Which calendar?

I guess this is not so much a problem as something to be aware of at Ancestry. I’m not sure how many datasets it applies to, but I’ve noticed in the Gloucestershire PRs that in recent weeks Ancestry transcriptions of early records have been changing from the Julian to the Gregorian calendar.

As always, you need to check the original image to be certain which year January to March dates fall into, but for those who don’t have a subscription it could get confusing.

 

I think it's probably pretty clear what I was referring to here. However, to avoid any ambiguity, with the Julian calendar the year began 25th March, and it is pre-1752 dates between 1 January and 24 March that I was referring to. So while a date may appear in a register as 10 January 1652 (for example) and was previously transcribed by Ancestry as 10 January 1652, now it is shown as 10 January 1653.

EDIT: I probably should also have said here, that although in England the change from the Julian calendar to the Gregorian calendar happened at the same time as the change to the year start and end, this wasn’t the case everywhere. So, although the old style dating is often equated in family history circles to the Julian calendar in England, I appreciate this isn’t strictly the case.

 

You are quite right about transcribers using different protocols, which is why, whenever possible it’s so important to check original records. Sometimes as well, differences can occur between a transcription from a PR and the same entry on the BT, and it can sometimes be harder to determine the correct year from a BT.

The reason for mentioning this now is because Ancestry is making changes to the protocol used, which might catch out some unsuspecting people.

A few old registers recorded some early events with the year starting 1 January rather than 25 March, and some kept changing from one to the other, all of which can add to the confusion.

 

What I find unsatisfactory is the lack of any explanations by platforms of how they deal with year numbering, as it's inherently confusing and some guidance is needed. The best example of this confusion is to ask yourself "how many days there were in February 1748 old style?".

Any search engine that uses a full date (day and month as well as year number) will be very awkward to design and to use if the year start switches during its range. Newspaper archives need to do this, so I had a look at the BL one to see how it coped. It uses 1st January throughout, but that is really the natural option for them. Their holdings don't go back further than 1700, with few titles before 1752, and all the ones I could find use dual dating for their masthead dates. But I could not find any help on this topic. I think I've come across some searches on Ancestry that allow full date entry, so this may apply to them as well.