Indexing Labs
I am doing review on U.S. New York-Eastern District Naturalization Records. My question is when reporting thru the labs, I select the reindex box and click on the appropriate boxes and then for other I type in what needs to be corrected. I find that I sometimes get back what I have requested to be done such as taking the record month, day and year from the affidavit of witness when it is either blank or a date from the declaration of intention at the bottom of the sheet is used (as this happens quite frequently). I may get it back but the reindexed data has the same data and has not been corrected. I know that it goes into some que for someone to look at but does the reason for the return get shown?
Thank you in advance.
Answers
-
I think I may have put something which may not have come out correctly. I know that you do use the declaration of intention at the bottom of the page when there is a picture. What I was trying to say that it looks like this. The record month and day and year are put in from this and not from oath of allegiance.
Sorry if I may have miss worded what I was trying to say. I do this quite frequently which is never my intent.
0 -
Honestly, don't report it via Labs. Just correct the date with the affidavit of witness date (not the OOA date, that belongs with the OOA) and submit it.
0 -
@erutherford I tried that, but it seems that for every two that I do, the next six are the way I described, or they are No, Extractable data.
I have another question that I should have included. When they have been sent them back to reindex and I know that that go into a que. Does the original indexer know that they are making an error or whichever word is used for error? If not, how do they know what area they may need help in?
0 -
For those six that are marked as what you described, say the DEC February 23, 1946 vs. the witness date of March 5, 1953, always use the witness date. Sometimes, you will have to go to the next image to find that witness date. In the case of reviewing, you can go to the +1 of your reference images (vertical toolbar/book-like image) to find that date.
With the NNEDs, you can do one of two things. The first is to throw it back on the pile. To do this, go to Batch ---> Reindex Batch and another indexer will pick it up or you can fix the mistakes. Sometimes I get frustrated, too, and will throw a batch marked NNED when I know it's something that has to be indexed, and I'll throw it back.
As far as I know, the original indexer does not know what happens to their batch after they submit it. It used to be, back when we had the desktop version, that once you submit a batch, it was reviewed and what mistakes you made were pointed out by the reviewer. I still remember my first project, Alabama WW II draft cards (where I found the best name ever; Burnt Corn, Alabama), and I would literally put 215lb instead of 215 for the weight. I kept a notebook of my mistakes and how to fix them for a few years before I threw it out (and it had been about that long since I needed it). The only things I know now for certain are there is a review from folks like us, then there is an extra extra review before the project is made live for researchers to see.
0 -
Hi Faith. Indexers and reviewers are not made aware of their errors, unless they are repeated (and I think critical errors). Sometimes if FS notices a pattern of a poor indexer or poor reviewer, they do reach out and offer them help. The lab feature isn't really necessary. It is an experiment, but, the computer system will also alert them to repeated errors by the same indexer or reviewer.
If you don't make more than 20% of changes to an index, then you are the only reviewer. Changing that date using the correct date from the Affidavit of Witnesses statement might just be enough to throw these to a second reviewer (3 of 16 fields). So, that batch will have at least 3 views (Indexer, Reviewer 1, Reviewer 2). If Reviewer 2 doesn't agree with Reviewer 1, then it goes to a 3rd Reviewer, and finally a FamilySearch team member.
0 -
@erutherford; @Melissa S Himes Thank you both for answering question. To me and this is my opinion only. If I was made aware of mistakes/errors that I was making during indexing, I would find it beneficial to have someone let me know what I was doing wrong, so that I could correct or prevent what I was doing wrong and make it easier for the reviewer. I find that at times when doing a new project that I have to leave the example tabs open at the top and refer back to them if I am unsure of something.
Thank you once again for your answers.
0 -
@FaithHamm The problem is that oftentimes the reviewers are the ones who are wrong. Back in the olden days when FamilySearch used the two indexers and one reviewer method and we learned our results, the reviewers hadn't read the instructions and were wrong. Then we could give feedback on the incorrect review. It does no good to confuse indexers by giving them incorrect feedback.
For instance, if a reviewer would tell an indexer to use that date from the front the of the petition where it says that they have resided in the US for a number of years from the date of the petition, (like that screenshot), the reviewer would be wrong. The field help tells us: Please index only the record date in this field. Do not index the petition date or arrival date.
At least with this new program, the batch can have up to four reviews before it gets published. The option for multiple reviews with no feedback should actually help with the quality of the publication of data.
0 -
@Melissa S Himes I can see where that can happen on getting confused and telling someone the wrong information. It is nice to know that should an indexer and reviewer not agree that is goes thru an extra step or two before it gets published.
Thanks for the information.
0