search

ArctosDB / arctos

Arctos is a museum collections management system
https://arctos.database.museum
Apache License 2.0
57 stars 13 forks source link

"Links" in records down? #4386

Open SerinaBrady opened 2 years ago

SerinaBrady commented 2 years ago

Hello there,

I have a skunk (ASNHC;Mamm:20741) that has multiple events and parts that correspond to those events. Previously, I have linked the parts to their event in their records. However, I cannot seem to do this anymore and I would very much like to. Any help is appreciated!

Thanks!

Links

dustymc commented 2 years ago

That was removed (the linker, the data are safe) as part of the Organism development. I can bring it back if necessary.

SerinaBrady commented 2 years ago

@campmlc @AdrienneRaniszewski Does MSB still use this for wolves or has the procedure changed?

campmlc commented 2 years ago

@dustymc looks like we've lost the ability to link from parts to events? See https://arctos.database.museum/guid/MSB:Mamm:224611

ewommack commented 2 years ago

@SerinaBrady there is a meeting that is going to address the new way Arctos is handling having this type of relationship. Very welcome to come and we would love to have you there for your info - the Entity Meeting on 14th of March 2022, on the Arctos Calendar. @ccicero and @Jegelewicz

Jegelewicz commented 2 years ago

AWG suggests that me @dustymc @lkvoong work on a clean up and transfer to the entity model.

Jegelewicz commented 2 years ago

I will contact @SerinaBrady and get her started on the entity model, then Lam and Dusty can help undo the old stuff and convert to entity.

Jegelewicz commented 2 years ago

Sent Serina an email today to ask for a meeting.

Jegelewicz commented 2 years ago

Team - I created the additional two records for Mexican Wolf 638 and they include only one event plus appropriate parts:

https://arctos.database.museum/guid/MSB:Mamm:341613 https://arctos.database.museum/guid/MSB:Mamm:341614

but the original record https://arctos.database.museum/guid/MSB:Mamm:231704 still has all of the events and parts because I cannot delete the events/parts that don't go there due to the "event/part links". I also have not created the entity with the entity magic because the complex record won't allow it.

Will need @dustymc to remove the event/part links in that record and then I can finish up. BUT also a few little things:

  1. There are three collectors - do they all belong on every record?
  2. There are three preparators - do they all belong on every record?
  3. https://arctos.database.museum/guid/MSB:Mamm:341614 covers two NK numbers - does that make sense?

Hopefully we can get this done tomorrow.

Jegelewicz commented 2 years ago

BTW - I added the GBIF occurrence IDs to the two new records so that we can try to maintain consistency there. I added them as "other identifier" because I don't want to mess with whatever @dustymc does with "GBIF Occurrence". Will do the same for iDigBio.

dustymc commented 2 years ago

I'll try to allow delete of linked parts later today, if I don't get there remind me and I'll do it manually.

whatever @dustymc does with "GBIF Occurrence".

This? https://arctos.database.museum/info/ctDocumentation.cfm?table=ctcoll_other_id_type#gbif_occurrence_id

If so I have absolutely no idea, all my GBIF links are dynamic, delete it?

Jegelewicz commented 2 years ago

This? https://arctos.database.museum/info/ctDocumentation.cfm?table=ctcoll_other_id_type#gbif_occurrence_id

If so I have absolutely no idea, all my GBIF links are dynamic, delete it?

Yeah - I thought that was there because it needed to be in the code table for you to automate but looking at usage - it is being used to link up with NMNH records.

Also, I'm really sorry, but my password isn't working in test and several requests for a reset haven't sent me any email. I've checked SPAM and still nothing. Any way to get me access? I hate to pester you when I know you need rest!

dustymc commented 2 years ago

I think the proxy setup is blocking email - new_passwerd should work for now.

dustymc commented 2 years ago

@Jegelewicz there's a new button in edit parts (in production)

Jegelewicz commented 2 years ago

Woo hoo! That made it a lot easier to get rid of stuff! But yes, could be dangerous!

