Actor-resource relationship (Bug?)

[virtu...@yahoo.com.br]

Hello everyone,

My name is Carlos and im the chief librarian at the Sao Paulo State Archives (Brazil). Using Atom sience first release. Great stuff and staff.

Using now version 2.4.1 and found something odd.
By disabling automatic insertion of producers as name (subject), all non-producers and non-subject names were excluded from main isad form, ending up hidden in the sidebar at the right.

Was this an option or we have a bug? Looks like a bug to me because, on the contrary, dates for all actors-resources relationship are, as expected, included in dates field with its proper qualifiers.

If it is a bug, my suggestion is the following:

1) automatically insert all names at access point in isad form, with its proper qualifiers, as it was before. Remove only the auto-generaton of subjects for producers, as done recently. By removing this feature, it seems that something more was lost.

2) insert all names (but producers and repositories) in the names box-filter. Otherwise all non-producers, non-repositories and non-subject names, i.e., all names involved in other actors-resources relationship, would remanin invisible in filters option as well.

Best wishes!
From Brazil,
Carlos Henrique Menegozzo

[Dan Gillean]

Hi Carlos, 

This is interesting, because I believe you are describing behaviors that you can only see if you are using multiple edit templates for data entry - for example, creating data using the RAD template (so you can add other actor relations, such as manufacturer, broadcaster, etc), but then changing the template to ISAD for display. Is this correct?

I suspect this because of the differences between the two standards, and how they are implemented in AtoM, based on the ICA's original requests. Some background: 

When we were first developing ICA-AtoM, we were collaborating with the ICA to implement ISAD(G) support as they wanted to see it, but also working with Canadian institutions to implement RAD support. We proposed to the ICA that we use an Events table in the data model to relate creation/modification/accumulation/etc events with an actor, a date, and if desired, a place - after all, creation does not happen out of thin air, generally you need someone who is doing the creating. This also worked well with the RAD standard, which groups creation dates in the same area as name of publisher/distributor/etc, place of publication/distribution/etc... 

In the ISAD(G) standard, Dates of creation (3.1.3, in the Identity statements area) and Name of creator(s) (3.2.1, in the Context area) are not related at all, and appear in different sections of the standard. The ICA did not want to see an Events module implemented in AtoM similar to the modal we use in the edit interface of AtoM's RAD template, where the two are related - however, we still needed a way to implement support for both ISAD, RAD, and future standards. 

In the end, we are using an Events table in the database to manage this information, but the user interface (in both the view and edit templates) obfuscates this for ISAD(G). When you create a description with dates and a creator in AtoM's ISAD template, in the database, 2 events are added - one with a creator but no dates, and a second one with dates but no creator. When using the RAD template, you have one row added to the Events table with both pieces of information. 

To get more specific about your comments: 

ISAD(G) only supports two event types - Creation, and Accumulation. It does not include additional relations (such as collection, distribution, modification, manufacturing, publication, broadcast, etc) that are included in the RAD template. Because dates and actors are separated in the standard, there is also no way in AtoM's ISAD template to directly associate an accumulation event with an actor. This means that if you are only using the ISAD template, there is nothing that is hidden from the body of the record. An example record I created in ISAD: 

If you are only using the RAD template, it's true that actors associated with other event types are not hyperlinked in the body of the record - but as you noted, they are available in the sidebar (AKA the right-hand context menu), and their names appear in the Dates of creation area associated with their events: 

It is only if you are using the templates in an unconventional way - i.e. creating your data in RAD, but then changing the display to ISAD, a completely different standard with different rules - that you end up with additional actor events only listed in the sidebar. For example, I took my modified record above and then flipped it back to ISAD for display: 

Now I can see what you mean - the information is present, but not in the body of the record - the names are only listed in the sidebar. Is this what you are doing?

I have previously discussed why we chose to remove the automatic addition of the creator name as a name (subject) access point in the following forum thread: 

• https://groups.google.com/d/msg/ica-atom-users/cV3LUvRyIPY/G0j2gZGvBgAJ
  

I think that the feedback we got from our community about this (mainly, that there are many cases where the Creator is not the subject of the records being described, and this shouldn't be automatically added) was valid, and I don't think we'd want to change this back. 

