Increase the character limit of citation remarks

[Jegelewicz]

Would it be possible to increase the character limit of citation remarks? While we would not always use remarks and don’t plan to copy verbatim text into remarks normally, there would sometimes be cases where this would be helpful. For example, there is a book produced by UAM that includes captions for some individual works in the collection that would be appropriate to include in the catalog record, and it makes the most sense to us to include these captions in citations. In fact, it’s the book Angie referenced above, Looking North. I believe we might want to be able to include a caption of about 2000 characters at most, though the majority of these captions would probably be about 1000.

Originally posted by @krgomez in https://github.com/ArctosDB/arctos/issues/1861#issuecomment-572755293

[dustymc]

Yes it's possible. At some point we explicitly decided to force brevity there; I'm not morally opposed to relaxing that, but I'm not entirely sure I see the point of reproducing the book either!

[marecaguthrie]

I absolutely understand and agree with the need for brevity. For published research that is readily available that makes complete sense. The text that we want to add from the Looking North book are essentially curatorial descriptions that add significant value to the understanding of the objects. We had thought about proposing a "curatorial description" attribute, but because these are published in a book (specifically about the museum collections) citations seems like the appropriate place to but this information (?) I worry that in 30 years the book may be difficult for researchers to locate (obviously the collection will have a copy) but because it is so valuable for the understanding of the object, and not documented anywhere else, it would be great to be able to keep the text associated with the object record.

On Thu, Jan 9, 2020 at 1:59 PM dustymc notifications@github.com wrote:

Yes it's possible. At some point we explicitly decided to force brevity there; I'm not morally opposed to relaxing that, but I'm not entirely sure I see the point of reproducing the book either!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/ArctosDB/arctos/issues/2435?email_source=notifications&email_token=AJKSRR7GA3KDNX2HNILR6OLQ46T6NA5CNFSM4KE7VQVKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEISCYIQ#issuecomment-572795938, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJKSRR4S7POPSZ7FBUJPGDLQ46T6NANCNFSM4KE7VQVA .

-- Mareca Guthrie Curator of Fine Arts & Associate Professor of Art University of Alaska Museum of the North 1962 Yukon Drive P.O. Box 756960 Fairbanks, AK 99775-6960 mrguthrie@alaska.edu

University of Alaska Museum of the North: www.uaf.edu/museum UAF Art Department: https://www.uaf.edu/art/ https://www.uaf.edu/art/ Colors of Nature: http://www.colorsofnature.org/

[Jegelewicz]

"curatorial description" attribute,

Perhaps that is still the solution. Adding the publication PLUS this attribute with the appropriate text. That way, if the text only applies to one part of a cataloged item, it will be associated only with that part.

[dustymc]

How possible/realistic would it be to attach an image of a page or two as Media? I think that's well within fair use.

In any case I agree that remarks concerning the relationship between the cataloged item and a publication belong in citation remarks.

I don't think packaging and display can be completely ignored, but I also think 'proper' data structure should win any conflict. Worse case, we just IF remarks.length > nnn THEN 'see cataloged record for information' ELSE {remarks} END or something.

[marecaguthrie]

We WOULD really love to have a "curatorial remarks" option that would serve to provide context for the artwork-much like an online exhibit label. I'm not sure curatorial remarks is the right term- what we would hope for would look like this example from the Cleveland Art Museum (called "description") https://www.clevelandart.org/art/1964.340#thumbnailonly We don't have these written and this would be a much larger project though.

The reason that I don't think what we were imagining for "curatorial remarks" would be a good fit in this instance is that "curatorial remarks" (or whatever we call it) would just display a single paragraph as a primary interpretation for the object , and in this case this isn't necessarily the text we want we want to display to provide context and background for the artwork. We need this text from Looking North closely associated with the artwork record, but only as one example of what has been written about the object- not necessarily as the definitive interpretation/explanation (although in some cases it might be). And we need to track that it came from a publication important to the collection. In that regard it does seem to be a better fit for citations.

Attaching an image of a page from the book does make a lot of sense - and perhaps that is something we can/should do to as well -but I would argue that this is critical information to understanding/valuing the object and we need the text itself in the catalog record. We did something similar to this (uploaded a pdf) with an artist statement, and it really isn't working for us- we need such important text more easily accessible in the record. This information from an art historian or the artist themselves is in some cases the only time any critical information about the object is documented. increasing character limit would solve our need to track this text with the object record.

I'm wondering now if citations be an appropriate place to put text from artist interviews and even artist statements?

On Fri, Jan 10, 2020 at 8:32 AM dustymc notifications@github.com wrote:

How possible/realistic would it be to attach an image of a page or two as Media? I think that's well within fair use.

In any case I agree that remarks concerning the relationship between the cataloged item and a publication belong in citation remarks.

