open-ils.supercat.authority.browse_center.by_axis.refs returns identical authority records

At my library we use authorities a lot, and I wanted to make sure I was understanding what you are seeing in this bug.

I tried recreating the steps you took and I think what you are seeing is not a bug, but an intended feature of the system.

When I right click on the bib tag 100 of bib number 100, I get results that partially look like the following...

Bohm, Georg, 1661-1733. Passion nach dem Evangelisten Johannes
   See from: Handel, George Frideric,1685-1759. Passion nach dem Evangelisten Johannes
   See from: Bohm, Georg, 1661-1733. Johannes-Passion
   See from: Bohm, Georg, 1661-1733. Passion according to Saint John
   See from: Bohm, Georg, 1661-1733. Passion according to St. John
   See from: Bohm, Georg, 1661-1733. Saint John Passion
   See from: Bohm, Georg, 1661-1733. St. John Passion
   See from: Bohm, Georg, 1661-1733. Janos-Passio

Bohm, Georg, 1661-1733. Passion nach dem Evangelisten Johannes
   See from: Handel, George Frideric,1685-1759. Passion nach dem Evangelisten Johannes
   See from: Bohm, Georg, 1661-1733. Johannes-Passion
   See from: Bohm, Georg, 1661-1733. Passion according to Saint John
   See from: Bohm, Georg, 1661-1733. Passion according to St. John
   See from: Bohm, Georg, 1661-1733. Saint John Passion
   See from: Bohm, Georg, 1661-1733. St. John Passion
   See from: Bohm, Georg, 1661-1733. Janos-Passio

Bohm, Georg, 1661-1733. Passion nach dem Evangelisten Johannes
   See from: Handel, George Frideric,1685-1759. Passion nach dem Evangelisten Johannes
   See from: Bohm, Georg, 1661-1733. Johannes-Passion
   See from: Bohm, Georg, 1661-1733. Passion according to Saint John
   See from: Bohm, Georg, 1661-1733. Passion according to St. John
   See from: Bohm, Georg, 1661-1733. Saint John Passion
   See from: Bohm, Georg, 1661-1733. St. John Passion
   See from: Bohm, Georg, 1661-1733. Janos-Passio

The repeated appearance of "Bohm, Georg, 1661-1733" is because the code is matching on that one authority record's auth tag 100, but also on some of that auth record tag 400 values. When there are matches on auth record's 4xx/5xx tags, the current code will repeat the display of the auth record with all of its 1xx/4xx/5xx. This looks like incorrect duplicate record, but that is how it meant to behave at this point.

There has been some talk about changing the code so the matches on auth tags 4xx/5xx do not appear so confusing, but no new code has been written at this point.

Perhaps what you are seeing is something different, so feel free to give you your feedback, but I suspect this is "a feature not a bug."

Yamil

Yamil,

Evergreen is searching both the the main and added entry headings, as you suggest. That's intentional. What you're seeing is particularly visible when there are only a few records (as with the concerto set), or when the added entries are very, very similar to the main entry.

We're completely open to suggestions on how to improve this, but searching the added entries in addition to the main entries is pretty important for cases where you know the added entry but not the main.

Do you have any thoughts or mock-ups on how that display might be improved?

I just wanted to state for the record, that Mike has previously asked for feedback from the Berklee library to come up with a new way for this code/feature to behave. We at Berklee have so far not provided any suggestions or mock ups to the community.

Thank you Mike once again for your offer to receive our feedback so new code could be written to try out new ideas. I have been thinking about this behavior and other "manage authority" changes since the last dev hack away in September. I hope to provide some feedback in the coming year.

Yamil,

I didn't intend to "call you out" directly. :) You gave a good summary of the underlying cause of the suboptimal display, so I wanted to address you WRT that.

Lucien, do you have any thoughts on how this display could be improved without removing the cross-reference search?

TIA!

Quite the opposite, I wanted to make it clear that you have repeatedly asked for feedback. That it was not the first time you have looked into improving this feature.

Okie Dokie, I think I got the why-ness of it all. Your suspicion is correct Yamil, what you describe is what our staff see with our live data as well. So it is intended feature behavior.

My guess is people only want to see one (unique by id) authority.
But then I am not an end user so I'll forward this issue to our catalogers to get some grassroots feedback.

Thank you.

Mike and others,

Should we turn this bug into a wishlist bug? We could collect feedback and experiments here.

Yamil

Encountered this working on bug #1852782.

I'm also not an end user, but in the context of managing authority links to bib records, where we display all of the 1xx/4xx/5xx values from each authority record, I see no value in repeating them. It definitely confused me at first.

A modification to or alternate version of authority.browse_center.by_axis.refs that groups and pages by authority ID instead of heading would appear to solve the issue for the linking UI's.

Alternatively, should the MARC editor linking UI /only/ display the matched headings instead of the full authority record?