However, I think we could potentially implement your suggestions without undoing the change in 2.4.1. 

I understand what you are suggesting with 1), but can you clarify what you mean by 2) please? Do you mean the names facet on the search/browse page?

Once I have a bit more clarity, I would be happy to file an issue ticket. However, since this change would provide a fix that is mostly only useful for edge cases (i.e. those combining the RAD and ISAD templates), I think it is unlikely we would address it without community support, either via a community code contribution (i.e. pull request), or sponsorship to implement the fix. If this is a priority for your institution to address and you might like to fix it or sponsor us fixing it, please feel free to contact me off-list so we can discuss the matter further. 

Regards, 

Dan Gillean, MAS, MLIS
AtoM Program Manager
Artefactual Systems, Inc.
604-527-2056

@accesstomemory

--
You received this message because you are subscribed to the Google Groups "AtoM Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ica-atom-user...@googlegroups.com.
To post to this group, send email to ica-ato...@googlegroups.com.
Visit this group at https://groups.google.com/group/ica-atom-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/ica-atom-users/7f84aadb-fe31-4eea-9693-f15f3dce18a1%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[virtu...@yahoo.com.br]

Hello Dan. 

 

Thank you very much for your answer. And I apologize for my english. Let me try to clarify the issue.

DAN: “I think that the feedback we got from our community about this (mainly, that there are many cases where the Creator is not the subject of the records being described, and this shouldn't be automatically added) was valid, and I don't think we'd want to change this back. ” 

-- I totally support this modification. God bless you guys.

DAN: “However, I think we could potentially implement your suggestions without undoing the change in 2.4.1.”

-- Exacltly. And it does not runs against ISAD. According to the rules “access points are based upon the elements of description” (ISAD 1.14), and no standards are prescribed other than ISAAR-CPF for authority records. I may add mentions of responsability other than accumulation/production/creation in many fields of description. If the event is presented in the description, the name and its proper qualifier are allowed as access points. If I am not mistaken, an earlier version of Atom worked that way (names implicated in Events endding up as access points with qualifiers). Now I realize that this was a possible case of a bug that a loved.

DAN: “I understand what you are suggesting with 1), but can you clarify what you mean by 2) please? Do you mean the names facet on the search/browse page?”

-- I think I do. I was talking about those drop boxes that appear on the left side while browsing archival records. One of them is “Names”. For names there are also “Archival institution” and “Creator”. All names are there? Seems to, but they are not. Subjects are, accumulators/producers/creators too. What about names implicated in Events other than these? My suggestion is to add them there, with the subject names all together.

In sum, my suggestions are:

1) First, add all names implicated in Events (other than accumulation/creation/production) as access point in ISAD description template. The rules allow it, since the Event information is presented in the body of the description (rule 1.14).

2) The second is a consequence of that: add every name (excepct accumulators/creators/producers) as a browse option in the "Names" drop box (wich in present version includes only names as subjects).

I have some other observations about “access point” and “keywords” in Atom that this discussion brings, but I you leave them for that other and specific thread you mentioned recently.

Regards,

Carlos

[virtu...@yahoo.com.br]

Sorry, forgot one aspect:

DAN: "This is interesting, because I believe you are describing behaviors that you can only see if you are using multiple edit templates for data entry - for example, creating data using the RAD template (so you can add other actor relations, such as manufacturer, broadcaster, etc), but then changing the template to ISAD for display. Is this correct?"

-- Yes and no. This result can be achived either by RAD CSV import, later converted into ISAD ; or by adding an actor-resource relationship to a ISAD description via user interface. Until auto-generation of subject names for creator was disabled in 2.4.1, these relations were ending up as access points in ISAD descriptions. I thought it was a fature designed inentionally,because it is consistent with ISAD rules. Now i realize that maybe that was the bug. It vanished when the auto-generation of subject names for creator was disabled.

If that was a bug unintentionally corrected, then here is a third suggestion. ISAD csv template for information object could support events, like RAD template. But the its display in ISAD description template would be restricted to access points area. This possibility would be perfect, specially for professionals that, like me, work rigth on the frontiers between archives and libraries.

