Jump to content


Photo

Double-dating, sorts right, won't calculate age


  • Please log in to reply
31 replies to this topic

#1 Stewartrb

Stewartrb

    Advanced Member

  • Members
  • PipPipPip
  • 35 posts

Posted 08 February 2015 - 10:19 AM

I've noticed the double-dates, while sorting correctly, don't provide an age calculation in the Age column.

                                               Age

Birth abt 1663.                         0

Death 15 Feb 1749-50.       (blank)

 

I should at least be able to see a rough age for this.



#2 Stewartrb

Stewartrb

    Advanced Member

  • Members
  • PipPipPip
  • 35 posts

Posted 08 February 2015 - 10:31 AM

I've noticed the double-dates, while sorting correctly, don't provide an age calculation in the Age column.

                                               Age

Birth abt 1663.                         0

Death 15 Feb 1749-50.       (blank)

 

I should at least be able to see a rough age for this.

I see now that I can enter dates in two styles.

 

I can have a 15 Feb 1749-50, which won't calculate an age.

 

And I can have a 15 Feb 1749/50 which *will* calculate an age.

 

I've seen double-dating transcribed both ways, using a dash as well as a slash.

 

It would seem the long "n" dash in Rootsmagic's case is to denote a date period.

 

While looking awful like a double-date, 15 Feb 1749-50 suggests a period between 15 Feb 1749 and (17?)50?

 

So... The "/" is for double dating and the "-" (long dash) is used to make a range.  (A range that looks too much like a double date if you ask me.)



#3 Stewartrb

Stewartrb

    Advanced Member

  • Members
  • PipPipPip
  • 35 posts

Posted 08 February 2015 - 10:39 AM

And, now my question, how do I find all dates that include this long "n" dash so I can verify that I intended that to be a date range or a double date?

 

Find Everything doesn't search through dates (and only finds the long dashes in notes and sources and such.)

 

Problem Search won't do it.

Search and Replace won't do it.

 

How do I find every instance of a date field using the long "n" dash in such a way that I can edit it then and there into a / if it's actually supposed to be a double date?

 

Thanks.



#4 kbens0n

kbens0n

    Advanced Member

  • Members
  • PipPipPip
  • 3459 posts

Posted 08 February 2015 - 10:49 AM

Reports->Lists->Fact List->Facts with text dates

---
--- "GENEALOGY, n. An account of one's descent from an ancestor who did not particularly care to trace his own." - Ambrose Bierce
--- "The trouble ain't what people don't know, it's what they know that ain't so." - Josh Billings
---Ô¿Ô---
K e V i N


#5 Stewartrb

Stewartrb

    Advanced Member

  • Members
  • PipPipPip
  • 35 posts

Posted 08 February 2015 - 11:51 AM

Reports->Lists->Fact List->Facts with text dates

Thanks for pointing that out.  It shows a lot of my older dates in an as of yet uncorrected form.

 

The problem with the long "n" dash dates is they aren't considered text.

 

They're legitimate (box is white, not yellow) dates.

 

The confusion between double (or dual) dating and time period is when something appears like "15 Feb 1749-50" which the software considers okay.  Does that really suggest to anyone "A period of time between 15 Feb 1749 and 50"?  Is that the year 50?  Or an assumed 1750?

 

Here's an article that says the hyphen in dual dating is best avoided so it's not mistaken as a time period.

 

I have, unfortunately, an unknown amount of these long "n" dash date ranges that I intended to be dual dates that I've entered over the last few months.  The hyphen was the style used for double dating in the older sources I used.  (NEHGR, Vol XXXIV, 1880, pg. 37, Births, Marriages, and Deaths in Lyme, Conn., is one example.)  

 

It seems an older way to indicate dual dating.  When Rootsmagic let me keep the same format as the source, and displayed it the same without error, I assumed it understood it to be a dual date.  (That's what I get for assuming, but it's something I need to fix too.)



#6 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6255 posts

Posted 08 February 2015 - 12:44 PM

I'm unsure that your example is a date range. The "-50" may be a date suffix which can be used in SortDate to order events having the same date. How that is interpreted in the Date field is unknown to me.

You could export your database to GEDCOM and use a text editor to find and replace these dates with the right format.

Alternatively, you could use SQLite to find and replace these dates in your database.

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#7 Stewartrb

Stewartrb

    Advanced Member

  • Members
  • PipPipPip
  • 35 posts

Posted 08 February 2015 - 03:01 PM

I'd rather not export to gedcom, fix the extraneous long "n" dashes with a utility like GedPad then reimport into RootsMagic.

 

I did download SQLiteSpy.  I have my database backed up, and loaded, I see in the EventTable numerous (64,884 returned) "Date" entries.

