Indexing process changed?
It seems like the batches are cycling through differently now. I have seen the same batch(es) multiple times. Has something changed in the indexing process to make this occur?
Answers
-
@genthusiast -- My opinion after indexing/reviewing over the past 10 years some 250K-300K records total- Indexers are making more mistakes- not reading directions, adding more information than is written plus Project Instructions and Examples are not always clear/explicit and can be contradictory. Also and most recently all the new Get Involved changes have thrown some of us for a loop - even at my advanced age, I understand that change is constant, necessary and can be very helpful but sometimes clicking extra buttons (taking extra steps) can be problematic, leading to discouragement, even momentary despair. That said, I want to emphasize that I appreciate and read everyone's contributions to this Community Forum --without it the indexing task/opportunity would be incalculably more difficult. Every day I learn something new and helpful. I am grateful for all of you - for your youth, your experience, your training, your education and your extreme patience. 😎😎
2 -
Can I get a response from an indexing moderator or someone willing to escalate this question to Indexing Administration please?
It would be nice to know whether the indexing process has changed. My previous understanding and experience with batches was that IF the batch were indexed multiple times it would not be me who did so - it would randomize to someone else. Now instead I keep getting the same batches every so often (great for stats but not for getting more indexing done)...
I have been mostly in Project: Ireland, Cork—Church Records, 1740–1913 [Part B]
My previous understanding of indexing batch process:
1 Indexer receives a random batch, (unless returns batch at some point) complete and Submit Batch.
1 Reviewer reviews that batch. If '>20%' changes are made in that batch it goes for 2nd Review.
2nd Reviewer reviews that batch. If '>20%' changes are made in that batch it goes for 3rd Review.
3rd Reviewer reviews that batch. Whether or not changes are made the batch is returned to FamilySearch for final Review.
What I am experiencing now:
I have received the same batches about 2 or 3 times. I am not reviewing - I am indexing.
0 -
This is pretty far-fetched but … it’s possible that those are batches returned by Reviewers for reindexing with the option “keep indexed data.” If you have seen the same batches again and again, you would be one of the Indexers involved in that cycle. I don’t believe it, but there is one explanation under the current system. A lot of “chance” has to come together for this to happen, which is why I don’t believe it, but it’s theoretically possible. I presume you’re keeping track of the Batch IDs so you’re sure about this strange phenomenon
1 -
No, I haven't been keeping the batch codes/ids - I just recall the batches ... some I am sure about.
Interesting. So are you saying that 'Reindex with Reasons' is an educative process for the Indexer? ... without the attached educative explanation ...
If that is the case - without an attached message from the Reviewer (which theoretically would be possible for me to receive in the inbox) - I won't have anymore detail from the Reviewer and will just put the index back the way I indexed it initially ('cause obviously I think I'm indexing correctly) ... so I suppose this game could go on endlessly. The same Reviewer gets my indexed batch does the same Review edits or Return with Reasons (keeping the data) ...
Wow, maybe your requested Reviewer feedback loop is being implemented ... or maybe it has been working that way and I just got lucky to have this experience ... ?
...just wish they would let me know ...
Obviously, I would turn this into an Idea that Reindex with Reasons: Keep indexed data - allow the Reviewer to attach a message delivered to the Inbox of the Indexer of that batch - that blinks red (ok just blinks then) when the batch is downloaded - or is required reading before the batch could be opened... Additionally, if I have an 'assigned' Reviewer - and we aren't seeing 'eye to eye' on Project Instructions or Entries - it would be nice to be able to 'note' (again I would probably vote for changing edit color) my reasons for edit of those particular Entries. If there isn't two way communication - this just becomes a frustrating but bemusing indexing scenario ...
0 -
Hi @genthusiast
No, I was referring to the ordinary non-Labs option under Batch>Reindex. The Reviewerget the options to keep the indexed data or delete it.
Regarding the Labs alternative “Reindex With Reasons” form, which I used to tout and be a big fan of, it is my understanding that there is no feedback from it to the Indexer (I’m not sure if there ever was) except in the most extreme cases. So I don’t even think about it anymore. If I found a case of malicious Indexing I would use it.
1 -
hmm ... well I wish the Reviewer(s) would just stop bugging me then 😀 ...
...and that I 'knew' if that is what is occuring or if something has changed about the routing of those ...
0 -
I don't really know what's going on. I was just imagining a scenario that might explain it. It sounds strange.
1 -
I have also asked this question before but not gotten any Project/moderator response (where did the indexing moderators go that could escalate items as needed?). Since the Project I've been working in recently: Ireland, Cork—Church Records, 1740–1913 [Part B]
I wonder if it is a subset, contains a subset or is a discrete continuation of the initial Project: Ireland, Cork—Church Records, 1740–1913
When Part B was released there were some 'peculiarities' opening batches initially (another reason for questioning what has changed).
0 -
I am curious if these same batches are all in the same project or is it random? Is it only the Ireland, Cork project records?
We could speculate as to why you would be seeing the same batch over and over but would be better if you could give us at least one shared batch code example. Do you keep records of all the shared batch codes you review? That would be helpful in seeing what was going on here.
1 -
Yes I believe they have all been in Part B. I don't normally keep the codes - but I can give one a little later (not with me right now).
0 -
I have already submitted the Batch code: M3ZL-4TC
If you can determine why - I have received that batch at least twice.
0 -
If you keep a record of your batches it will help you verify if they are the same batch and not duplicates of the same record. Occasionally there are duplications because the camera operator wanted to get a better image. These would be the same record but different batch codes.
0 -
Yes I'm aware - if I see that same batch yet again I guess I will have my answer. It just seems like the same batches are cycling and it would be 'nice' if indexing admin passed on when changes in process are made. And if there were any direct communication here in Community ... More shared information keeps indexers informed/happy.
1 -
We have heard nothing about any cycling. We do answer questions with the information we have. :) Adding a personal opinion here. I am inclined to think that what you are seeing is from the digitizer trying to get the best image they can or even human error in digitizing. Those would result in duplications. Those scenarios are only my personal opinion though.
1 -
I am inclined to think that what you are seeing is from the digitizer trying to get the best image they can or even human error in digitizing.
So the digitizer for indexing projects is instructed to include extraneous images - not duplicate/pick the one best image to replicate the film/original record? Yes - indeed that would mean a separate batch code. Maybe the machine operator just allowed the machine to create duplicates or forgot to remove extras? I would like to think it wasn't the operator introducing the duplication. But yes maybe this does point to duplicated images in this Project - this is the first I have experienced this issue (thus the questions).
0 -
@genthusiast There are often duplications of images, especially in these old records where they can't snap a good shot due to the quality of the original or they way it might be in a book (bound and trying to capture the innermost text without ruining the old book binding), or they play with the lighting to get a better image, etc. Very occasionally you will get an image that has the operator's notes telling you why they have chosen to re-photograph.
I have worked on old parish records where the same record has been photographed numerous times to capture the whole image or to get a better image. So, there is definitely some repeated information collected by each indexer/reviewer who received those batches. Since each batch is indexed, information may be repeated, but by doing so, no name is ever left out either.
3 -
I understand. But that is not the situation with these images/batches (no imaging notes attached). The images are the same (quality and content - yes I am sure) - so if they are duplicates (from the indexing batcher/creator) - they were duplicated for no purpose. I index them exactly the same (or close to it).
0
