talk@lists.collectionspace.org

WE HAVE SUNSET THIS LISTSERV - Join us at collectionspace@lyrasislists.org

View all threads

Upgrading Location and Movement

MF
Megan Forbes
Thu, Jan 19, 2012 3:31 PM

CSpace Implementers:

We've gotten a lot of feedback about the current implementation of location
and movement, so have made upgrading this functionality a priority for the
next release. Based on feedback, user stories, and use cases, Angela and I
developed two alternative implementation options for your consideration and
comments. You are also welcome to respond to this email with new
suggestions for implementation that aren't covered by our proposed
solutions.

Option I

Separate the current location and movement tab into two separate tabs, one
called Location and one called Movement. The new tab would have the same
functionality as all current tabs.

There are both pros and cons to this approach; for example, it would get us
out of the current situation where an object or group of objects must be
"moved" in order to be given an updated location. It would unfortunately
not solve the problem of a user being able to inadvertently change the
current location of a group of objects, if s/he did not know that the
location record were tied to multiple objects.

Option 2

Separate the current location and movement tab into two separate services.
Movement would remain a top-level service with its own tab. Location would
be modeled on the sub-record approach we have taken with Contact, where the
Location information group would be its own service, able to be appended to
any procedure (e.g. Cataloging, Intake, Acquisition, etc.).

On the plus side, this approach would provide a clearer link between an
object (or procedure) and its location. We would likely need to couple this
approach with a new UI for batch processing, so that we didn't lose the
ability to make changes in current location to more than one object at a
time. The sub-record model may cause some difficulties with report writing
or import/export, as the information group is not part of the record but
related to it.

The current wireframe for Acquisition shows how this approach would look,
with location as a repeatable group within the procedural record:
http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition
.

Additional Ideas

It has been suggested that making the location information group - whether
as a separate procedure or as a separate service - "write once" would help
to mitigate the potential for data or audit trail losses that are possible
with the current implementation and both new options. In other words, once
that data is filled out and the record is saved, it can no longer be
edited. To add a new location, update the current location, fix a typo,
etc., a new location record or group of fields would need to be created. We
are interested in hearing your thoughts on adding this piece of
functionality to either of the new options or to the existing functionality.

Feedback
**
Please submit your comments, critiques, questions, or alternative solutions
to the talk list by Friday, January 27th. I created a wiki page that
includes a description of the current implementation, both options, and a
few sample workflows to show how these new approaches differ:
http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade.
We do also have a wiki page for location and movement use cases - if you
have a use case that isn't helped by either the current or proposed
options, please post it!
http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases
.

Thanks everyone,
Megan

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800
Direct 718 777 6834

CSpace Implementers: We've gotten a lot of feedback about the current implementation of location and movement, so have made upgrading this functionality a priority for the next release. Based on feedback, user stories, and use cases, Angela and I developed two alternative implementation options for your consideration and comments. You are also welcome to respond to this email with new suggestions for implementation that aren't covered by our proposed solutions. *Option I* Separate the current location and movement tab into two separate tabs, one called Location and one called Movement. The new tab would have the same functionality as all current tabs. There are both pros and cons to this approach; for example, it would get us out of the current situation where an object or group of objects must be "moved" in order to be given an updated location. It would unfortunately not solve the problem of a user being able to inadvertently change the current location of a group of objects, if s/he did not know that the location record were tied to multiple objects. *Option 2* Separate the current location and movement tab into two separate services. Movement would remain a top-level service with its own tab. Location would be modeled on the sub-record approach we have taken with Contact, where the Location information group would be its own service, able to be appended to any procedure (e.g. Cataloging, Intake, Acquisition, etc.). On the plus side, this approach would provide a clearer link between an object (or procedure) and its location. We would likely need to couple this approach with a new UI for batch processing, so that we didn't lose the ability to make changes in current location to more than one object at a time. The sub-record model may cause some difficulties with report writing or import/export, as the information group is not part of the record but related to it. The current wireframe for Acquisition shows how this approach would look, with location as a repeatable group within the procedural record: http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition . *Additional Ideas* It has been suggested that making the location information group - whether as a separate procedure or as a separate service - "write once" would help to mitigate the potential for data or audit trail losses that are possible with the current implementation and both new options. In other words, once that data is filled out and the record is saved, it can no longer be edited. To add a new location, update the current location, fix a typo, etc., a new location record or group of fields would need to be created. We are interested in hearing your thoughts on adding this piece of functionality to either of the new options or to the existing functionality. *Feedback* ** Please submit your comments, critiques, questions, or alternative solutions to the talk list by Friday, January 27th. I created a wiki page that includes a description of the current implementation, both options, and a few sample workflows to show how these new approaches differ: http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade. We do also have a wiki page for location and movement use cases - if you have a use case that isn't helped by either the current or proposed options, please post it! http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases . Thanks everyone, Megan -- Megan Forbes Collection Manager Museum of the Moving Image 36-01 35 Avenue Astoria, NY 11106 movingimage.us 718 777 6800 Direct 718 777 6834
MT
Michael T. Black
Tue, Jan 24, 2012 1:59 AM

Hi Megan,

Option 2 sounds like the better/safer of the two options for us.  PAHMA would need to be able to support multiple current locations (the wireframe at http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition makes it appear that it could do so by having Location be a repeatable section), and to see (at least the most recent) previous locations.  Presumably this information would be in the right-hand column under related procedures?

