What is the convention about Gregorian Calender dates?

Question

Since the Gregorian Calender was introduced, (Between 1582 and 1870 or so) often dates given in the first 10 weeks of the year are given two years (15 Jan 1703-4).

Does WikiTree have a convention as to which year (since only one can be entered) is used when reporting birth/death/marriage dates?

Comments

I know it is always important to enter the date as you find it. Therefore a comment in the bio is very important.
Have been using the 1st part 1703 from 1703/4.  I answered so I could receive any other comments. Its an important issue.

Answer 1

Michael and Tom,

Really important question. Thanks for bringing it up.

Here is the current, terribly incomplete answer from http://www.wikitree.com/wiki/Date_Fields#Julian_vs._Gregorian_Calendar

 

WikiTree cannot accept dates such as 1753/54 so one year needs to be chosen.

Our general convention, as with names, is to use the standard that the person themselves would have known and used.

(Is this a sufficient answer? When the standard calendar changed from the Julian to Gregorian in 1752, was there a period of uncertainty?)

What do you think? Is there a cutoff date we can use?

Comments

As if we could know what they thought then about the date....The convention changed depending on where you were. it started in 1582/3 in the Vatican, Alaska change it after 1860 something. I'll use the earlier date,  noting the older date in comments.

---

Tom, do you want to do some research on conventions, and then we can offer more guidance to others?

---

I'll be glad to, given somewhere to find guidance....

---

Tom - I'd recommend this Wikipedia article and the linked pages...  http://en.wikipedia.org/wiki/Old_Style_and_New_Style_dates
Cheers

Answer 2

As has been noted, standardization in this (as in most other genealogical issues) is categorically impossible. Over the course of the three-century or so transition, a person might very easily have gone from a region that used OS to a region that used NS (and/or vice versa) multiple times over the course of his or her life. Short of a tool that converts the dates, I think the only recourse is to note in the bio which style is used, where that can be ascertained.

Comments

I'm wondering if we should make the WikiTree convention be to use the Gregorian calendar in all cases (when possible), regardless of what people in the given time and place would have been using.

I'm all for explaining things in the bio, and that will still be necessary, but since dates are essential for searches and merges, one convention would have a lot of value operationally.

Oh, and regarding a tool that converts, that would be relatively easy to add later -- if we know the direction we're converting from. That is, if all dates are assumed to be Gregorian we can generate or display them as Julian automatically.

---

That makes a lot of sense to me, but I'd recommend also instructing people not to obsess about it! ;)

---

I did some research many years ago when writing a calendar program.
Here is a link with some times of adoption of the Gregorian Calendar and its' implications.
 http://www.webexhibits.org/calendars/year-countries.html
Exact dates are somewhat useless unless you know the convention being used in a certain time and place. So what if the date of an event in Italy is reported in England as being on the 20th of October 1582. Which calendar system is being referred to? England had not yet adopted the Gregorian system.
The Julian system is still used today, mainly in Astronomy. It is basically the number of days since Jan 1, 4713BC.
Paul

---

Great link, Paul.

What would you think of making our convention be to use the Gregorian calendar if it all possible (including using a table like the one on http://www.webexhibits.org/calendars/year-countries.html to do the conversion, if you can)? And, of course, explaining what you've done in the text of the page.

Then, down the road, we work on a little program to do automatic conversions from Gregorian to Julian or other standards. I'm starting to think this program wouldn't be easy, but still feasible in a limited way, and helpful. With your experience, Paul, what do you think?

