FTAnalyzer questions.

[Bryman]

You need to contact Genome.
Genopro have not updated their software for years. Genome has been updating the skins to fix any reporting errors like this..

Reply from Genome:

I have compared the two versions and cannot find any code present in 2013.05.05 that is missing from 2014.04.05
There is a facility introduced in 2013.07.30 that allows additional Gedcom event and attribute tags to be output from specially formatted Occupation entries.
So for example if you create an Occupation entry for an individual with a job title of |EVEN|Lost Cousins and start and end dates of say 1961 then the output from the Export to Gedcom skin will indeed have

1 EVEN
2 TYPE Lost Cousins
2 DATE 1961
as you require.
which is the second of your suggestions, Tim, but not used by me as it felt like a fudge (not really anything to do with Occupation).
I now have a Lost Cousins tag on the GenoPro Properties menu, as suggested nearly a year ago and working well before this skin replacement.
I am confused. Do I have a non-standard version of the 2013.05.05 skin? I have not applied modifications to any code.

 

[Tim]

Hi Bryman,

Is it worth sending Genome your copy of the 2013.05.05 skin? He can check to see if it is different?

Perhaps you used the 1st suggestion, which was to add the code to the skin? Does your properties look like this? Maybe you can post yours?

 

[Bryman]

Hi Bryman,

Is it worth sending Genome your copy of the 2013.05.05 skin? He can check to see if it is different?

Perhaps you used the 1st suggestion, which was to add the code to the skin? Does your properties look like this? Maybe you can post yours?

I haven't added/changed any code - and don't even know what language is used.
However, my properties look exactly like your attachment, word perfect.

 

[Tim]

Ok, well it sounds like you implemented what I did wrt to GenoPro.

Did I send you my modified skin?

 

[Bryman]

Ok, well it sounds like you implemented what I did wrt to GenoPro.

Did I send you my modified skin?

I don't know. How do I tell? What modification did you make?

I have looked in my skins folder and found a version.txt file which just contains the text "2013.05.05".
However, there are two .js files with names "Gedcom.js" dated 13/05/2013 and "Gedcom orig ver.js" dated 06/05/2013 but I don't have anything to let me look inside.

Was Genome the originator of those changes or were they a Tim special?

 

[Alexander Bisset]

The .js files are probably Javascript in which case ANY text editor eg: Notepad will be able to open them.

 

[Bryman]

The .js files are probably Javascript in which case ANY text editor eg: Notepad will be able to open them.

Thank you Alexander. I tried to use the default program, Windows Based Script Host, and it bombed.
Using Notepad, I found the added 17 lines of code (without indent) but cannot follow the overall logic as there are no comments.
Does this seem familiar, Tim? Are there any other modifications?
Does Genome know about this code and would he be willing to incorporate it in the release version?

I am not happy with having to make my own changes to every version released and would much prefer a formal 'user exit' module approach or included text if it is not agreed for general release.

 

[Tim]

Thank you Alexander. I tried to use the default program, Windows Based Script Host, and it bombed.
Using Notepad, I found the added 17 lines of code (without indent) but cannot follow the overall logic as there are no comments.
Does this seem familiar, Tim? Are there any other modifications?
Does Genome know about this code and would he be willing to incorporate it in the release version?

I am not happy with having to make my own changes to every version released and would much prefer a formal 'user exit' module approach or included text if it is not agreed for general release.

Reporting Lost Cousins as facts is non standard and therefore I had to add the code to get the Lost Cousins extracted to the gedcom.

I don't think he will add it to the standard release, I think that's why option 2 came about. The one that includes this |EVEN|Lost Cousins format

 

[AndyMick]

Alexander: I was looking at the Data Errors tab on my latest file, and went to correct the error FTA reported (born before mother was 13) and found that the birth shows correctly in FTM2011 (file created from the Gedcom I was sent) in that the child was born to wife #1, but you refer to the name and age of wife #2, step mother to said child. Is this a feature?! Would you be wanting the whole Gedcom file or extracts?

