Family Historian User Group

Since upgrading to V6 I've had a problem with trees exported to Ancestry.

I've used the Export Gedcom File 1.7 plugin, tried both (Std) and (FTM) export modes, with (ANSI) encoding and (NONE) conversion options. I then run the resulting Gedcom through ResPrivata to remove some details of living people (I've used this for years).

That final Gedcom imports into Ancestry ok, but the resulting tree has 4 digit numbers appended to all of the place names like this example (scroll the picture down):
The numbers seem to bear no relation to the individual or place name entries in the Gedcom:

Ellen Deamer:
0 @I469@ INDI
1 NAME Ellen /Deamer/
2 SOUR @S303@
2 SOUR @S163@
2 SOUR @S108@
2 SOUR @S165@
1 SEX F
1 BIRT
2 DATE ABT 1844
2 PLAC Little Wymondley, Hertfordshire, England
3 SOUR @S2114@
2 SOUR @S303@
2 SOUR @S165@
1 CHR
2 DATE 11 FEB 1844
2 PLAC Little Wymondley, Hertfordshire, England
3 SOUR @S2114@
2 SOUR @S303@
1 CENS
2 DATE 30 MAR 1851
2 PLAC Hitchin (reg), Hertfordshire, England
3 SOUR @S2112@
2 ADDR Village, Little Wymondley, Hertfordshire
2 AGE 7y
2 SOUR @S250@
1 OCCU Scholar
2 DATE 30 MAR 1851
2 PLAC Hitchin (reg), Hertfordshire, England
3 SOUR @S2112@
2 SOUR @S250@
1 CENS
2 DATE 7 APR 1861
2 PLAC Hitchin (reg), Hertfordshire, England
3 SOUR @S2112@
2 ADDR Private House, Little Wymondley, Hertfordshire
2 AGE 17y
2 SOUR @S249@
1 OCCU Straw Plaiter
2 DATE 7 APR 1861
2 PLAC Hitchin (reg), Hertfordshire, England
3 SOUR @S2112@
2 SOUR @S249@
1 CENS
2 DATE 2 APR 1871
2 PLAC Hitchin (reg), Hertfordshire, England
3 SOUR @S2112@
2 ADDR Cottage - Titmore Green, Little Wymondley, Hertfordshire
2 AGE 27y
2 SOUR @S240@
1 CENS
2 DATE 3 APR 1881
2 PLAC Hitchin (reg), Hertfordshire, England
3 SOUR @S2112@
2 ADDR Little Wymondley, Hertfordshire, England
2 AGE 37y
2 SOUR @S238@
1 DEAT
2 DATE 13 NOV 1883
2 PLAC Little Wymondley, Hertfordshire, England
3 SOUR @S2114@
2 SOUR @S165@
1 FAMC @F202@
1 FAMS @F102@
1 NOTE Births x2 - Female Dearmer 1844 Q4, Hitchin v6 p315 and p316
2 CONT IGIe - christening for Helen Deamer 11/2/1844, Little Wymondley, parents David and Mary.
1 EVEN
2 TYPE Chris Index
1 EVEN
2 TYPE Mar Cert
1 EVEN
2 TYPE 1851 UK Census
1 EVEN
2 TYPE 1861 UK Census
1 EVEN
2 TYPE 1871 UK Census
1 EVEN
2 TYPE 1881 UK Census
1 EVEN
2 TYPE Death Cert
1 CHAN
2 DATE 15 DEC 2014
3 TIME 17:01:00

Little Wymondley:
0 @S2114@ SOUR
1 TITL Little Wymondley, Hertfordshire, England
1 AUTH Place Rec Id [6]
1 TEXT Map Plot:
2 CONT Latitude: N51.93268
2 CONT Longitude: W0.23552
1 NOTE Source Type: Place Details
1 CHAN
2 DATE 20 DEC 2014
3 TIME 13:37:30

Hitchin (reg)
0 @S2112@ SOUR
1 TITL Hitchin (reg), Hertfordshire, England
1 AUTH Place Rec Id [4]
1 TEXT Map Plot:
2 CONT Latitude: N51.9485178
2 CONT Longitude: W0.2674978
1 NOTE Source Type: Place Details
1 CHAN
2 DATE 15 DEC 2014
3 TIME 17:29:17

How are others exporting their Gedcom's to Ancestry, and has anyone seen this problem?

