IMPORTANT PROJECT TIMELINE UPDATE
Comments
-
I suspect that introducing the transition in the middle of the project is the only reasonable way. One needs actual data (which was released in April) with the OCR field values, new software for the human-checking process, and some human-checking results. Then study how much the human checking improves the OCR results while fixing a large number of problems that inevitably crop up. This is likely to take a few months. Then decide if the new system is good enough.
0 -
@Tom Gibson_1, thank you for your comments. Also, when you run into the issues that you have described above, please report the issue in Report a Problem, and please include your name and also the state in your report so engineers will be able to find it easier in the future.
All work that has been done and being done by the reviewers, such as yourself, is necessary and very valued! We appreciate all that you have done in this project!
0 -
QUOTE: Okay, I guess we will see if the planned accelerated processes produce a usable product.
It won't.
QUOTE: Corrections after publication sounds like an inefficient and incomplete process to me. Many of these erroneous records may never be found in a search and therefore in turn, never be corrected!
I've yet to find my mother, despite her living in an obvious household. I've tried many, many, many variations on her name, and she's still unrecoverable. Along with her three younger siblings.
QUOTE: "Isn't the whole point of the index to aid researchers in actually finding the correct people in the pages? Obviously we will have to deal with whatever changes in processes that are arbitrarily decided by the higher ups, but to keep changing strategies in the middle of this project for the sake of speeding up publishing and possibly sacrificing accurate search results sounds like bad strategy to me."
Completely agree. Plus: It's a strategy that's going to alienate thousands of both volunteers and paying customers. So, it seems like Family Search is shooting itself in the proverbial economic foot.
0 -
I am concerned that (apparently) relying on interested and lucky researchers to search for, find and repair badly indexed records is unlikely to result in all records being found and fixed. The old practice of systematically marching through every record ensured that at least every name was seen, touched, and hopefully reasonably well-indexed.
I could never find my family in the Family Review for Connecticut US 1950 census after the Name Review during the earliest days/moments (1% completed) of that part of the project. I'll be curious to see if I can find them when FamilySearch publishes the records. I did find them on the NARA site and Ancestry.com because I could search on more detailed parameters and via the names of nearby families. I also tried the latter during the Family Review but was not successful. So I wonder what the FS index for my Surname is. How will I repair it if I can't find it?
0 -
I found it interesting today they've added some stats for Family Review 'completions,' when selecting a review choice - New York 6% - 547,538 images left, (37,273 already completed)
FWIW: my personal stats show that I've done a 'reasonable' amount of those completions! Of course, those stats are "images" and personal stats are "families" - not certain how that is computed (Heads of hous ehold?), (basically, I only did my NY 'birth' county.)
It appears that New York needs a lot more help!!!! But, then again, with this 'new' procedure being adopted soon, it may NOT be needed.
0 -
I have had a similar experience as John Empoliti trying to review my family on the 1950 census. My state is now complete and still my family is not findable! I can see them on Ancestry but not on FamilySearch. Sure hope this gets fixed!
1 -
@Mary Louise Lillian So glad that you keep checking and care so much.
Although the states will show that they are 100% complete, it will still take a little time on the back end to prepare them for publication. We all wish it would be quite instantaneous, but the reality is that it takes a team of people behind the scenes to get it ready for all of us to research.
Ancestry and NARA will have the records but they will be a computer aided index and not reviewed. Ancestry's will be updated as FamilySearch published the reviewed records.
1 -
(A follow-up to my post of 2 Aug above)
A quick & rough numerical analysis of NY Family Review possible completion time:
4 Aug 11:30PM 6% 548,001 images left, (36,810 already completed)
5 Aug 10:30PM 6% 547,636 images left, (37,175 already completed)
365 images completed in 'one day' - let's round it up to 400 images: At that rate, the remaining 500K+ images will be completed in about 1,370 days - or nearly 4 YEARS
"It appears that New York needs a lot more help!!!! But, then again, with this 'new' procedure being adopted soon, it may NOT be needed."
(NOTE: Please don't ask me about the inconsistency between the "images left" on the two posts, they were both simply copy & paste)
0 -
It APPEARS MANY more folks are now doing New York Family Review.
Since the report above of ~400 images/day, as of yesterday, indexers were doing about 1,600 images per day for two days = COMPLETION IN ABOUT ONE YEAR!!
BUT, this is again confusing: yesterday at 9 Aug 5:20PM, it was:
7% 543,069 images left, (41,742 already completed) - BUT,
TODAY: 6% 544,603 images left, (40,208 already completed)
(Just copying & pasting)
0 -
Most family history researchers quickly learn to take a census and any associated index with a grain of salt. There are just too many opportunities for humans to be fallible and introduce error into the process.
Starts with enumerators who cannot spell or write legibly, continues with household reporters who do not give the correct facts to the enumerator, and finally with human perception differences found in the indexing effort. Still, I am grateful to those indexers who came before me and provided the assistance needed to break down some of my family's brick walls. For the 1940 and 1950 US Census projects, I have tried to redeem some of my debt to others and pay it forward.
While I am less confident in finding what I need in OCR and computer generated indexes, I will be grateful for what I get. When the indexes fail, I regularly resort to page by page personal review of census images (this was the only way I could locate my g-g-grandfather in the 1850 Census!). I suspect I will have to do additional page by page searches in the 1950 Census to get all of what I would like to uncover.
So, to all of you great indexers out there, thanks for the help! Crossing my fingers it was the many quality volunteers who reviewed the pages important to my family history so they are easy to find.
3