Unauthorized headings should not be displaying "See Also" references

[Kathy Lussier]

Evergreen version: 2.5.3

When browsing authority records, if a user enters a search for an unauthorized term, the browse list should display a "See" reference to the authorized heading. When the user then goes to the authorized headings, they should see all of the "See Also" references associated with that heading.

In Evergreen, when a user enters an unauthorized term, they see a "See Also" reference followed by a "See" reference. That initial "See Also" reference should not be displaying in the list.

For example, in the authority record for Nikki Poppen, we have a Eagen, Nikki Poppen- listed in the 400 field as an unauthorized entry. When a user searches Eagen, Nikki, they should find a "See" reference directing them to the proper authority term.

http://www.screencast.com/t/JWWfzcygfuN

They do receive this "See" reference, but, at the same time, they also get a "See Also" reference to the record for Bronwyn Scott, the author that is listed in the 500 field for Nikki Poppen's record. http://www.screencast.com/t/fURnBelE

This "See Also" reference should only display with the "Poppen, Nikki 1967-" heading. It should not display with all of the unauthorized headings associated with Poppen's record.

[Mike Rylander (mrylander)]

First some background on how things are happening behind the scenes ...

Unauthorized headings (Brownyn, Scott in this example) are linked to the bibs making use of them based on direct string matching because, by definition, unauthorized headings don't have an authority id. If that unauthorized heading is in use by bibs and is linked to an authorized heading, as this is, and you are in the browse page that would show the authorized term (in this case, the lowercase version of "eagen nikki poppen") then we show the unauthorized heading.

The logic there is that because bibs are making use of the heading, and there's a relationship between the two, discoverability is increased for the user that knows about the Eagen heading but may not know about the Brownyn heading.

Put another way, if there were no bibs using the unauthorized heading or the unauthorized heading were not linked via $0 to the authorized one, then it wouldn't be surfaced. This may be different from the logic we'd use in a pure authority browse interface, but I can't say I'm certain about that case -- we're specifically attacking the problem of getting a layperson to the records that are in use, augmented by as much authority linking data as we have access to.

So, the reason we show that is specifically because of the $0 link between the 500 (Brownyn; unauthorized heading) and the authorized (Eagen) heading.

With that in mind (bib browse + authority linking), does the current behavior make more sense to you?

[Kathy Lussier (klussier)]

Hi Mike,

Sorry for the delay in commenting. Yes, I agree that we should try to save clicks for the user, and I agree that giving direct access to those See Also references makes sense here.

However, I do think a couple of changes could improve this display:

1) It seems more intuitive to me if the See reference displayed above all of the See Also references. We previously discussed this at http://irc.evergreen-ils.org/evergreen/2014-04-14#i_88365.

2) The scope notes seem to be showing the same behavior. When the browse is displaying an unauthorized subject heading, it will display the scope note from the linked cross reference. In this case, the addition of notes from a crossreference clutters the screen and doesn't add value for the end user.

[Srey Seng (sreyseng)]

Hi Mike,

Explanation in comment #1 (https://bugs.launchpad.net/evergreen/+bug/1307603/comments/1) provides great background information.

However, I did notice that there appears to be a little mix up in references to authorized heading and unauthorized heading. For instance, in the explanation, you refer to "Brownyn" as an unauthorized heading and "Eagen" as an authorized heading, when in fact, if you had to categorize them one or the other, it is the opposite.

Given that, does it in any way affect the explanation?

Thanks!

[Kathy Lussier (klussier)]

Adding a note that, although I said in comment #2 that I see the value in saving clicks for the user, we've found in practice that including these linked cross-references in the browse list often leads to unexpected results.

For example, the attached screenshot shows an example with the Christianity heading. The Christianity authority record has a 550, broader term reference to Religions and another 550 to Church History. Religions has a 550, broader term reference to Civilization, so we now get a "See Also" reference to Civilization. Church history, on the other hand, has a 550 for History, and we therefore have a See Also reference for History listed under Christianity. Neither of those are cross-references I would choose to display under the Christianity heading.

[Kathy Lussier (klussier)]

Marking this as a duplicate of bug 1638299 since the 3.0 authority infrastructure project addressed all of the concerns in this bug. Scope notes now only display along with their main entry heading and the list no longer includes linked cross references.