Regards,
Carlos

[virtu...@yahoo.com.br]

Last addition: the same result with events and ISAD descriptions could be achieved in three ways.

1-- Via RAD CSV import with later conversion to ISAD;

2-- Manually via user interface;

3-- via Event CSV import combined with ISAD CSV template.

[Dan Gillean]

Hi Carlos, 

Thank you for this detailed response! And for reminding me of the other methods of adding these relations - I had temporarily forgotten!

I have attempted to capture our conversation in two issue tickets I've just filed. 

The first is a bug report ticket, based on your proposals 1 and 2: 

• https://projects.artefactual.com/issues/12718
  

I have also added this to an internal list I maintain of community-reported bugs for us to attempt to address in the next release. With each release, our AtoM team has a limited budget of time to address unsponsored reported bugs, which we try to prioritize based on severity. Inclusion in this list does not guarantee it will be fixed in the next release, but it will be on our radar and we'll get to it if we can. If this is a priority issue for you, please consider having your institution sponsor the work involved. 

The second ticket I've created is a Wish list ticket for your proposal #3: 

• https://projects.artefactual.com/issues/12719
  

Note that this is in fact more complex to implement than it seems at first, because name access points are not currently treated as events in the database. Adding the ability to define a type (and create an event relation) for name access points involves a signficant number of database and form changes - I've tried to add some thoughts about this on the related ticket. 

This would definitely require sponsorship for us to be able to implement. We rely on our community to support feature enhancements and major changes. In case you are not familiar with our development and maintenance model for AtoM, please see the following page with more details: 

• https://wiki.accesstomemory.org/Development/Philosophy
  

Thank you again for this report, and for your patience and attention to detail! Let me know if you feel I have missed something, or misunderstood, in the related tickets. 

Regards, 

Dan Gillean, MAS, MLIS
AtoM Program Manager
Artefactual Systems, Inc.
604-527-2056

@accesstomemory

--
You received this message because you are subscribed to the Google Groups "AtoM Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ica-atom-user...@googlegroups.com.
To post to this group, send email to ica-ato...@googlegroups.com.
Visit this group at https://groups.google.com/group/ica-atom-users.

To view this discussion on the web visit https://groups.google.com/d/msgid/ica-atom-users/cfbc57c5-0b9b-43f6-b739-02b7abcddf30%40googlegroups.com.

[virtu...@yahoo.com.br]

Hi Dan
Perfect! Thanks for your attention!

One last clarification. Seems to me that one aspect can be done in a rather simple way.

By "qualification" i dont mean anything but the event itself. For exemple: Oxford University Press (Publication), General Motors (Manufecturing) etc. That is something that **Atom was already doing in earlier versions**.

For me, Event csv import combined with ISAD csv import will be fine. The core issue: just reestablish behavior of access points in isad descption **as it was in earlier versions, except the auto-subject for creators**. Although not elegant, this feature would do the trick.

Great team you guys have there!
Cheers,
Carlos

[Dan Gillean]

Hi Carlos, 

Thanks for the clarification. Based on the way I have described the issue, I think if we are able to fix bug #12718 then the previous qualifications will be included when non-creation events are shown in the view page, under name access points. 

With this fix, you will still need to manage the events via import as you mention, or via editing the related Actor relationship area (i.e. these name access points will only appear in the view page of the description, not in the edit page, unless you are using the RAD template - and then they will be shown in the Dates of creation area). But I think that such a fix is possible, and should provide you with everything you are expecting. 

Cheers, 

Dan Gillean, MAS, MLIS
AtoM Program Manager
Artefactual Systems, Inc.
604-527-2056

@accesstomemory

--
You received this message because you are subscribed to the Google Groups "AtoM Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ica-atom-user...@googlegroups.com.
To post to this group, send email to ica-ato...@googlegroups.com.
Visit this group at https://groups.google.com/group/ica-atom-users.

To view this discussion on the web visit https://groups.google.com/d/msgid/ica-atom-users/b68d88df-ef43-4f81-b089-ac2b082d65c6%40googlegroups.com.

[virtu...@yahoo.com.br]

Thats it. Thank you very much!