Improve/expand use of item type icons (from 1776)

[gigamorph]

Problem Description: There is a disconnect between item type (eg. sound recording) and icon (book/textual work). Can we start using more icons? If this object type is sound, isn't there an audio icon that could be applied? This expansion should probably happen on Work Types rather than Object Types.

Expected Behavior/Solution: Use the most relevant icon for each object/work type. The mapping from classification to icons is here: https://docs.google.com/spreadsheets/d/1gsVcsR-_b1kPbGi-Xz7w_GUJ8BHvmOjFW28ZZ9JGt6Y/edit#gid=0

Requirements:

• Icons will match their item type correctly(ex: records of books will contain the book icon, objects will contain object icon and so on)

Needed for promotion: If an item on the list is not needed, it should be crossed off but not removed.

• [x] updated list of icons
• [x] Analyze number of object types in the data/generate list (Peter, Kelly) - not all, but analyzed the top classifications
• [x] creating an ordered system solution for icons (peter and kelly)
• [x] Decision about mapping/effort
• [x] icon/data mapping
• [x] Review new icons with Heather
• clothing
• newspaper/periodical
• photographs
• furniture
• [x] Request new icons from POW - Heather
• [ ] UI design that accommodates multiple icons - Heather
• [x] requirements for calling the correct icon (Kam, Peter, Rob)
• [x] Data constants/ML https://github.com/project-lux/lux-marklogic/issues/59
• [x] examples for UAT testing
• [x] Wireframe/Mockup - Mike
• [x] Committee discussions - Sarah
• [x] Feasibility/Team discussion - Sarah - [ ] Backend requirements - TBD - [ ] Frontend requirements- TBD
• [x] Questions
• What to do when there are multiple types/icons?
  • only show one, based a given priority
• Should there be more specific icons for YPM specimens (e.g, plant specimens)
  • no
• What new icons are needed?
  • see icons below
• Who needs to review/sign off on icons?
  • icons below are approved

UAT/LUX Examples:

Priority:

1. Scores and Notation
2. Photographs
3. Furniture
4. Journals & Newspapers
5. Clothing
6. Numismatics
7. Cartography
8. Video
9. Audio
10. Software & Electronic Media
11. Visual Works
12. Textual Works
13. Objects
14. Specimens
15. Collections

Dependencies/Blocks:

• Blocking: https://git.yale.edu/lux-its/lux-web/issues/1746 (closed)
• Blocking: https://github.com/project-lux/lux-frontend/issues/22
• Blocked by: https://github.com/project-lux/lux-marklogic/issues/55

Related Github Issues:

Related links:

• https://www.bugherd.com/projects/284041/tasks/1350
• https://lux-front-sbx.collections.yale.edu/view/object/735d8b00-b93e-48b3-8327-4eb5986a2814
• https://docs.google.com/spreadsheets/d/1gsVcsR-_b1kPbGi-Xz7w_GUJ8BHvmOjFW28ZZ9JGt6Y/edit#gid=573072421 (Most Common Object Types and Work Types)
• https://docs.google.com/spreadsheets/d/1gsVcsR-_b1kPbGi-Xz7w_GUJ8BHvmOjFW28ZZ9JGt6Y/edit#gid=573072421 (Type Icon Association)
• https://git.yale.edu/lux-its/lux-web/tree/main/client/src/resources/images/type (Viewing Icons in Git)

Wireframe/Mockup: Place wireframe/mockup for the proposed solution at end of ticket.

Priority set by LUX core team.

[prowns]

first pass of a straight ranking of the icons:

• scores-notation: 1
• photographs: 2
• furniture: 3
• journals/newspapers: 4
• clothing: 5
• numismatics: 6
• cartography: 7
• specimen: 8
• video: 9
• audio: 10
• software-electronic-media: 11
• visual-works: 12
• textual-works: 13
• object: 14
• collections: 15

[roamye]

labeling this issue as blocked as: https://github.com/project-lux/lux-marklogic/issues/55 is blocked since we are awaiting a new data set.

[roamye]

This ticket is no longer blocked as https://github.com/project-lux/lux-marklogic/issues/55 was closed today during IT team meeting.

From team meeting we will need to add AAT references.

[kkdavis14]

@prowns I hate to throw any part of this back to the design phase, but did we ever have a design for fossil, that is different from the butterfly/specimen? I can imagine YPM would like to distinguish and fossils getting a butterfly icon doesn't work that well.

