• Only registered members can see all the forums - if you've received an invitation to join (it'll be on your My Summary page) please register NOW!

  • If you're looking for the LostCousins site please click the logo in the top left corner - these forums are for existing LostCousins members only.
  • This is the LostCousins Forum. If you were looking for the LostCousins website simply click the logo at the top left.
  • It's easier than ever before to check your entries from the 1881 Census - more details here

Publicising Lost Cousins

Alexander Bisset said:
This is why I thought a specific FTAnalyzer field would be useful then on the ancestors page you could highlight them in red or something to show they haven't been checked. Then once user had confirmed an entry they would be ok. It would give an extra level of protection to the core data.
Click to expand...
All entries should be checked, not just those entered through FTA. In fact, I suspect that there will be fewer entries that need altering when FTA is used than when users enter the information manually!
 
Ah then perhaps a flag on all entries that shows whether the user has clicked to check them or not so that it's something that applies to all entries added.

I'm running a poll at present on the website to ask if people have heard of Lost Cousins and if they have joined or not. Only 27 replies so far so not representative but 16 Lost Cousins members 8 haven't heard of it before 2 members who have not entered anyone and one just joined as a result of reading about it in FTAnalyzer.

Oh and beta 7 available.
 
Last edited:
BTW with the data entered so far has there been any issues noted? I noticed a problem with missing Schedules in 1911 census and that affected my test user and Tim (who is now working through them to fix the Missing Schedule numbers). Beta 2 had a value of Missing for Schedule number and I'd not checked to prevent that being used and unfortunately the website accepted it. However Tim noticed and we caught and squashed that bug. Hence the reason for limited beta testing before live.
 
It's uploading my results now - it seems to be struggling with Canadian 1881 and US1940 entries probably because of how Ancestry lists the appropriate information:
It seems for the 1940 US census, the Roll either reads as "T000_0000" or "m-t0000-00000", and the ED is processed as "00-00" depending on the number. (not a single US1940 census entry was uploaded to LC)
For the Canadian 1881 census, it failed every entry where the Roll started with "C_..."
 
How does your data show in the census report does it record it as a good census entry?

When you say it fails is it saying in the text box it fails (good) or is it uploading the wrong data to Lost Cousins (bad).

I strongly suspect that I will need to add extra census recognition for non UK census entries.
 
Alexander Bisset said:
How does your data show in the census report does it record it as a good census entry?

When you say it fails is it saying in the text box it fails (good) or is it uploading the wrong data to Lost Cousins (bad).

I strongly suspect that I will need to add extra census recognition for non UK census entries.
Click to expand...

It records it as good data, available for uploading and then fails in the text box, it doesn’t upload the information to Lost Cousins.

I did noticed that some of the entries are duplicated on LostCousins that FTAnalyzer added, mostly where the individual had an adjusted birth date presumably because I’ve found them on the 1939 register.
 
jorghes said:
It records it as good data, available for uploading and then fails in the text box, it doesn’t upload the information to Lost Cousins.

I did noticed that some of the entries are duplicated on LostCousins that FTAnalyzer added, mostly where the individual had an adjusted birth date presumably because I’ve found them on the 1939 register.
Click to expand...
Ok that’s good that they didn’t upload. The duplicates do you mean they had previously been manually entered by you? The difference being the age? I’m assuming that you hadn’t tagged those people with a custom fact as having a Lost Cousins entry? Thus as far as the program was concerned they hadn’t been entered?

My understanding is that the Lost Cousins website doesn’t allow you to enter two people with identical names and census references so I’m assuming the name is slightly different or the age is different?

It may be an idea for me to add a checker to read the my ancestors page and try to work out if the person is already there. Some form of checklist against your existing entries.
 
jorghes said:
It's uploading my results now - it seems to be struggling with Canadian 1881 and US1940 entries probably because of how Ancestry lists the appropriate information:
It seems for the 1940 US census, the Roll either reads as "T000_0000" or "m-t0000-00000", and the ED is processed as "00-00" depending on the number. (not a single US1940 census entry was uploaded to LC)
For the Canadian 1881 census, it failed every entry where the Roll started with "C_..."
Click to expand...
Could you give me some example full census references that have been recognised but aren’t uploading. I’m thinking it’s the ones with T627 and m-t0627 not sure about Canadian ones though.

I suspect it’s recognising it as a valid census pattern but submitting the t627 bit which Peter mentions should be stripped. So I’ll need to check how it’s recognised and tweak the recognition routine. So examples will help.
 
Alexander Bisset said:
Could you give me some example full census references that have been recognised but aren’t uploading. I’m thinking it’s the ones with T627 and m-t0627 not sure about Canadian ones though.

I suspect it’s recognising it as a valid census pattern but submitting the t627 bit which Peter mentions should be stripped. So I’ll need to check how it’s recognised and tweak the recognition routine. So examples will help.
Click to expand...

Sure:
US1940: Roll T627_1141; ED 8-14; Page 9A
Or: Roll: m-t0627_2227; ED 1-7; Page 19A

Canada 1881:
Roll: C_13266; Page 67; Family 301

As for the duplication, I noticed in two instances - once where the name on LostCousins and the name on my GEDcom where different spelling - all bar one seemed to pull a LostCousins "check this entry" error:
Evelyn v Eveline - and thus while the years were the same, there were two entries.
The other instances were mostly where a full birthdate had been added, thus changing the year of birth - in one instance the census said "1901" but the correction read "1900".
I may not have had a LC reference fact for them in my GEDcom, but that could have been for multiple reasons, such as omitting that person as I thought that they were not part of the family (which has happened several times) and later discovering that they married in.
 