Jegelewicz commented 2 years ago

OK, original record fixed up and entity magically created: https://arctos.database.museum/guid/Arctos:Entity:134

Jegelewicz commented 2 years ago

@campmlc @SerinaBrady

Team - I created the additional two records for Mexican Wolf 638 and they include only one event plus appropriate parts:

https://arctos.database.museum/guid/MSB:Mamm:341613 https://arctos.database.museum/guid/MSB:Mamm:341614

and the original record https://arctos.database.museum/guid/MSB:Mamm:231704 has only one event and the appropriate parts.

I used the three records to magically create the entity https://arctos.database.museum/guid/Arctos:Entity:134

But a few open questions:

  1. There are three collectors - do they all belong on every record?
  2. There are three preparators - do they all belong on every record?
  3. https://arctos.database.museum/guid/MSB:Mamm:341614 covers two NK numbers - does that make sense?
dustymc commented 2 years ago

Nice, would be good to get https://github.com/ArctosDB/arctos/issues/4605#issuecomment-1110130471 through if possible, I think it's a lot more clear (and would show the metadata).

Jegelewicz commented 2 years ago

I just added that to today's agenda....

SerinaBrady commented 2 years ago

Hi!

Thank you so much, it looks great!

1.) I think all three collectors are valid for each record. 2.) I would say that the preparators would only be on the final preparation (skull, etc.) 3.) I think 2 NK numbers make sense/are okay.

Serina

From: Teresa Mayfield-Meyer @.> Sent: Wednesday, May 4, 2022 7:36 PM To: ArctosDB/arctos @.> Cc: Serina Brady @.>; Mention @.> Subject: Re: [ArctosDB/arctos] "Links" in records down? (Issue #4386)

ATTENTION: This email originated from outside of Angelo State University. Do not click on links or open attachments unless you recognize the sender and know the content is safe.

@campmlchttps://urldefense.com/v3/__https://github.com/campmlc__;!!KGfBWX4!qJQh3Sdokr3qTJ60Tny_5hn4OqCttUDVoqkonaiuZxqej-DFxmkAR8mqwD__syv2q4c-QKnx56SKE1zDgEJ0h7Du0diRUg$ @SerinaBradyhttps://urldefense.com/v3/__https://github.com/SerinaBrady__;!!KGfBWX4!qJQh3Sdokr3qTJ60Tny_5hn4OqCttUDVoqkonaiuZxqej-DFxmkAR8mqwD__syv2q4c-QKnx56SKE1zDgEJ0h7COfymKZQ$

Team - I created the additional two records for Mexican Wolf 638 and they include only one event plus appropriate parts:

https://arctos.database.museum/guid/MSB:Mamm:341613https://urldefense.com/v3/__https://arctos.database.museum/guid/MSB:Mamm:341613__;!!KGfBWX4!qJQh3Sdokr3qTJ60Tny_5hn4OqCttUDVoqkonaiuZxqej-DFxmkAR8mqwD__syv2q4c-QKnx56SKE1zDgEJ0h7D_UlTVxw$ https://arctos.database.museum/guid/MSB:Mamm:341614https://urldefense.com/v3/__https://arctos.database.museum/guid/MSB:Mamm:341614__;!!KGfBWX4!qJQh3Sdokr3qTJ60Tny_5hn4OqCttUDVoqkonaiuZxqej-DFxmkAR8mqwD__syv2q4c-QKnx56SKE1zDgEJ0h7DR5i1t_g$

and the original record https://arctos.database.museum/guid/MSB:Mamm:231704https://urldefense.com/v3/__https://arctos.database.museum/guid/MSB:Mamm:231704__;!!KGfBWX4!qJQh3Sdokr3qTJ60Tny_5hn4OqCttUDVoqkonaiuZxqej-DFxmkAR8mqwD__syv2q4c-QKnx56SKE1zDgEJ0h7ALmH8J6Q$ has only one event and the appropriate parts.