The "write once" functionality for locations sounded great to our registrars.

Also, a request from our registrars and collections managers: They wonder if it would be possible to implement the concept of "active" and "inactive" location records?  The point is to hide inactive locations and to only show active locations.  In general, current locations would be considered active and previous locations (and erroneously created location records, as in your "write once" example) would be considered inactive.  Without this, some objects (i.e., frequently moved/exhibited/lent objects) would be drowning in a sea of visible location and/or movement records.  Ideally, the current location(s) and perhaps the immediately previous location(s) would be active/visible, and others/older location records would require more work (e.g., running a report) to get at.

Michael

On Jan 19, 2012, at 7:31 AM, Megan Forbes wrote:

CSpace Implementers:

We've gotten a lot of feedback about the current implementation of location and movement, so have made upgrading this functionality a priority for the next release. Based on feedback, user stories, and use cases, Angela and I developed two alternative implementation options for your consideration and comments. You are also welcome to respond to this email with new suggestions for implementation that aren't covered by our proposed solutions.

Option I

Separate the current location and movement tab into two separate tabs, one called Location and one called Movement. The new tab would have the same functionality as all current tabs.

There are both pros and cons to this approach; for example, it would get us out of the current situation where an object or group of objects must be "moved" in order to be given an updated location. It would unfortunately not solve the problem of a user being able to inadvertently change the current location of a group of objects, if s/he did not know that the location record were tied to multiple objects.

Option 2

Separate the current location and movement tab into two separate services. Movement would remain a top-level service with its own tab. Location would be modeled on the sub-record approach we have taken with Contact, where the Location information group would be its own service, able to be appended to any procedure (e.g. Cataloging, Intake, Acquisition, etc.).

On the plus side, this approach would provide a clearer link between an object (or procedure) and its location. We would likely need to couple this approach with a new UI for batch processing, so that we didn't lose the ability to make changes in current location to more than one object at a time. The sub-record model may cause some difficulties with report writing or import/export, as the information group is not part of the record but related to it.

The current wireframe for Acquisition shows how this approach would look, with location as a repeatable group within the procedural record: http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition.

Additional Ideas

It has been suggested that making the location information group - whether as a separate procedure or as a separate service - "write once" would help to mitigate the potential for data or audit trail losses that are possible with the current implementation and both new options. In other words, once that data is filled out and the record is saved, it can no longer be edited. To add a new location, update the current location, fix a typo, etc., a new location record or group of fields would need to be created. We are interested in hearing your thoughts on adding this piece of functionality to either of the new options or to the existing functionality.

Feedback

Please submit your comments, critiques, questions, or alternative solutions to the talk list by Friday, January 27th. I created a wiki page that includes a description of the current implementation, both options, and a few sample workflows to show how these new approaches differ: http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade. We do also have a wiki page for location and movement use cases - if you have a use case that isn't helped by either the current or proposed options, please post it! http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases.

Thanks everyone,
Megan

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800
Direct 718 777 6834


Work mailing list
Work@lists.collectionspace.org
http://lists.collectionspace.org/mailman/listinfo/work_lists.collectionspace.org