---

The code for calculating Julian dates is fairly simple. I still have the formulae from the Turbo Pascal program I wrote in the dos days. The difficulty comes in the numerous and varied dates of adoption. Many countries even adopted the change on a region by region basis. This would require greater geographical knowledge than I possess. And Wikitree locations are not "normalised" so interpreting location data would be fraught with problems.
The main advantage of Julian dates is the ease of calculating times differences. Subtract one date from another and that is the number of days between them.
As I see it the dates we use on Wikitree should be as they were reported  in the time and place they were first used.
And we haven't mentioned the Islamic, Hebrew or Chinese calendars either (amongst others). What about the Mayan calendar? Oh, wait, the world comes to an end this year, so we dont have to worry about that one!
Bottom line, like Baptism info, or Burial info, which a lot of people put into the Biography section, perhaps just an explanation of the date system used, or a note to say the date indicated may be inaccurate due to the calendar used.
Makes me wonder about all sorts of events. Christmas, December 25th. By what calendar? Are we celebrating Christmas 10 days out (or 11 or 12 or 13 depending on when the calendar was adopted).

---

There seem to be several questions here.   One deals with how we note dates
> which fall within the 14 days or so which were lost during the change from
> Julian to Gregorian calenders.
>
> Unless a date is noted as Old Style or New Style, we have no way of knowing
> which it is written as, thus there is no need to change it.  I have yet to
> encounter a date mentioned as Old or New Style.
>
> My original question dealt with dates that are written so as to indicate one
> of two years, ie., 1703-4.  Since this date is before England and the
> English speaking world changed the date (1752), it indicates to me simply a
> confusion, that the writer didn't know in which year the event occurred.
>
> I am going to recommend that when we find a date such as 1703-4 we adopt the
> convention of writing the earlier date in the profile and make a note in the
> biography that it was originally written as (in this case) 1703-4.
>
> If we find a date with the Old Style/New Style notation, this should be
> mentioned in the biography.

---

I agree.  Record Georgian dates if possible and enter a comment in the notes.  Obviously during a period of transition, it may require more research than can easily be accomplished to know which system was used in a given date.  Uncertainty is inevitable in older genealogy and the notes are the place to discuss it.  I favor giving as much relevant information as possible marked as uncertain if that is appropriate.

---

I think we should have a site wide selected date method... people like me aren't smart enough to know there is a difference *doh!*

---

I updated the explanation on http://www.wikitree.com/wiki/Date_Fields#Julian_vs._Gregorian_Calendar to call for using the Gregorian date if at all possible, e.g. 1719 when you see the date written as 1718/1719. Let me know if that needs something more. Thanks!

---

To me, there is an issue of how the date is stored and how it's displayed. That is, I have dates like 1 1650, which I take to be March 1650 new style. (It's not clear to me if someone would have written 12 1649 on 24 March and 1 1650 on 25 March, but that is for a different thread!)

Optimally, there would be a page (or hideous popup dialogue) where one would input the date to see how it would look in other forms, e. g.,


Enter the date: _________________

Displ.

[].....This would look like 1650-03-11 if we projected the modern calendar back

[].....This would look like March 1 1649/50 in England at the time
                                                                                              
.......This might look like 12 1 1649 at the time if anyone understood that stupid system...grrr!


Uber-optimally, checking a Displ. box would make the dates show in that form


I don't think it is necessary to do much more than this. If there is some contingent that has a lot of date issues, maybe add them. I don't think there are too many people labouring under the Napoleanic calendar, so Julian would serve a great many needs.

The point would be to get people thinking about date conventions. Unless you are at a courthouse or graveyard or have an old bible with consistent date notations, there is not a lot of hope for accuracy. Seeing how a year or month can change dramatically, might help future transcriptions to be more thoughtful.

Refs:

http://www.fourmilab.ch/documents/calendar/
http://www.postgresql.org/docs/8.4/static/datatype-datetime.html (funny observation at the bottom).

Thanks for reading this!


bkm

---

Thinking about it a bit more, it would be a benefit to all if dates between 1582 and 1752 could be displayed in both systems (maybe as a dreaded popup). Some dates are too garbled to ever be known, but many errors could be prevented if it was obvious how the computer was storing the information.

This seems a neutral way to handle the Julian / Gregorian issue.