 

EventID EventType OwnerType OwnerID FamilyID PlaceID SiteID Date SortDate IsPrimary IsPrivate Proof Status EditDate Sentence Details Note 3971 2 0 2888 0 0 0 D.+18230918..+00000000.. 6656083871432441868 1 0 0 0 42009.6769834028     "He died in his 83 year."

 

I see there are "D" dates, "DA", "DB", "DR", and various "T" dates.  (That must mean something.)

 

(I note when using the / the sort date correctly uses the correct (later) year.  With the - the sort date is the entire date phrase.)

 

So I created two test events using a dual date of 12 Mar 1693/4 and also entering 12 Mar 1693-94.

 

This is how it looks in SQLiteSpy:

 

EventID EventType OwnerType OwnerID FamilyID PlaceID SiteID Date SortDate IsPrimary IsPrivate Proof Status EditDate Sentence Details Note 64975 1018 0 90 0 0 0 D-+16930312..+00940000.. 6582685966125760533 0 0 0 0 42043.6034904282   Test Description for event with dual date using - Test Note for event with dual date using -

 

EventID EventType OwnerType OwnerID FamilyID PlaceID SiteID Date SortDate IsPrimary IsPrivate Proof Status EditDate Sentence Details Note 64974 1018 0 90 0 0 0 D.+16930312/.+00000000.. 6583248922673676300 0 0 0 0 42043.6024762268   Test Description for event with dual date using / Test Note for event with dual date using /

 

It would appear I need to find the D- dates, and for those I intended to be dual dates, I'd have to change "D-" to "D." and rather than ending in ".." it needs to end in "/.".  While the slash date doesn't include the 94, which I also need to remove, it appears the "/." is code for the program to interpret the year 1693 as 1693/4.

 

I took a look and I only have about 30 instances of this kind of D- date.  So I could probably edit by hand.   (I can do that in SQLiteSpy, right?).

 

Is there a cheat sheet with how the dates are coded?  What the various D's do, the +'s and -'s, etc.?  I could probably guess my way through it, but if there's a write up somewhere?

 

Seems it wasn't as bad as I imagined.  And fixing it won't require too much work.



#8 kbens0n

kbens0n

    Advanced Member

  • Members
  • PipPipPip
  • 3459 posts

Posted 08 February 2015 - 03:21 PM

So I created two test events using a dual date of 12 Mar 1693/4 and also entering 12 Mar 1693-94.


Just as an aside, I believe it's more of a "standard" for that first date style to be represented as 12 Mar 1693/94 (two trailing digits after the slash, as is the typical convention with dashed "range" representations) instead of 12 Mar 1693/4.

---
--- "GENEALOGY, n. An account of one's descent from an ancestor who did not particularly care to trace his own." - Ambrose Bierce
--- "The trouble ain't what people don't know, it's what they know that ain't so." - Josh Billings
---Ô¿Ô---
K e V i N


#9 Stewartrb

Stewartrb

    Advanced Member

  • Members
  • PipPipPip
  • 35 posts

Posted 08 February 2015 - 04:25 PM

Just as an aside, I believe it's more of a "standard" for that first date style to be represented as 12 Mar 1693/94 (two trailing digits after the slash, as is the typical convention with dashed "range" representations) instead of 12 Mar 1693/4.

The Gedcom 5.5 (and 5.5.1) standard agrees with using two digits.  

 

It appears RootsMagic will only show one if it can.  But then it goes all out with 1699/1700 (rather than 1699/00).

 

And GedcomX appears to ignore the matter altogether.

 

The gedcom 5.5(.1) standard:

 

YEAR_GREG: = {Size=3:7} 
[ <NUMBER> | <NUMBER>/<DIGIT><DIGIT> ] 
The slash "/" <DIGIT><DIGIT> a year modifier which shows the possible date alternatives for pre-1752 date brought about by a changing the beginning of the year from MAR to JAN in the English calendar change of 1752, for example, 15 APR 1699/00. A (B.C.) appended to the <YEAR> indicates a date before the birth of Christ.



#10 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6255 posts

Posted 08 February 2015 - 04:34 PM

Have a look at this page which has info on the RootsMagic Date formats: http://sqlitetoolsfo...om/Date Decoder

 

With just some 30, it might be fastest to edit them manually with SQLite. The trick then is to find them in RootsMagic to confirm you have done so correctly. If "D-+" is distinctive, then it might be better to edit in RootsMagic by tracing back from a SQLite query filtered on that string. The SQLite View "EventWay" generated by script on the page http://sqlitetoolsfo...tsMagic screens could be so filtered. 


Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#11 Renee Zamora

Renee Zamora

    Advanced Member

  • Support
  • PipPipPip
  • 8474 posts

Posted 08 February 2015 - 07:50 PM

If you go to Search>Person List>Find and enter: Any Fact - date- contains - "–" (without quotes) it should find them all. Click Next to rotate through them. 


Renee
RootsMagic

#12 Stewartrb

Stewartrb

    Advanced Member

  • Members
  • PipPipPip
  • 35 posts

Posted 08 February 2015 - 09:32 PM

If you go to Search>Person List>Find and enter: Any Fact - date- contains - "–" (without quotes) it should find them all. Click Next to rotate through them. 

 

I was happy to learn about the database structure and date formats, which gave me a nice appreciation for the variety of dates I could enter.

 

Naturally, this in-program method for searching to fix was the simplest given I actually had very few I had to fix.  

 

Thanks.

 

P.S. Out of curiosity is there any thought to standardize to a double digit dual/double date format?



#13 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6255 posts

Posted 08 February 2015 - 10:46 PM

If you go to Search>Person List>Find and enter: Any Fact - date- contains - "–" (without quotes) it should find them all. Click Next to rotate through them.

How did we miss the obvious?

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#14 Don Newcomb

Don Newcomb

    Advanced Member

  • Members
  • PipPipPip
  • 1045 posts

Posted 10 February 2015 - 07:49 AM

As long as we're on the subject, does RM7 still insist that double dates prior to 1583 are incorrect? Many well respected genealogical publications use double dates prior to 1583 as a means of clarifying that Feb 14 would be in 1446 but is really in 1445. (e.g. 14 Feb 1445/6)   As far as I'm concerned, RM is just being pedantic in not allowing this. 



#15 Laura

Laura

    Advanced Member

  • Members
  • PipPipPip
  • 4276 posts

Posted 10 February 2015 - 10:41 AM

I am glad RM doesn't allow Double dates for dates before the Gregorian calendar even existed.

The publications may allow projecting the Gregorian calendar year backward, but it is confusing.

If I see a date 14 Feb 1445/6, I assume that the researcher didn't know for sure what year the person was born as I am expecting a date for the calendar in use at the time in that location.

#16 Don Newcomb

Don Newcomb

    Advanced Member

  • Members
  • PipPipPip
  • 1045 posts

Posted 10 February 2015 - 07:12 PM

If I see a date 14 Feb 1445/6, I assume that the researcher didn't know for sure what year the person was born as I am expecting a date for the calendar in use at the time in that location.

 

The problem is that some of the people who "don't know any better" are some of the most respected names in medieval genealogical research. 



#17 kbens0n

kbens0n

    Advanced Member

  • Members
  • PipPipPip
  • 3459 posts

Posted 10 February 2015 - 09:06 PM

The problem is that some of the people who "don't know any better" are some of the most respected names in medieval genealogical research.
[/quote

They're certainly welcome to create their own convention ...but I suspect that their respect is not derived from that method of dates (o;


---
--- "GENEALOGY, n. An account of one's descent from an ancestor who did not particularly care to trace his own." - Ambrose Bierce
--- "The trouble ain't what people don't know, it's what they know that ain't so." - Josh Billings
---Ô¿Ô---
K e V i N


#18 Don Newcomb

Don Newcomb

    Advanced Member

  • Members
  • PipPipPip
  • 1045 posts

Posted 11 February 2015 - 06:59 AM

 

 

They're certainly welcome to create their own convention ...but I suspect that their respect is not derived from that method of dates (o;

 

In this case, I'd like to ask if RM7 still believes that 25 Dec 1580 comes after 2 Feb 1580?   :unsure:



#19 kbens0n

kbens0n

    Advanced Member

  • Members
  • PipPipPip
  • 3459 posts

Posted 11 February 2015 - 04:18 PM

In this case, I'd like to ask if RM7 still believes that 25 Dec 1580 comes after 2 Feb 1580?   :unsure:


Paste them into the Sort Date: of two separate facts and you'll get your answer ;-)

---
--- "GENEALOGY, n. An account of one's descent from an ancestor who did not particularly care to trace his own." - Ambrose Bierce
--- "The trouble ain't what people don't know, it's what they know that ain't so." - Josh Billings
---Ô¿Ô---
K e V i N


#20 Don Newcomb

Don Newcomb

    Advanced Member

  • Members
  • PipPipPip
  • 1045 posts

Posted 11 February 2015 - 07:38 PM

Paste them into the Sort Date: of two separate facts and you'll get your answer ;-)

 

I would but I have not yet seen any reason to upgrade to RM7. I can tell you that in RM6, December 1580 comes after Feb 1580. OTOH,  Dec 1584 comes before Feb 1584/5. However, it comes after Feb 1584.  I asked the question to learn if RM had fixed that problem with v. 7.