 

[Alexander Bisset]

An export of just that two families should suffice. I suspect the problem is the way that the children are recorded against the two families. It is likely that the "step" relationship isn't being recognised, thus it suspects it is their mother rather than step mother.

The code only gives the birth fact as an error if the mother or father is a natural parent. The check being (Natural, Unknown or Private). If the relationship is step then it is expecting the GEDCOM to have a relationship flag to indicate that if that's missing it would be why there is an error highlighted.

The basic principal is that if you are recording a child against a parent that isn't a natural parent if you haven't recorded the nature of the relationship then it will assume its a natural parent. If you have recorded a step relationship it may be that the method of recording differs from what I was expecting.

 

[AndyMick]

I've extracted the Gedcom and I'm struggling to comprehend it - so no wonder FTA doesn't report what I expected. Alvin (not a chipmonk I hope) has 2 FAMC entries - for both his father's wives. There's no mention of any step-relationship. I don't know whether the GEDCOM has come from FTM or from an Ancestry tree. What also surprised me was the Ancestry tree showed the spouses in the wrong order - so I suspect a gigantic mess up somewhere, probably Ancestry. I know when I tried to create a copy of some folks on this tree on Ancestry, things really messed up.

This can certainly be classed as a GEDCOM issue, not FTA. Sorry to have wasted your time, Alexander.

 

[Bryman]

Alexander, I noticed that you had updated FTA to level 3.6.0.0 so downloaded it to have a look at the new Referral Report.
First impression is that it looks very good. Thank you.
However, my report appears with no values in the Census Reference column. Why is that? Have I forgotten to initialise something? I cannot see anywhere that I need to request that information.

 

[Bryman]

I have not been through the full referral process, ticking all the appropriate relatives for a potential new member, so thought that I would try a dummy run using the new FTA Referral Report as a guide. I did not have a non-member readily available so decided to simulate a referral for a 4th cousin who I have been matched with recently. Although he is already a LC member, but not a forum member, the match was only for two relatives and the FTA report showed more than 50 potential matches! Perhaps there is some way that my cousin can still benefit from my information even though he is already registered with LC.

However, SEVERAL hours later, I have a few comments/suggestions regarding this process. How did anyone manage to do it before this report was available? Even now, it is not easy and there are a few changes which I think could help considerably.

1. As I mention in a previous post, I had no Census References and they would help to confirm that the right relative is selected. Multiple occurrences of similarly named individuals can lead to considerable confusion. To be really helpful, the Census Reference also needs to be displayed in the selection list displayed from LC. (Request to Peter for simple enhancement, please.)

2. The report produced from FTA did not show individuals in the same sequence as the list shown by LC. This really was the main reason for having the FTA report. In fairness to Alexander, I did not have all of the anticipated records in the gedcom file due to omissions in records exported from my FHS so don't know whether that might have made a difference. I have reported the bug and await a reply/fix. All individuals had both RESI and CENS data in the FHS but did not show that way in the gedcom. Sometimes RESI, sometimes CENS, sometimes neither, but never both!

3. The report from FTA shows the Age of each individual (as is reported in the census) but LC shows Birth Year (calculated from that age). It would help to make life easier if both lists showed the same information.

4. Some individuals in the FTA report are shown with their maiden surname although they were married at the time of the census. This makes it harder to match records for the selection process. It would help greatly if FTA could anticipate the change of name and use the married name appropriate to the time of census, or show both.

 

[Alexander Bisset]

Alexander, I noticed that you had updated FTA to level 3.6.0.0 so downloaded it to have a look at the new Referral Report.
First impression is that it looks very good. Thank you.
However, my report appears with no values in the Census Reference column. Why is that? Have I forgotten to initialise something? I cannot see anywhere that I need to request that information.

There are limited formats that can be understood by the census references. How are you recording the census references in your file? An example GEDCOM with a census reference that isn't showing up would help.

 

[Alexander Bisset]

You could export the report to excel and email it to your 4th cousin. That's essentially what the limit to matches checkbox is for to restrict the results to only the matches so you can send just the matches to someone.