It looks like Ancestry is mishandling the Source Citation attached to every Place field with FH V6.
1 BIRT
2 DATE ABT 1844
2 PLAC Little Wymondley, Hertfordshire, England
3 SOUR @S2114@
2 SOUR @S303@
2 SOUR @S165@
The 3 SOUR @S2114@ is NOT a Source Citation for the BIRTh Fact, but a Source Citation for the PLACe field that links to a Source Record holding the details derived from the removed non-standard FH V6 Place Record.

If you don't need those details in Ancestry, then in the Plugin on the Extra Options tab change the Place Record 0 @P%d+@ option to Remove entirely (Min).

The way to preserve that setting for future use is to use the (CEA) Custom Export Alpha mode and configure all the options to match what Ancestry accepts.

If you find any other problems with Ancestry then let me know and I can easily add a tailored Ancestry export mode to the Plugin.

Mike,

Thanks, changing the Place Record 0 @P%d+@ option to Remove entirely (Min) has resolved that problem.

There's another (minor) problem, that was there before, in that all Census entries have an extra Source, "Unsourced Citations" which don't exist in my FH tree (scroll down the screen grab to the 1851 census entry) :
Clicking on the "Unsourced Citations" against the 1851 Census displays:
implying it's related to the age.

The Properties-Facts tabs display shows:
i.e. only one source attached to the 1851 Census.

The GEDCOM text for the related individual, and their 1851 Census is (after processing via the Export plugin and ResPrivata):

"1 NAME Ellen /Deamer/
2 SOUR @S303@
2 SOUR @S163@
2 SOUR @S108@
2 SOUR @S165@
1 SEX F
1 BIRT
2 DATE ABT 1844
2 PLAC Little Wymondley, Hertfordshire, England
2 SOUR @S303@
2 SOUR @S165@
1 CHR
2 DATE 11 FEB 1844
2 PLAC Little Wymondley, Hertfordshire, England
2 SOUR @S303@
1 CENS
2 DATE 30 MAR 1851
2 PLAC Hitchin (reg), Hertfordshire, England
2 ADDR Village, Little Wymondley, Hertfordshire
2 AGE 7y
2 SOUR @S250@
1 OCCU Scholar
2 DATE 30 MAR 1851
2 PLAC Hitchin (reg), Hertfordshire, England
2 SOUR @S250@"

Any thoughts on this little glitch? From my point of view at the moment it's just a cosmetic thing, but might have unrealised implications later on.

Have you noticed that the Occupation fact following the Census also says Age 7 even though there is NO AGE tag in your GEDCOM extract for OCCU.

Looks like Ancestry is not handling the AGE tag on the CENS correctly. Perhaps you should report that to Ancestry.

Try just removing the AGE tag line from CENS in the GEDCOM and see if the problem goes away.

Mike, yes you're right. I removed the "2 AGE 7y" tag from the 1851 CENS entry for that individual and uploaded the GEDCOM to Ancestry again, and it now looks ok, with no "Unsourced Citations" shown against the census.

The age is still shown underneath the 1851 Census and 1851 Occupation so I'm guessing that Ancestry is calculating that itself.

Is the age in the FH GEDCOM a calculated age or taken from the actual age transcribed from the census via AS? The age given on the census may be wrong of course, so people searching for the census entry based on a calculated age may not find the source.

I'll report this to Ancestry, but won't be holding my breath for any resolution!

In the meantime is there a config option in the Export Plugin to exclude the Census age, as a temporary (ho ho) workaround?

The AGE tag only applies to explicit entries in the Age field of each Fact (next to the Date field) near the bottom of the Facts tab.

AGE tags are such a fundamental feature of the GEDCOM Specification that I never considered needing to remove them, although it could be done quite easily with a Plugin update.

Can you please investigate whether that AGE tag problem is widespread or only applicable to CENSus events, or maybe only when there is a Source Citation, or perhaps both CENSus and Citation. In other words, look for explicit Age field/AGE tag entries in other Facts and see how Ancestry handles them. That is the kind of software forensics I often have to do to get Plugins to offer the best solutions.

OK Mike, I'll put together a smaller "test" tree and see what Ancestry does with it. Might take a few days, but I'll post back with results.

Yes, the old problem that not all programs interpret GEDCOM correctly and/or in the same way. You also need a program that can report your data how you want it. Ancestry trees are notorious for missing data or "hiding" it or in this case getting it wrong. Hence important to design your data structure accordingly, or at least know how to amend your Gedcom file.

See Best format of PLACe and ADDRess for Map Geocoding (11937) post 03 Jan 2015 02:27 for examples where even FH does not include all information in Reports (although you can display everything if only by the ALL tab).