I used the three records to magically create the entity https://arctos.database.museum/guid/Arctos:Entity:134https://urldefense.com/v3/__https://arctos.database.museum/guid/Arctos:Entity:134__;!!KGfBWX4!qJQh3Sdokr3qTJ60Tny_5hn4OqCttUDVoqkonaiuZxqej-DFxmkAR8mqwD__syv2q4c-QKnx56SKE1zDgEJ0h7BuULNsGQ$

But a few open questions:

  1. There are three collectors - do they all belong on every record?
  2. There are three preparators - do they all belong on every record?
  3. https://arctos.database.museum/guid/MSB:Mamm:341614https://urldefense.com/v3/__https://arctos.database.museum/guid/MSB:Mamm:341614__;!!KGfBWX4!qJQh3Sdokr3qTJ60Tny_5hn4OqCttUDVoqkonaiuZxqej-DFxmkAR8mqwD__syv2q4c-QKnx56SKE1zDgEJ0h7DR5i1t_g$ covers two NK numbers - does that make sense?

— Reply to this email directly, view it on GitHubhttps://urldefense.com/v3/__https://github.com/ArctosDB/arctos/issues/4386*issuecomment-1118059420__;Iw!!KGfBWX4!qJQh3Sdokr3qTJ60Tny_5hn4OqCttUDVoqkonaiuZxqej-DFxmkAR8mqwD__syv2q4c-QKnx56SKE1zDgEJ0h7BdEefnlg$, or unsubscribehttps://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AUR7UZX7YCJQFAMNFRYLTIDVIMJZJANCNFSM5PUJAEHA__;!!KGfBWX4!qJQh3Sdokr3qTJ60Tny_5hn4OqCttUDVoqkonaiuZxqej-DFxmkAR8mqwD__syv2q4c-QKnx56SKE1zDgEJ0h7Bev8Eg_Q$. You are receiving this because you were mentioned.Message ID: @.***>

Jegelewicz commented 2 years ago

I'll let y'all take it from here. let me know if I can help with anything else!

campmlc commented 1 year ago

This looks great! Cheryl Parmenter should be the only preparator on all three records - the other two should only be on the collection event. I removed them. I removed Sevilleta -not the right project. NK 261319 doesn't link to anything anywhere - I'll check with the paper copies to confirm. These records should only have one NK each - because the NK corresponds to the collecting event.

On Wed, May 4, 2022 at 6:36 PM Teresa Mayfield-Meyer < @.***> wrote:

  • [EXTERNAL]*

@campmlc https://github.com/campmlc @SerinaBrady https://github.com/SerinaBrady

Team - I created the additional two records for Mexican Wolf 638 and they include only one event plus appropriate parts:

https://arctos.database.museum/guid/MSB:Mamm:341613 https://arctos.database.museum/guid/MSB:Mamm:341614

and the original record https://arctos.database.museum/guid/MSB:Mamm:231704 has only one event and the appropriate parts.

I used the three records to magically create the entity https://arctos.database.museum/guid/Arctos:Entity:134

But a few open questions:

  1. There are three collectors - do they all belong on every record?
  2. There are three preparators - do they all belong on every record?
  3. https://arctos.database.museum/guid/MSB:Mamm:341614 covers two NK numbers - does that make sense?

— Reply to this email directly, view it on GitHub https://github.com/ArctosDB/arctos/issues/4386#issuecomment-1118059420, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADQ7JBCGI2ZIFLW5MTIYIJTVIMJZJANCNFSM5PUJAEHA . You are receiving this because you were mentioned.Message ID: @.***>

Jegelewicz commented 8 months ago

I think this is done? Closing.

dustymc commented 8 months ago

Did you make entity-linked records or ??? Just curious...

Here's the raw data, Just In Case.

specimen_event_links.csv.zip

EDIT: Actually I'm going to reopen, that's still keyed in, please confirm that I can drop that table.

Jegelewicz commented 8 months ago

Did you make entity-linked records or ???

Nope - Any chance you can also provide GUIDs with that data? This was so long ago - I don't know if I obligated my self to fix everything or just the one thing requested initially. If I know which records need work, I can at least approach the collections and try to get things cleaned up.

dustymc commented 8 months ago

also provide GUIDs

temp_specimen_event_links_lnks.csv.zip

Jegelewicz commented 8 months ago

OK - that is over 500 records that need to be looked at! I am guessing it would take me a full-time month to work through them, create entities or additional records as needed and get it all corrected.

Also, how would I correct these? UTEP has one record with links (and I think they were made in error), but I can see no way to get rid of them.

https://arctos.database.museum/guid/UTEP:Herp:19805

dustymc commented 8 months ago

would I correct these?

