FT Analyzer - could it possibly do...

[emjay]

Not sure this answers any of your issues but it has made me have a lightbulb moment.
I'm wondering if there is a way to mass load this copied data from Lost Cousins as "Lost Cousins facts" into a gedcom file, which you can then load into your software?

I'll raise this on the FTA website.

Obviously not an energy saving (fluorescent) bulb. They tend to warm up slowly rather than giving that "ZING" moment.

 

[Tim]

I've done some initial testing, and the theory is holding up so far.

 

[Tim]

Well some good news, I've managed to match 525 names from my 1,617 Lost Cousins data entries to my gedcom file.
Which I don't think is to bad. Remember though that I've used year of birth, which is calculated from the age at LC and I've used actual birth year from the gedcom. I can now alter the birth year (+/-1) and see how many more matches I get.
So now up to 788 matches.

And there's middle names

 

[webwiz]

Can you be more specific about how you have done this? And if it's all on Excel would you be prepared to share the spreadsheet?

 

[Tim]

Can you be more specific about how you have done this? And if it's all on Excel would you be prepared to share the spreadsheet?

Of course, which bit would you like me to be more specific on?

I could share the spreadsheet but its all my data, which isn't a great test because mines pretty much aligned.
However, I would need one to test on? If you PM'd me your gedcom and the LC data in a spreadsheet, I can send you some data back.

 

[webwiz]

OK stand by. I've got the two files ready but I can't see how to send a PM on this forum. I can start a conversation but can't see how to upload a file to it. I notice there is an upload button on this page.

 

[Alexander Bisset]

There must be many people in a similar position to me. I have over 4000 individuals in my FTM database and over 1000 entries on LC. But I have never recorded any census or LC facts in FTM. So my GEDCOM is useless for reporting which individuals are missing from LC. The work load in updating 4000 entries on FTM with these facts would probably be greater than comparing the FTM entries against the LC ones to find which are missing on LC. Can you see any solution? I can't, unless Peter gave you access to his database.

Even with access to the Lost Cousins database this wouldn't help at all as there is NO infomation stored on the Lost Cousins database that would link to someone's GEDCOM file. Think about it you would need to have specifically recorded something unique to a census record in your GEDCOM file that specifically tied that individual to that Lost Cousins record.

Now since you say you haven't recorded a census entry nor a Lost Cousins entry then the likelihood of you having a census reference stored in exactly the same format as on Lost Cousins is next to zero. The amount of work to identify and add records to a spreadsheet is going to be the same if not more work than simply adding a Lost Cousins fact. Then what do you gain you have a spreadsheet to maintain separately from your family history program that is just a bad idea as it becomes more and more work to keep up to date.

Tim has raised the issue on the FTA website and I've declined it as it is simply a bad idea. This is because there is NO unique individual ID in the data on Lost Cousins so you'd need to manually work out which is which and manually add the individual ID. Tim has suggested you could create a GEDCOM from this but I don't know what that solves as you'd then have to import it to your tree and hope that the hit or miss merge process worked and matched people.

Without a specific link between your data and the Lost Cousins data there is NOTHING I can do with FTA to make the job easier. However you can bite it off in small chunks and add the custom tags as you work through your tree and eventually you'll get there. It is in my opinion dramatically less work and FAR FAR safer to add a custom tag than to try to create a file from a spreadsheet play around with it import it into your tree and hope nothing goes wrong.

Trying the spreadsheet way is fraught with danger DO NOT ATTEMPT IT WITHOUT A BACKUP!!

 

[Alexander Bisset]

Well some good news, I've managed to match 525 names from my 1,617 Lost Cousins data entries to my gedcom file.
Which I don't think is to bad. Remember though that I've used year of birth, which is calculated from the age at LC and I've used actual birth year from the gedcom. I can now alter the birth year (+/-1) and see how many more matches I get.
So now up to 788 matches.

And there's middle names

That's the problem Tim it's guesswork not all of those 788 matches are going to match the right person. In my mind it is far safer to manually add a tag to the right person than to try automating it and never again being sure if its right or not.

 

[webwiz]

yes it's obvious when you think about it. nothing for it but to do the hard slog. although my plan was not quite as daft as you suggest, I would have exported the entire file to a gedcom, updated the gedcom with LC flags (if that had been feasible) and from it built a new and different FTM file, so no risk to the original one.

 

[Tim]

Guesswork? I've checked the file. I don't have anyone with names that match (including middle name) that were born in the same year. But no matter.
Webwiz, that was exactly the plan.

 

[Alexander Bisset]

Yes but the method when applied to others files is not going to provide a reliable match without considerable technical skill with spreadsheets and a large effort in creating and manually checking a spreadsheet. That's fine for people who are technically competent with Excel but not ever something I want to venture near supporting for novice Excel users. There is a simple reliable foolproof method at present that involves tagging individuals.

Mass tagging of individuals is fine but without manually checking each one it could produce random matches for the unwary. I have no doubt at all that with your Excel skills you have got something that works well for you but I cannot see how FTAnalzyer could possibly improve on your spreadsheet. This process is about getting tags into your Family Tree program some how. A merge into a new FTM file is still a separate file to maintain which was my original concern.

At the end of the day the end result is the same your family tree program needs tags to indicate who has a Lost Cousins record. How the user gets that data into their particular family tree program is not something FTAnalyzer can do as there are so many different family tree programs out there.