Has anyone invited "Professor Joe" to join us here and have a heart-to-heart?
US1910Project
Answers
-
Some of the errors I've had to correct were for people who were deceased before 1910. IOW the name of the project is misleading.
0 -
I get frustrated as they create women in the 1910 census under their married names, with or without a spouse. A simple search with a child would likely find the family that would not be found using a women's married name vs maiden. I am regularly merging 1910 project created names. I understand it is volunteer based and I work in a couple projects, but quality control or instructions could be greatly improved. How to make them look harder to find the families before creating records.
1 -
Quality control is very lacking, or the students are novices. They often create a duplicate profile and don't even attach the record. Sometimes they don't assemble the family either.
0 -
Has anyone invited "Professor Joe" to join us here and have a heart-to-heart?
0 -
Here's another example of unnecessary duplication caused by the 1910project:
Ellen McCarthy
1863 – 25 May 1943 • LBXR-ZWT
This woman and her many children were already in Family Tree and now there are duplicates of her and 7 of her children. I am in the process of merging them all.
I agree with others that somehow this "project" should be halted.
2 -
Yes. He said he may join in on this thread at some point 🤷♀️
0 -
I see these people are also working in the 1900 census, can't Familysearch do something to stop them ?
1 -
If you go to:
Katy Dombrowiak
1876 – Deceased • GFRM-1NW
You will see that the US1910Project created TWO families in family search for her!!!!
Look at the sources. You can see that US1910Project is credited with TWO sources from the Project...the 1900 and 1910 US census, one for each family!!
This is ridiculous!!
So she has TWO husbands plus TWO children that need to be merged now due to various spellings of the last name on the two different census records.
1 -
The damage done not only by this project but by the creation (prior to the introduction of Family Tree) of one ID for every "event" where a person is included (e.g. separate William Brown IDs created against each of the baptism records of his ten children that includes his name) makes our bad practices pale into insignificance.
4 -
@Paul W Not to mention all the triplicate William Browns and wife and children when the baptismal register was indexed 3 times.
2 -
That's actually not true to say that FamilySearch won't do anything because it is a BYU project. The project has received a great deal of attention at the highest levels of FamilySearch. It has been scrutinized and reviewed numerous times. There are just more pieces to this pie than most understand. I'm not getting into a debate as I understand better than most, the issues that have been raised here and elsewhere in the community. But I also see, as does FamilySearch management that this project is also doing some really good things. I am working with a couple of folks here to get some info for you about what is going well and why this project continues. We are working with Joe to keep you better informed and we would love to have him come out to the community and talk with us. However, we have to receive him kindly and not beat him up because we are frustrated. Sam 😃
4 -
Family Tree is based entirely on historical records and all such issues are temporary. It is all work in progress, please excuse our mess.
1 -
I was talking about the us1910project, not Familysearch, although Familysearch is OK with the us1910project.
0 -
I must admit I'm a bit baffled by your response here, as it does not appear to be addressing the matter being raised. Also, is your comment, "...please excuse our mess" meant to be an apology on behalf of FamilySearch?
0 -
I can't speak for dontiknowyou, Paul, but "Please excuse our mess" is often seen in US English when something is under construction or repair. It's used both on physical signs, in a store or other location, and on websites that may be undergoing an update.
1 -
US1910Project is contributing, albeit at a rather inept novice level. Should we demand perfection from newbies, or should we tolerate their baby steps?
All of Family Tree is under construction, a work in progress. Along with the privilege of being able to see and correct all other contributors' mistakes there comes, I think, an obligation to be tolerant. Noblesse oblige.
1 -
The us1910project is causing more harm than good, and they flat out said they're not going to change course on their procedures. I get what you're saying, but we're going to have to respectfully agree to disagree.
4 -
Of course be tolerant of newbies / genuinely inexperienced users, but not of project instructions from an academic source, which do not appear to be deterring this mass creation of duplicates that might take years to clean up.
Family Tree instructions are not to create duplicates unnecessarily, so the project needs far more oversight than has been provided hitherto.
5 -
Is there a thread about the mess in the 1950 census?
5 -
Thanks!
0 -
Regarding the 1910 project, I am again finding duplicates being created unnecessarily. Latest case in point:
Rose L Nicholas
29 June 1876 – 28 April 1972
• KWVP-P8M
I know this example is from last January, but still . . .
1 -
So Sam, please let us know what are the "really good things" that are coming out of the 1910 Disaster.
3 -
Brace yourselves, guys, the US1910Project account has been renamed USCensusProject.
4 -
I agree--census projects were not well thought out and should be stopped until the substantial problems are fully addressed.
For starters, it's never a good idea to create a Family Tree person based on just one record. That's a sure way to introduce errors.
And census records are some of the most error-prone historical records (names, ages, relationships and birthplaces can be and often are misstated). And that's just on the images.
Errors are increased because the project uses indexes of census records, not the records themselves. So indexing errors reduce the quality even further.
All that being said, this project could have worked in a private tree. It isn't the most efficient way to build a tree, but a person doing this could have cleaned up the many errors after the initial data dump. But in a shared tree, the data doesn't just stay the way it was added. People are constantly changing it. That's what a shared tree is for.
And the trouble is, when the data is low-quality to start with, in my experience people tend to make incorrect changes--attach wrong sources, create wrong relationships, do bad merges. That just exacerbates the initial problems.
The other thing that troubles me is that the project is essentially adding my family to the tree for me. Why should they do that, especially with so many errors? I and my family want to build our tree!
And if FamilySearch thinks Family Tree is not growing quickly enough, there are much better ways to grow the tree with high-quality data--ways that save time in the long run.
I'll end with what I hope is a win-win suggestion. If census projects continue, their data should be put in a separate database, outside Family Tree. Volunteers could clean up the messes there. Then, when the data is clean, it can be added to Family Tree. That doesn't solve a core problem (that I would like to build my own tree rather than having some computerized project do it), but at least it would substantially cut down on the errors and messes.
I'd love to hear other users' thoughts on this idea.
5 -
You would have thought that the BYU RLL project should have been required to develop a process to eliminate duplication before they just started to add to FT without doing so.
What value does this project add when people can already look up their family connections in FS Records, Ancestry.com, etc.?
2 -
About six weeks ago you wrote:
I am working with a couple of folks here to get some info for you about what is going well and why this project continues. We are working with Joe to keep you better informed and we would love to have him come out to the community and talk with us.
Sadly, still no positive news involving the benefits of this project, including nothing from Professor Joe. I feel these negative comments will just go on and on until users can have some evidence that that benefits outweigh the detriment of the exercise.
I have watched Joe's YouTube video on the project and he makes a strong case for it. Unfortunately, a lack of supervision appears to have led to a situation (masses of duplicates) that was never envisaged or, I'm sure, was intended.
For those who work daily to reduce the number of duplicates in the tree, it makes depressing reading that the scope of the project now appears to be extending beyond the 1910 census.
Further to a request to provide it, I'm afraid I can't locate the link to Joe Price's YouTube video. Would be grateful if any Community member having it could post it here.
3 -
Is this the one, @Paul W? https://www.youtube.com/watch?v=NErxYeiwhVI
0 -
No, afraid not. The one I was referencing was a few years old, I believe. This one concentrates on adding the census sources to existing profiles (although adding "missing" relatives at the same time), but the "original" clip described the wider aim of carrying out this work based on communities. (i.e.blanket coverage of everyone found in a specific area / township.)
Update - I think it must be the one at https://www.youtube.com/watch?v=c64LKzc_wig
1
Clear
No Groups Found