Matching the sequence is a bit of a nightmare as it sort of relies on having census references. The ONLY thing the website has is the census reference and not the place name so understandably it sorts by that, without census references in the GEDCOM the only thing FTAnalyzer has is the census place. So unless it can get the census ref from the GEDCOM it's going to struggle to match the format exactly. It should hopefully be at least in household blocks as per website.

Changing age to birth year is a good idea.

Showing surname at date is also a good idea I thought I'd already done that but clearly not.

 

[Bryman]

There are limited formats that can be understood by the census references. How are you recording the census references in your file? An example GEDCOM with a census reference that isn't showing up would help.

What do you define as Census Reference? I tend to use either "HO107/129/2/8/11" or "1841 UK Census, Piece 129, Book 2, Folio 8, Page 11" depending on where it is being used. I assume that the former is more similar to that used/displayed by LC but should the prefix be included or will FTA strip that off?

However, I don't see anything like that in the gedcom file so perhaps I have not specified this information in the right field in my FHS.
What record and format does FTA expect to see in the gedcom file? Knowing that, I can then experiment to find where I must place the information to get it exported correctly.

 

[Tim]

Hi Bryman,

The format should look like this in the gedcom

1 RESI Age: 11 Relation to Head of House: Daughter
2 DATE 1901
3 PAGE Class: RG13; Piece: 3507; Folio: 141; Page: 47

So it's probably the PAGE Class that you are missing.

Changing age to birth year is a good idea.

Why can't you have both? Age at the time of the census is recorded in FTM gedcoms at least.

1. As I mention in a previous post, I had no Census References and they would help to confirm that the right relative is selected. Multiple occurrences of similarly named individuals can lead to considerable confusion. To be really helpful, the Census Reference also needs to be displayed in the selection list displayed from LC. (Request to Peter for simple enhancement, please.)

I have also requested to Peter and Alexander that they add the LC census ref to the referrals page and to the FTA Referrals report in this format 3507/141/47 (depending on census year) respectively.

 

[Bryman]

The format should look like this in the gedcom

1 RESI Age: 11 Relation to Head of House: Daughter
2 DATE 1901
3 PAGE Class: RG13; Piece: 3507; Folio: 141; Page: 47

So it's probably the PAGE Class that you are missing.

Not only but also! My gedcom records look more like . . .

1 RESI
2 TYPE Primary Residence
2 DATE 02 APR 1911
2 PLAC street, town, county, country
2 SOUR @source00123@

1 CENS
2 DATE 02 APR 1911
2 PLAC street, town, county, country
2 SOUR @source00123@
3 PAGE RG14/12345/123
3 NOTE 1911 census - Piece 12345, SN123 - living at street, town, county

0 @source00123@ SOUR
1 TITL RG14/12345/123 @@ 1911 UK Census
1 PUBL Publication Date: 02 APR 1911
2 CONT Media: Census
1 NOTE 1911 census - Piece 12345, SN123 - living at street, town, county

It seems that the information is there in the FHS but just not always output to the gedcom, or at least not in the required format.
Initially, I used GR so added census information in the form of notes.
Those notes were then copied into Genopro as comments but still free form.
Then the information was copied from those comments into appropriate fields as it seemed appropriate.

I appreciate Alexander's difficulty when users are left to their own devices. There is a reason to have standards. It is a pity when such standards are not made clear to all users.

 

[Bryman]

I have just had a reply from GenoPro saying that either a RESI record or a CENS record is output (not both), depending on whether there is a corresponding Source which is marked as a Census. Gedcom records are not produced just because there is a Source entry. However, it might be that the matching of Source identifier stops at the first blank in the field, which suggests that the format needs to be "RG14/12345/123" rather than "Class: RG14; Piece: 12345; SN: 123".

 

[Bryman]

On further investigation, I have found that the data output on the PAGE record in the gedcom file comes from yet another field within my FHS.
That field is free format and I have entered data in the form . . .

Piece aaa, Book bbb, Folio ccc, Page ddd etc.

Is this incompatible with FTA and do I need to change to an alternative or can FTA accommodate?