I don't think packaging and display can be completely ignored, but I also think 'proper' data structure should win any conflict. Worse case, we just IF remarks.length > nnn THEN 'see cataloged record for information' ELSE {remarks} END or something.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/ArctosDB/arctos/issues/2435?email_source=notifications&email_token=AJKSRRZCQHWQ2Q5RPOTFE5DQ5CWLDA5CNFSM4KE7VQVKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEIUUSLA#issuecomment-573131052, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJKSRRYQ3RP3Q3JDIEJMLLDQ5CWLDANCNFSM4KE7VQVA .

-- Mareca Guthrie Curator of Fine Arts & Associate Professor of Art University of Alaska Museum of the North 1962 Yukon Drive P.O. Box 756960 Fairbanks, AK 99775-6960 mrguthrie@alaska.edu

University of Alaska Museum of the North: www.uaf.edu/museum UAF Art Department: https://www.uaf.edu/art/ https://www.uaf.edu/art/ Colors of Nature: http://www.colorsofnature.org/

[Jegelewicz]

I'm wondering now if citations be an appropriate place to put text from artist interviews and even artist statements?

With an attached publication that makes great sense. @dustymc can we in crease the character limit for citation remarks now or is it something that should wait until after PG? I don't see any reason not to do this.

[dustymc]

There was a reason, but I can't remember if it's just appearances or if it breaks something.

One way to find out....

UAM@ARCTOS> alter table citation modify CITATION_REMARKS VARCHAR2(4000);
Table altered.

arctosprod@arctosutf>>  alter table citation alter column citation_remarks type varchar(4000);
ALTER TABLE

[Jegelewicz]

@marecaguthrie can you test this out for us?

[krgomez]

Works great. Thanks!

[marecaguthrie]

I have such a big grin on my face. This is FANTASTIC and makes me feel like we are doing right by the collection! Thank you so much!

On Mon, Jan 13, 2020 at 9:00 AM Karinna Gomez notifications@github.com wrote:

Works great. Thanks!

[image: Screen Shot 2020-01-13 at 8 54 23 AM] https://user-images.githubusercontent.com/40579515/72279316-5ac59280-35e2-11ea-85c4-5ba93c3d8445.png

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ArctosDB/arctos/issues/2435?email_source=notifications&email_token=AJKSRR7BRF3DTZJNDCXEOMDQ5ST4ZA5CNFSM4KE7VQVKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEIZVW3I#issuecomment-573791085, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJKSRRYWCEMM45DTOEBGKQLQ5ST4ZANCNFSM4KE7VQVA .

-- Mareca Guthrie Curator of Fine Arts & Associate Professor of Art University of Alaska Museum of the North 1962 Yukon Drive P.O. Box 756960 Fairbanks, AK 99775-6960 mrguthrie@alaska.edu

University of Alaska Museum of the North: www.uaf.edu/museum UAF Art Department: https://www.uaf.edu/art/ https://www.uaf.edu/art/ Colors of Nature: http://www.colorsofnature.org/

[dustymc]

Mostly unrelated to this, I can't possibly over-emphasize how important normalization is. With that, you've added some information which is ONLY available at the intersection of two data objects (cataloged_item and publication), both of which are only vaguely related to the data object to which that information pertains. In other words, there's about one place (http://arctos.database.museum/guid/UAM:Art:UA1987-003-026) to find that information, and it's fairly obscure. Recorded as metadata of the agent (http://arctos.database.museum/agents.cfm?agent_id=21290198), that information is available from ANY data object which uses the Agent (currently http://arctos.database.museum/info/agentActivity.cfm?agent_id=21290198, but external sources may use it as well).

We can always talk about UI - how and where information is accessed or displayed - or you can build links into text fields eg

See <a href="http://arctos.database.museum/agent.cfm?agent_id=21290198">Henry%20Varnum%20Poor</a> Agent Record for more information.

Is there some way we can make these data a lot more useful, and save you some work in the process?

[marecaguthrie]

I can see how the example Karinna posted would seem appropriate to link to the agent, as it does contain biographical information and context relevant to the other works by Poor in the collection. However, it specifically provides context for this work in particular and the work is visually reproduced adjacent to the caption in the book cited (thus the basis of illustration citation type). While we do actually want to be doing a lot more with the agent profiles (as the people involved are critical to understanding the works themselves), I do think this belongs in the catalog record. In this case I think we would link the citation to both the object and the agent (if that becomes an option?) because having the text in the object record makes the record so much richer and more valuable.

Dusty, I want to be sure I understand your concerns correctly. Are you saying that if citation information about an agent is listed in a citation for an object record it is therefore not readily available when someone looks at the agent?

Would it be possible to add a new author role in publications, possibly called “Subject”? This would give us a way of keeping track of artist bibliographies, artist interviews, exhibition reviews, maybe even artist statements?. Sometimes the Agent is both the subject AND the author and it would be great to be able to make that clear.

