If this is a bulk import report, send the spreadsheet to services@wildme.org with the email subject line matching your bug report
In which Wildbook did the issue occur? Zebra
What operating system were you using? (eg. MacOS 10.15.3) PC
What web browser were you using? (eg. Chrome 79) Edge
What is your role on the site? (admin, researcher, etc) researcher
What happened? Individuals not matching even though am sure they are in the software. When I try to open encounters of the same individual, I get collaboration denied. I have access with the persons that worked on Ol Pejeta zebras however there is public@localhost which was used to transfer all our old database to wildbook that I donāt have access. I guess that might be another problem.
What are some steps we could take to reproduce the issue? I have shared screenshots of some individuals that didnāt get a match but the same individual has 3 encounters in wildbook.
Can you share links to the encounters or match pages of the zebras youāre not seeing correct matches for? When you share the link, can you also let me know which individual should be showing up as a positive match?
Iām seeing the same issue with public encounters requiring a collaboration first. Iāll need a little more time to work on that one. Thank you for letting us know.
For the public encounters, the issue was that the public username wasnāt in the correct format, so Wildbook was treating it as a regular user. Iāve updated it but itās going to take a little while for indexing to complete. Iāll check in tomorrow on its progress.
Quick update: we found two bugs related to viewpoints and social groups contributing to this that weāre in the process of fixing. Iāll update this post when I have an update.
This should be fixed now. Matching was excluding annotations that were missing a viewpoint (which the encounters for the individuals you shared didnāt have). Now that this is corrected, re-running those matches should now show you the correct individual in the top results. Thanks for flagging this!
It looks like of the 6 encounters we have of 05_005, only one has a location ID. I just re-ran the match, but this time, I checked the box for āmatch data without location IDā to see if the correct matches show up and the top result is one of his encounters without a location ID.
The same thing happened with this second encounter. Only one of 04_006ās encounters had a location ID and re-running the match to look for encounters without a location ID showed one of the encounters without a location ID as the top result.
Since those encounters on both individuals without location IDs are public submissions, you should be able to add a location ID when you come across these if you know the area they were in based on the photo/their usual location.
Hi Anastasia,
Thanks for helping me out but I guess the problem isnāt location ID. The one in the link shows no match yet it has 2 encounters in wildbook. I have added location ID to the 2 encounters and yet it doesnāt show any match.
Itās OP11_204.
Kindly check for me what else should be the cause of mis match.
Thank for the new example! Weāve uncovered a bug thatās the underlying cause of these missed matches. Iām writing up a new ticket and will post a link to it here when itās ready.
Thanks again for reporting this. The fix for this ticket has just been completed in Zebra Wildbook. Can you let me know if you continue to see any unusual behavior with matches?
I couldnāt find the individual page for OP05_017 in an individual search. Can you link to the individual page or double-check that the ID is correct? Once I have this, Iāll be able to keep looking into the match issue.
Once the zebra is assigned an ID on this encounter, you should start seeing it appear in results for Ol Pejeta and its sub-locations.
All that aside, it does look like initial search results are still looking too broadly for candidates (11,000 vs 2,000 that are actually in Ol Pejeta). Weāll keep working on this to make sure you start seeing better matches. Thanks for your feedback and for your patience!
