Awful transcriptions

[ElizabethHerts]

I've come across a few in the past 24 hours, but this one made me cringe:

Bertha Pomlmoom
male
born Burton on Strent

https://www.ancestry.co.uk/discoveryui-content/view/5585742:2352?tid=191210385&pid=392490356143

[Olde Crone]

My favourite is (can't remember his name but will remember his occupation for ever)

Samartive Beguine Bleaner.

I followed a local family called Jepson. There were hundreds of them, a wellknown local surname. Ancestry rendered it as Jepson Jobson Sepson, Lepson, in fact any alteration to any of the six letters. I must have send dozens if not hundreds of corrections to Ancestry.

OC

[Pinefamily]

I came across one the other day. In the BT's for Ashleworth, Gloucestershire, there was an entry, "ye son of John Roberts and wife Jane". Transcribed as Ye Roberts!

[Phoenix]

I have taken to browsing the early Hampshire registers because the handwriting is so poor. I'm impressed that the indexers can make anything of some entries... and note that in many case they cannot, so the entries have not been indexed at all.

[ElizabethHerts]

A lot of the Ancestry transcribers seem to have no knowledge of place names and no curiosity to learn about the geography of the country.

[Olde Crone]

Phoenix, oh, I agree, some originals are impossible to read. My smartive man though, the original was very clear and I wonder why no one ever double checked the nonsense. If English isn't your first language why are you doing transcriptions.

OC

[kiterunner]

I assume the transcribers have to produce a certain number of records per hour, so they have no time to learn about geography, unless they do it for free on their own time. As for why they use transcribers who don't have English as their first language (if they do) it would be because it's cheaper.

[Phoenix]

You probably get just as good accuracy with someone who doesn't know an area as someone who does, since the temptation to bring your knowledge to bear must result in well-meaning errors, where you have read into it what isn't there.

[Olde Crone]

I think the Ancestry transcription I'm referring to, was done by prisoners in the USA (?) and many of them didn't have English as their first language. As prisoners they would have been paid peanuts of course, resulting in a poor job, for which Ancestry subscribers paid a lot! Accuracy was never the prime objective.

OC

[Phoenix]

I have just been battling with the Hampshire Record Office catalogue (no longer the CALM database, and frankly pretty poor) where I am fairly sure that Lovey should actually be Covey. Now that, surely, was not transcribed by prisoners?

I can take mistranscriptions. What irritates me beyond measure is missing pages. When we did it by hand, we would count the pages, check they followed on, check we had transcribed everyone. This, surely, is actually nothing to do with transcribers, who presumably start with the digital images? It happens so frequently that I despair.

[kiterunner]

By the way, when I was volunteering for Ancestry's World Archives Project I transcribed lots of records which were not in English because there weren't enough (or any) volunteers who were fluent in the languages we needed.

[Jill]

There seem to be an awful lot of people born in Mexico living in London, because Middx often gets transcribed as Mexico.

[Olde Crone]

Kate

But you as a volunteer and a serious researcher, have an interest in accuracy which means you are more likely to puzzle over something you don't understand or doesn't look right. I can excuse mistakes made by volunteers (usually) but not glaringly-obvious mistakes for which someone has paid good money. Doubke keying might have been a good start.

OC

[kiterunner]

Anyway, they (Ancestry, at least) are switching to using AI for transcriptions wherever possible now.

[Mary from Italy]

Oh dear, really?