[kkdavis14]

Also, I've reordered the priority list slightly and place specimen lower down. I updated the priority list in the ticket description to reflect this change.

[kkdavis14]

to start @clarkepeterf

~~scores&notation: 300026427 photographs: 300046300 furniture: 300037680 journals: 300215390 & newspapers: 300026656 (there are two different AATs here, both should get this icon) clothing: 300266639 numismatics: 300054419 cartography: 300028052 (cartographic materials)~~

how does the new way of doing data constants affect the mapping we did for these types in the spreadsheet?

[kkdavis14]

Kelly to create concordance of {icon type: [AAT, AAT, AAT]}

[kkdavis14]

ObjWorksIconAATs.csv @clarkepeterf see attached csv for icon type >> AAT(s)

[prowns]

On agenda for 5/13/24 Data WG.

[kkdavis14]

@prowns I think this can move forward without having the YPM fully classifications sorted, FYI

[prowns]

@roamye - I am moving this to prioritization.

[kamerynB]

@kkdavis14 @clarkepeterf This issue is blocked until the frontend removes data constants and implements AATs correct? It would be time consuming to update this only to change it again.

[prowns]

Yes - need list of AATs, so yes blocked until that is available. @clarkepeterf will add additional information here from teams messages with @kkdavis14

[clarkepeterf]

@kkdavis14 I remember discussing a while ago that some of the classifications we want to use for icons do not have AATs

• Does that mean this is not possible with AATs?
• Or is it possible to add AATs or some other constant identifier for those classifications?
• Or do any of those classifications overlap with other classifications that do have AATs (therefore we can rely on the other classification to determine which icon to use)?

[kkdavis14]

@clarkepeterf yeah, many YPM taxon will not have AATs and it isn't possible to add them (as the concepts don't exist in AAT). If YPM has given a GBIF, we could use that as the constant identifier, this probably covers most of the cases.

[clarkepeterf]

@kkdavis14 can we assume that there is a GBIF for the YPM taxon that do not have AATs? Do we need to confirm or do some work to ensure that is the case?

[kkdavis14]

@clarkepeterf I don't think we can assume it. I'll run a script to check the uris from the icon mapping sheet that are YPM taxon and see which have GBIF equivs and which don't. The GBIF equivs could then be added to my csv of April 30.

There's not going to be a way to ensure they have GBIF equivs--if they're not there, they probably don't exist either.

Could expand to include Wikidata if not GBIF if no AAT. What do you think?

[prowns]

Can we default to "Objects" for any YPM items that don't have a more specific designation?

[kkdavis14]

I think it does this already, yes.

[clarkepeterf]

@kkdavis14 as long as we have some sort of unique identifier that will stick with the record across datasets, it meets our requirements. Could be AAT, GBIF, wikidata, or something else. It just needs to continue to exist in the record over time

[clarkepeterf]

@kkdavis14 and the pipeline needs to be able to include that unique id in the Object/Work that references the concept.

E.g. :

{
    "id": "https://lux.collections.yale.edu/data/object/abcdefghijklmnopqrstuvwxyz",
    "type": "HumanMadeObject",
    "_label": "Some YPM Artifact",
    "classified_as": {
        "id": "https://lux.collections.yale.edu/data/concept/abcdefghijklmnopqrstuvwxyz",
        "type": "Type",
        "_label": "Some YPM Taxon",
        "equivalent": [
            {
                "id": "<AAT, GBIF, Wikidata, or etc.>"
            }
        ]
    }
}

[kkdavis14]

@clarkepeterf We only do AATs as data constants right now. Adding Wikidata or GBIF would need Rob's approval. Also, I don't think EVERY AAT, etc, is being injected as a data constant.

For example, a YPM taxon building this LUX record: https://lux.collections.yale.edu/view/concept/c8f28d7d-bcd7-4c7d-8d09-2ddd9dcd9170 (Crinoidea) has an AAT equiv, but that isn't a data constant on any of the objects with that classification, e.g.: https://lux.collections.yale.edu/data/object/da0b2c21-3eae-43bf-a480-975ec0ac6f40

Recommend adding this specific part of this topic to an upcoming ITS meeting that we're all attending @prowns

just a reminder that "It just needs to continue to exist in the record over time" is reliant on the unit's changing their data. e.g. they decide it's a different GBIF or AAT, it will break our icon code. We would just have to find them and fix them if/when that happens.

ran the script to check this out anyway and ~100 of the taxon on the icon type list have no external equivalents at all, so they won't be able to get a new icon beyond the basic "object".