Split records and add Entities, but I think that probably requires some sort of communication with the collections.

made in error

That would definitely involve communication with the collection - minting forever-identifiers for mistakes wouldn't be awesome.

Jegelewicz commented 8 months ago

That would definitely involve communication with the collection - minting forever-identifiers for mistakes wouldn't be awesome.

That one only has one event (really) and I think they misunderstood why event linking was used. I wouldn't be creating an entity for it.

campmlc commented 4 months ago

These are all MSB Mexican wolves, from when we attempted a new system, with @dustymc to deal with repeat samples from the same individual by moving all events and parts to a single catalog record. We had a master's student do her entire thesis project on this before we realized the system of multiple events linked to parts in a single record really didn't work. So we need to figure out how to undo a lot of these. Yes, it will take dedicated time, which likely will require grant funding and/or dedicated personnel. So, can we hold off for now? I don't know an easy way forward, but we can meet to discuss? Thanks for grabbing the data!

campmlc commented 4 months ago

@SerinaBrady is no longer using Arctos, so this is now an MSB problem.

jldunnum commented 4 months ago

Can't determine what exactly is being proposed in this string but as Mariel states, if it has to do with our very complex Mexican wolf issue please do nothing that will compound the problem or hamper tracking past modifications to those records. Don't have bandwidth to take this on currently. Thanks

dustymc commented 4 months ago

e all MSB Mexican wolves

 guid_prefix |                                scientific_name                                | count 
-------------+-------------------------------------------------------------------------------+-------
 BYU:Herp    | Ambystoma mavortium                                                           |     7
 BYU:Herp    | Caecilia tentaculata                                                          |     1
 BYU:Herp    | Gegeneophis ramaswamii                                                        |     2
 KNWR:Herb   | Betula kenaica                                                                |     3
 KNWR:Herb   | Betula pendula                                                                |     3
 MLZ:Bird    | Anas diazi                                                                    |     2
 MLZ:Bird    | Aphelocoma californica oocleptica / Aphelocoma californica nevadae intergrade |     2
 MSB:Mamm    | Ammospermophilus leucurus                                                     |     7
 MSB:Mamm    | Canis lupus baileyi                                                           |  5924
 MSB:Mamm    | Dendrolagus matschiei                                                         |    10
 MSB:Mamm    | Macropus rufus                                                                |     5
 MSB:Mamm    | Microtus pennsylvanicus                                                       |     2
 MSB:Mamm    | Ovis aries                                                                    |    13
 MSB:Mamm    | Peromyscus maniculatus                                                        |     4
 MSB:Mamm    | Puma concolor                                                                 |     9
 MSB:Mamm    | Tamias minimus                                                                |    10
 MSB:Mamm    | Uncia uncia                                                                   |    21
 UAM:Mamm    | Myodes rutilus                                                                |     2
 UTEP:Herp   | Pantherophis emoryi ?                                                         |     2

There's also full data above.

what exactly is being proposed

I'm not proposing anything, I'm asking for help. I have data from an apparently-failed experiment. I'm pretty sure nobody other than me can get at it. I need to do SOMETHING with it.

jldunnum commented 4 months ago

Im not sure what to do at the moment either. This is a result of us trying to work with Arctos moving to an event based model. Now here are with many events languishing inside single records.

campmlc commented 4 months ago

We tried multiple events per record and that didn't work. For one, we can't add more than one accession to a record, but these wolves have multiple accessions over their lifetimes. For another, the number of parts becomes unmanageable over the years, which is why we attempted to link parts to events, and we also recorded all the event info again in part remarks. But then the data entry became so convoluted that it could take an hour to figure out how to enter a single record, and required separate training and workflow from regular data entry. No bulk tools are available. Now we can use Entities as a method to link all indivduals through a shared organism ID, which means we can go back to entering these types of samples using our normal methods, with one catalog number per sampling event. But we still have these records that were converted to the multiple event model left in limbo, or at least, the linkages between parts and events in these multi-event records no longer works. That's not a deal-breaker - we can figure out parts to events because we duplicated all the event info in part remarks. But it's still not ideal.