Agents are so important to our collection- they are the creators of our objects- and much of the research that is done in the arts centers around the agents as much as it does the objects. We have several things on our “wish list” regarding Agents in order to meet the best practices for core requirements for cataloging works of art (biography, creator description, culture, nationality, gender) that we haven’t posted on GitHub yet. Having citations link to Agents as subjects of the publication would fit beautifully with this.

On Mon, Jan 13, 2020 at 9:15 AM dustymc notifications@github.com wrote:

Reopened #2435 https://github.com/ArctosDB/arctos/issues/2435.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ArctosDB/arctos/issues/2435?email_source=notifications&email_token=AJKSRR4AKAFB5HLLGGQK773Q5SVVFA5CNFSM4KE7VQVKYY3PNVWWK3TUL52HS4DFWZEXG43VMVCXMZLOORHG65DJMZUWGYLUNFXW5KTDN5WW2ZLOORPWSZGOV56YAEA#event-2944237584, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJKSRR3KQ6MTV3J3NO2RRKDQ5SVVFANCNFSM4KE7VQVA .

-- Mareca Guthrie Curator of Fine Arts & Associate Professor of Art University of Alaska Museum of the North 1962 Yukon Drive P.O. Box 756960 Fairbanks, AK 99775-6960 mrguthrie@alaska.edu

University of Alaska Museum of the North: www.uaf.edu/museum UAF Art Department: https://www.uaf.edu/art/ https://www.uaf.edu/art/ Colors of Nature: http://www.colorsofnature.org/

[dustymc]

new author role in publications

Needs an Issue, but seems reasonable-ish to me.

understand your concerns

Essentially I'm just trying to help you get the most benefit for the least work, which generally boils down to recording data as few times and as formally as possible in the most fundamental data object possible. The real power of Arctos is in how it can connect to the world (including the rest of Arctos) and put things in a wider context, but doing so requires the data to support those connections. (It also means that other users, in Arctos and beyond, benefit from your work, and vice-versa.)

As text in a remarks field, "Some Guy wrote two books" is available to whoever finds that remarks field, which is displayed in a very limited context, and that's it. If he writes another book, you'll have to find (which won't be trivial - he'll be listed as 'S. Guy' and "Guy, S." and "Some Gai" and ...) every reference to him and correct them (which would be a tremendous amount of work - enough that it just never happens 'in the wild'). There will never be a way to KNOW we're talking about THAT "Some Guy" - the data will always be ambiguous.

As remarks on the data object that is Some Guy's agent record, "Some Guy wrote two books" is available to anything that uses the agent record (as collector, author, donor, identifier, etc., and perhaps beyond Arctos).

To make corrections, you fix one field and it propagates out to all of that other stuff.

Add a wikidata ID or ORCID to the Agent, and those data become available to anything that can tie into those resources (and we can be absolutely sure which "Some Guy" we're talking about). All of https://www.wikidata.org/wiki/Q3785010 (including all of the stuff linked from that view - https://www.wikitree.com/wiki/Emery-3911 is his grandmother, for example) is available to Arctos via the wikidata agent address, it's just a matter of writing the code to access it.

With the books as publications, you can do all of the above, plus access things like CrossRef to find publications which cited Some Guy's publication and publications cited by Some Guy, and other resources that can be accessed by publication identifiers (mostly DOI at the moment, but no reason we couldn't tie into other systems), etc. "Corrections" handle themselves - just add the third book and count the links when you need to.

The reality of all that's widely variable, but it's getting better fast. For this particular example the DOI system isn't very interesting, but things like nationality and gender (and those data for related people - family, co-authors, etc.) are readily available. I suspect all of those things linked from wikidata would also love to have the context of your catalog record, which in turn would make it easier to document the impact of your collection. Yay everybody!

We need to find a way to fund a 'do magic from wikidata' button!

[Jegelewicz]

We need to find a way to fund a 'do magic from wikidata' button!

AMEN Brother!

[marecaguthrie]

Thanks for going into more detail Dusty. This is fascinating! I am totally unfamiliar with wikidata. I see why this could be powerful.

On Mon, Jan 13, 2020 at 3:35 PM Teresa Mayfield-Meyer < notifications@github.com> wrote:

We need to find a way to fund a 'do magic from wikidata' button!

AMEN Brother!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ArctosDB/arctos/issues/2435?email_source=notifications&email_token=AJKSRR53ETM3V5VBJXC226DQ5UCG3A5CNFSM4KE7VQVKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEI22SOQ#issuecomment-573942074, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJKSRRYWOGHWIPQZNQAKAYTQ5UCG3ANCNFSM4KE7VQVA .

-- Mareca Guthrie Curator of Fine Arts & Associate Professor of Art University of Alaska Museum of the North 1962 Yukon Drive P.O. Box 756960 Fairbanks, AK 99775-6960 mrguthrie@alaska.edu

University of Alaska Museum of the North: www.uaf.edu/museum UAF Art Department: https://www.uaf.edu/art/ https://www.uaf.edu/art/ Colors of Nature: http://www.colorsofnature.org/