The worst Ancestry transcription ever?

[Drayke]

But transcribers have to transcribe what is ACTUALLY seen to be there, and not correct it to what they might know (or suspect) it OUGHT to be.

And therein lies the rub ... "seen to be". With some of the more, shall we say, exotic manuscripts, different people can see different things. A bit like those ink blot pictures, where you see a butterfly and I see the Isle of Mull ...

I wasnt suggesting at all that they interpret it and change the name that is there. What I was suggesting is that if they are struggling to read the text, they should think about the name they transcribe and make sure that the name makes sense. If it doesn't, they have a responsibility to look for a name that makes sense and compare it to the writing to see if it matches or at least look at words with similar letters on the same page to match it.

If they cannot do that or can't at least use a manuscript writing identifier which you can find on the web, then they shouldn't be transcribing. Whilst I appreciate having the registers searchable, there isn't much point if the transcription is so obscure that you can't find the person.

For example, attached is the entry in the register. Firstly, Ursley clearly looks like 'Ursley' and not 'Custoy' and secondly, they didn't even bother to actually finish the last name and simply wrote Sal instead of Sallsberye which is also clear as day.

Even matching letters on the same page would provide them with the correct transcription. The word 'unto', literally next to Ursley, has the 'u' which is identical to the 'u' in Ursley yet they transcribe it as a C? If the 'u' in unto looks like it does, why did they think the 'r' in Ursley was a 'u' when it looks nothing like the 'u' in unto but more like the 'r' in Edward (or even more closely to the 'r' in from) that they transcribed correctly?

So based on all the ways they can identify the correct letters in the script I'm afraid I disagree. Using the excuse that "they saw something different" is just a cop-out and can't be used as the justification here if common sense is used, for this particular instance anyway.

[Copper1]

Point 1: Why doesn't someone submit an FOI to F/H companies to declare what "criteria" they have used in the past to employ somebody transcriber's?
Point 2: Is there a minimum requirement of C+ grade in (now), the King's English?

Don't let's forget back when, within 2 hours of the 1901 census going live, all manner of abusive language was found to have been used to apply to the occupations of police officers and prison staff.- allegedly but never denied because the task had been out-sourced to prisoners! How do we know now how qualified transcriber's were in the past? Too much secrecy and companies hiding behind 'commercial sensitivity'.

[BumbleB]

Transcribers - just to point out that FreeBMD transcribers are all named, although we are allowed to use "pet names" as we do on RootsChat, BUT we are all supervised by our particular Syndicate.  If you have found a transcription error then it can be reported, and a request made for a correction.  Assuming that we, the transcribers, agree that a mistake has been made, then a correction is carried out.  BUT there again we have the easy option as there are no time limits, nor do we get paid. 

I DO NOT have any information related to transcribers on Ancestry, FindMyPast, FS or any other body.

[RJ_Paton]

Point 1: Why doesn't someone submit an FOI to F/H companies to declare what "criteria" they have used in the past to employ somebody transcriber's?

Because a Freedom of Information Request would be quite rightly ignored by Commercial companies as they are not subject to it.

[Drayke]

Because a Freedom of Information Request would be quite rightly ignored by Commercial companies as they are not subject to it.

Not if that request came from the National Archives or other archival places that release these documents to companies to transcribe. They could even make a standard that needs to be conformed to as part of the release.

What is even more odd is when those documents have already been transcribed to a free searchable database and yet Ancestry doesn't provide that information to the transcribers to help them. How better would the welsh wills or the PCC wills searching be on ancestry if those transcribers knew that if they are struggling with a transcription they can simply go to the National Library of Wales or TNA websites and search for those wills from there and confirm their transcriptions.

Ancestry is the worst offender in this issue as well with not just transcription errors, but entire lines or pages simply ignored by the transcribers. It wouldn't be such an issue if Ancestry made available ways to insert a line so as the missing lines can be added but they dont do this at all. Same goes for the ability to redefine whether the record is baptism, marriage or death.

As to who these companies use, I am sure they could find many family historians that would be willing to help for free (similar to how FS do theirs) that have at least a better knowledge of reading manuscripts and an interest to ensure that those transcriptions are as accurate as possible.

[RJ_Paton]

Not if that request came from the National Archives or other archival places that release these documents to companies to transcribe. They could even make a standard that needs to be conformed to as part of the release.

Such a "request" is entirely different from a FOI application and is in the area of Commercial Contract transactions. It is also impossible to guarantee its success as humans always make subjective decisions within the parameters set them and double checking every single transcription would inevitably a) increase the costs to the end users (us) and b) increase the time taken to release new datasets.

The problem is that as long as these commercial companies are profitable there is unlikely to be any changes in the way they operate their bottom line is to make money and if they can keep costs as low as possible they will continue to do so. It would need a massive rebellion by their subscribers to force any change even the most minor of alterations.

[jbml]

One of the most entertaining ones I had for struggling transcribers turned out not to be an ancestor after all and had to be pruned from my tree. But I'm guessing he was a Welshman - Ieuan Day.

The Northamptonshire clerks had great difficulty with that "Ieuan" ... and the transcribers even greater difficulty trying to make sense of what the Northamptonshire clerks had written.

I had "Hewin", "Hewen", "Hughin", "Hugin" ... and by FAR my favourite ... "[Heaven?]"

[Drayke]

It is also impossible to guarantee its success as humans always make subjective decisions within the parameters set them and double checking every single transcription would inevitably a) increase the costs to the end users (us) and b) increase the time taken to release new datasets.

The problem is that as long as these commercial companies are profitable there is unlikely to be any changes in the way they operate their bottom line is to make money and if they can keep costs as low as possible they will continue to do so. It would need a massive rebellion by their subscribers to force any change even the most minor of alterations.

Yet, I would have thought that offering the transcription to volunteers would decrease commercial companies costs therefore increasing their profits. Why do they need to pay for transcribers when there would be plenty of family historians or others that would gladly offer the services for free anyway. As it stands now with the way transcriptions are on Ancestry, most users end up being the transcribers for free anyway.

FamilySearch has been doing this for years as has FreeREG, FreeBMD, FreeCEN, Family History Societies and other great websites with far greater accuracy and speed than what these commercial companies achieve by paying random people that have no knowledge of such documents.

[melba_schmelba]

Just going through some registers from the 1660s-90s on Ancestry, and my mind was boggling at how bad the transcriptions were, I think I was correcting about 80% of the entries. The person doing it had clearly had no training in understanding old handwriting, whoever was in charge of them, clearly did not care how accurate the transcriptions were, and provided no assistance, and also, clearly Ancestry also does not care how bad the transcriptions are, just that 'they are done'. There are many paleography courses free online, there is really no excuse for it other than greed and laziness on the part of Ancestry and the companies they emply to do it .