill*) apparently not. Furthermore FMP is rather picky when entering forenames, so if you use Henry G Smith, you may get no results, but using Henry Smith or Henry George Smith you will.