Dual dates not standardizing correctly
If I enter 1603/04 as a date (under Vitals events) this standardizes as 1604. Likewise if I enter 1 February 1603/04 (or any other day in February). However, entering "February 1603/04" standardizes the date as 1603. Why the inconsistency, other than a programming fault?
Comments
-
As stated in the thread of another post I instigated, these reports are no longer being acknowledged in the way they were in the old GetSat forum.
Is it now best to ignore the advice (given upon the demise of GetSat) to use "Ideas" to report such problems? FamilySearch employees' activity here is now practically non-existent, so I wonder if these matters should now be raised as Support cases instead.
1 -
Paul,
"I wonder if these matters should now be raised as Support cases instead"
I certainly hope not. There are many distinct advantages to posting a potential issue in a forum format. But there most definitely seems to have been an abandoning of this forum by both patrons as well as FS employees.
Regarding your original question, I can't remember when the Julian/Gregorian calendar change over occurred. But it doesn't seem that it would behave quite like you have observed.
I wonder how the "/" is being interpreted when you enter it. Is it seen as an either/or type thing, a from/to, or a year/month? Furthermore, if you add the month, that meaning might change.
Just running a quick test on this, when you enter 1603/04, note that there are two suggested standard values -- 1604 and April 1603 (with 1604 being the default assigned). Prefixing it with 1 February eliminates the possibility of the 04 being the month so you end up with a single default standard of 1 February 1604. And yes, removing the 1 prefix ends up giving the single standard of February 1603.
So my question is, when you enter 1603/04 what is the "04" for? Is it an alternate year or is it a month? The two suggested standards provided cover both of these possibilities. When you enter February 1603/04, since you did not enter a day, it looks like it is considering the "04" as some kind of day indicator. And when you actually record a specific day 1 February 1603/04 it is just assuming the "04" is some kind of alternate year (e.g., years 1603 or 1604) or extra non-standard text.
It seems to be attempting to get a correct match but there seems to be a lot of variables in the mix.
0 -
Thank you for your comments, Jeff. Having never been involved in computer programming, I hadn't really considered how the "variables" might be read by the set program. To me, it was just a plain anomaly that the input of just the month and year produced a different default standardized date than if the full date or just the dual-dated year is entered.
I would emphasise I am not being "picky" here - the parish registers I am taking this detail from often show the dates very unclearly - the day in the month being written in Roman numerals. For example, "xviij Februaris 1603". At best, it is often unclear whether the date is actually recorded as xviij or xvij - hence the reason for my just inputting month and year on many occasions. Unfortunately, there is no option in these cases to apply consistency, as I cannot enter a display date of February 1603/04 and be able to standardize as February 1604. February 1603, as you appear to have confirmed, is my only option.
I would hope this could be fixed but realise (from the lack of participation and "thumbs up" to my comments in this thread) this is an issue that does not widely affect other users, so would probably be considered of low priority by FamilySearch.
0 -
Yes, didn't we have issues with this originally? The dual date was only recognized if the repeat after the slash was the 4 digit year? i.e. it had to be entered as 1603/1604?
Maybe not everything was catered for.
As Jeff says, there are a lot of possibilities if you start thinking (a bad habit?). It is, for instance, far from clear to me that 1603/04 should be interpreted as a Dual Date - dual dates can only apply to Jan thru Mar, yet there is no month in 1603/1604, so how do you, the inputter, know it's a dual possibility? Well, maybe it's between 31 Dec 1603 and 1 March 1603 but has no (legible) month so could be Jan 1603 or Feb 1603... Oh dear....
0 -
When I saw this at first I thought 1603/04 meant April 1603.
0 -
After reading the comments here, I think I should record the format as 1603/1604 (rather than 1603/04) in future, in case the latter causes confusion to other users.
However, inputting February 1603/1604 makes no difference to the problem with standardizing: February 1603 is the only option in the drop-down, whereas if I input "1603/04" I have the choice of 1603 OR 1604 and any full date (1-28 February) only gives an option with the year standardizing as 1604!
0 -
Trouble is, Paul, if I and many others I'm sure, see "February 1603/1604", our assumption will be that you mean 1603 or 1604 and you don't know which. Which isn't the same thing as what you intend.
It kinda makes me wonder how many groups (countries?) actually understand the 1603/04 Dual Dating convention. Not something I'd thought about.
One other possibility - which I haven't tried - is to write "Feb 1603 OS", where "OS" stands for "Old Style" (i.e. New Year starts 25 March). However, I doubt this is going to be any better as OS and NS are English acronyms, and therefore not unversally understood.
0 -
Yea, it's important to realize that it wasn't just Julian vs. Gregorian calendars. There are a handful of other calendar types being used in different countries even today. Also the changeover between the Julian and Gregorian calendars was gradual starting around 1582 and ending around 1752 depending on the country you were in. And there are other little "gotchas" when it comes to dating things. For example, an act of Parliament made the dates of 3 September 1752 to 13 September 1752 nonexistent!
So our standards folks have their work cut out for them trying to work the details of alternate calendars that can exist.
0 -
I don't want to continue to make a big deal out of the issue, but my original post still stands. No matter what the ambiguity / confusion this might be causing users unfamiliar with this kind of dating system, it's the fact that in one situation (say February 1603/1604) the date is standardized as 1603, whereas in the other two (say 1 February 1603/1604 or just plain 1603/1604) the input is standardized as 1604. So, I was purely requesting consistency here - preferably, that all three can be standardized as 1604 in this example.
Incidentally, I was surprised when I first came across original records recorded in this way (with dual dating). Obviously the clerk knew that the general population (in England) accepted 1st January as being the first day of the year (in secular terms, if you like) and was acknowledging 25th March only applied to the church calendar.
Incidentally, I believe the instruction on most indexing projects must be to index the date as the "later" year (1604 in this case), since many computer programs do not allow for dual dates (only the ??/??/???? format). I have found this to be the general situation in Find My Past. However, when a register has been indexed twice, it is common to find one version with (say) 01 February 1603 and the other with 01 February 1604 as the event date.
Certainly, when coming across these events, the original records need to be consulted to confirm the actual position regarding these 1 January - 24 March events (usually pre 1753, in England).
0 -
Yes, it took me a while to realise that dual dating was a thing employed then, pre-1752, and not a concoction created in the 20th century. Just not always employed.
As you say, there must have been some sort of change in public view that led people to begin to view 1 January as the start of the New Year in England. Indeed, Scotland moved to 1 January as the start of the New Year in 1600, so there must have been scope for confusion when the families living in the Borders came down to Carlisle to do their shopping... :-)
0