Ok, interesting I'd only ever seen the 1940 as Roll; Page; ED before. I'll add a recognition to that format. I noted I'd left the T627_'s on so that will be the reason they failed.

I don't understand the Canadian Reference though it needs a district, sub-district, page and family and that doesn't appear to have those 4 bits of info.
 
Alexander Bisset said:
Ok, interesting I'd only ever seen the 1940 as Roll; Page; ED before. I'll add a recognition to that format. I noted I'd left the T627_'s on so that will be the reason they failed.

I don't understand the Canadian Reference though it needs a district, sub-district, page and family and that doesn't appear to have those 4 bits of info.
Click to expand...

I'll try again for the US1940 and Canadian 1880 once I have finished adding the LC facts to my GEDcom - I don't want to find out if it will duplicate the 1600 odd entries it attempted to add today - the majority of which were entered.

The Canada reference was taken straight from FTAnalyzer - so perhaps what it collects as the census reference doesn't include the correct information?
 
Could you send me the GEDCOM for just the individual with that Canadian reference so I can see what the raw data looks like please.
 
Thanks the full reference was:

Year: 1881; Census Place: Aldborough, Elgin West, Ontario; Roll: C_13266; Page: 67; Family No: 301

Beta 9 out that should cater for that properly now.

I've still to add a bunch of other checks as per info Peter kindly sent me.
 
Last edited:
Beta 9 looks like it is still having issues with the US1940 Census.

This time it looks like it is failing some of the Census entries with four digits in the Roll, and those with a Roll reference that begins "M-T0627-0" (those are zeros as opposed to small "o"s).

For example (off the scroll):
Record 292 of 440: US Federal Census 1940 - Failed to add I03615: Cleo Reynoldson b.28 FEB 1900, Roll: 1141, ED: 8-18, Page: 1B.
Record 295 of 440: US Federal Census 1940 - Failed to add I12120: Ralph Kelly Murray b.13 SEP 1924, Roll: M-T0627-00046, ED: 37-111, Page: 1B.
Record 317 of 440: US Federal Census 1940 - Failed to add I06949: Leone Reynoldson b.15 NOV 1927, Roll: 2236, ED: 6-17, Page: 4A.

But:
Record 304 of 440: US Federal Census 1940 - I02071: Robert H Dawbarn b.27 MAR 1895, Roll: 189, ED: 1-138, Page: 7A added.
Record 305 of 440: US Federal Census 1940 - I06047: Laura Helena Hunt b.6 MAY 1874, Roll: 2638, ED: 31-631, Page: 16A added.

I also had no luck again with the Canadian census from 1880, but I presume that is simply because Ancestry does not add the applicable information into the citation.
 
jorghes said:
Beta 9 looks like it is still having issues with the US1940 Census.

This time it looks like it is failing some of the Census entries with four digits in the Roll, and those with a Roll reference that begins "M-T0627-0" (those are zeros as opposed to small "o"s).

For example (off the scroll):
Record 292 of 440: US Federal Census 1940 - Failed to add I03615: Cleo Reynoldson b.28 FEB 1900, Roll: 1141, ED: 8-18, Page: 1B.
Record 295 of 440: US Federal Census 1940 - Failed to add I12120: Ralph Kelly Murray b.13 SEP 1924, Roll: M-T0627-00046, ED: 37-111, Page: 1B.
Record 317 of 440: US Federal Census 1940 - Failed to add I06949: Leone Reynoldson b.15 NOV 1927, Roll: 2236, ED: 6-17, Page: 4A.

I also had no luck again with the Canadian census from 1880, but I presume that is simply because Ancestry does not add the applicable information into the citation.
Click to expand...

The 1940 ones are interesting It should already trim the M-T0627 entry however I've not added all of Peter's rules yet to trim leading zeros. Again seeing the original GEDCOM for those 3 would help test what's happening.
 
Can I just add my own congratulations to Alexander, having downloaded and extracted FTA Beta 9 (I found the thread somewhat latish and by the time had digested same, Alexander was up to Beta 9). Then, to my amazement (and amazement is about the right word) after reading what needed to be done, followed through and then suddenly found 159 Ancestors auto added onto my Lost Cousins Ancestors page within seconds (20 at most). I have a few Census items to check but my overall total has been augments by 159 ancestors, and that is just from the EW 1881 Census. What a boon for LC members and a reward also for Peter and his efforts to publicise Lost Cousins.

As Tim says, and I most certainly echo, Great Job.
 
Beta 10 now out added extra checks for formats of census references that are allowed by Lost Cousins (also strips out things like the M-T0627 stuff) so more references should be allowed.

Latest state of poll is:
4 - I'm a member and have entered everyone at Lost Cousins I've found (I suspect they haven't but...)
46 - I'm a member and I've entered some people but could do more
10 - I'm a member but only for the newsletter
35 - I'm not a member and haven't heard of Lost Cousins before
8 - I'm not a member but have heard of Lost Cousins

So 103 entries which is likely reasonably representative of the 750 people who have liked the page and the 1700+ people who have used FTAnalyzer in the last month. Of the testers mean number of new entries added is around 150 - although one person added over 1000 new entries.

If just 10% of the current FTAnalyzer users use this new feature that will add an approximate 25,500 entries. If just 10% of the Lost Cousins Membership use the program as a result of a newsletter article then we are looking at a likely 1.3 million new entries added. That of course assumes just the people that have already entered lots of people add the extras. If those people how haven't bothered entering Lost Cousins have closer to the 1000 mark as one person did then it could be several million new entries.

All of course of which is why I'm doing LOTS of betas before making this live.
 
Last edited:
You must log in or register to reply here.
Back
Top