Hi Megan, Option 2 sounds like the better/safer of the two options for us. PAHMA would need to be able to support multiple current locations (the wireframe at http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition makes it appear that it could do so by having Location be a repeatable section), and to see (at least the most recent) previous locations. Presumably this information would be in the right-hand column under related procedures? The "write once" functionality for locations sounded great to our registrars. Also, a request from our registrars and collections managers: They wonder if it would be possible to implement the concept of "active" and "inactive" location records? The point is to hide inactive locations and to only show active locations. In general, current locations would be considered active and previous locations (and erroneously created location records, as in your "write once" example) would be considered inactive. Without this, some objects (i.e., frequently moved/exhibited/lent objects) would be drowning in a sea of visible location and/or movement records. Ideally, the current location(s) and perhaps the immediately previous location(s) would be active/visible, and others/older location records would require more work (e.g., running a report) to get at. Michael On Jan 19, 2012, at 7:31 AM, Megan Forbes wrote: > CSpace Implementers: > > We've gotten a lot of feedback about the current implementation of location and movement, so have made upgrading this functionality a priority for the next release. Based on feedback, user stories, and use cases, Angela and I developed two alternative implementation options for your consideration and comments. You are also welcome to respond to this email with new suggestions for implementation that aren't covered by our proposed solutions. > > Option I > > Separate the current location and movement tab into two separate tabs, one called Location and one called Movement. The new tab would have the same functionality as all current tabs. > > There are both pros and cons to this approach; for example, it would get us out of the current situation where an object or group of objects must be "moved" in order to be given an updated location. It would unfortunately not solve the problem of a user being able to inadvertently change the current location of a group of objects, if s/he did not know that the location record were tied to multiple objects. > > Option 2 > > Separate the current location and movement tab into two separate services. Movement would remain a top-level service with its own tab. Location would be modeled on the sub-record approach we have taken with Contact, where the Location information group would be its own service, able to be appended to any procedure (e.g. Cataloging, Intake, Acquisition, etc.). > > On the plus side, this approach would provide a clearer link between an object (or procedure) and its location. We would likely need to couple this approach with a new UI for batch processing, so that we didn't lose the ability to make changes in current location to more than one object at a time. The sub-record model may cause some difficulties with report writing or import/export, as the information group is not part of the record but related to it. > > The current wireframe for Acquisition shows how this approach would look, with location as a repeatable group within the procedural record: http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition. > > Additional Ideas > > It has been suggested that making the location information group - whether as a separate procedure or as a separate service - "write once" would help to mitigate the potential for data or audit trail losses that are possible with the current implementation and both new options. In other words, once that data is filled out and the record is saved, it can no longer be edited. To add a new location, update the current location, fix a typo, etc., a new location record or group of fields would need to be created. We are interested in hearing your thoughts on adding this piece of functionality to either of the new options or to the existing functionality. > > Feedback > > Please submit your comments, critiques, questions, or alternative solutions to the talk list by Friday, January 27th. I created a wiki page that includes a description of the current implementation, both options, and a few sample workflows to show how these new approaches differ: http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade. We do also have a wiki page for location and movement use cases - if you have a use case that isn't helped by either the current or proposed options, please post it! http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases. > > > Thanks everyone, > Megan > > > > -- > Megan Forbes > Collection Manager > > Museum of the Moving Image > 36-01 35 Avenue Astoria, NY 11106 > movingimage.us 718 777 6800 > Direct 718 777 6834 > > > _______________________________________________ > Work mailing list > Work@lists.collectionspace.org > http://lists.collectionspace.org/mailman/listinfo/work_lists.collectionspace.org
PS
Patrick Schmitz
Tue, Jan 24, 2012 7:12 PM

Note: we have the notion in all our authorities of items that are inactive,
and these (I think) are already being filtered in calls for term-completion.

In addition, we can fairly easily define a new set of states for any
document type (including the Location records). We currently have most of
the records in a simple set that includes active and "deleted", and we
filter the deleted ones in the normal UI (but allow access with reports, as
you suggest). We just need some UI to mark the Location record as
"inactive". Or, you can just delete them since we only soft-delete them
anyway.

Patrick


From: work-bounces@lists.collectionspace.org
[mailto:work-bounces@lists.collectionspace.org] On Behalf Of Michael T.
Black
Sent: Monday, January 23, 2012 5:59 PM
To: Megan Forbes
Cc: CollectionSpace Talk List; CollectionSpace Work list
Subject: Re: [Work] Upgrading Location and Movement

Hi Megan,

Option 2 sounds like the better/safer of the two options for us.  PAHMA
would need to be able to support multiple current locations (the wireframe
at
http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisi
tion makes it appear that it could do so by having Location be a repeatable
section), and to see (at least the most recent) previous locations.
Presumably this information would be in the right-hand column under related
procedures?

The "write once" functionality for locations sounded great to our
registrars.

Also, a request from our registrars and collections managers: They wonder if
it would be possible to implement the concept of "active" and "inactive"
location records?  The point is to hide inactive locations and to only show
active locations.  In general, current locations would be considered active
and previous locations (and erroneously created location records, as in your
"write once" example) would be considered inactive.  Without this, some
objects (i.e., frequently moved/exhibited/lent objects) would be drowning in
a sea of visible location and/or movement records.  Ideally, the current
location(s) and perhaps the immediately previous location(s) would be
active/visible, and others/older location records would require more work
(e.g., running a report) to get at.

Michael

On Jan 19, 2012, at 7:31 AM, Megan Forbes wrote:

CSpace Implementers:

We've gotten a lot of feedback about the current implementation of location
and movement, so have made upgrading this functionality a priority for the
next release. Based on feedback, user stories, and use cases, Angela and I
developed two alternative implementation options for your consideration and
comments. You are also welcome to respond to this email with new suggestions
for implementation that aren't covered by our proposed solutions.

Option I

Separate the current location and movement tab into two separate tabs, one
called Location and one called Movement. The new tab would have the same
functionality as all current tabs.

There are both pros and cons to this approach; for example, it would get us
out of the current situation where an object or group of objects must be
"moved" in order to be given an updated location. It would unfortunately not
solve the problem of a user being able to inadvertently change the current
location of a group of objects, if s/he did not know that the location
record were tied to multiple objects.

Option 2

Separate the current location and movement tab into two separate services.
Movement would remain a top-level service with its own tab. Location would
be modeled on the sub-record approach we have taken with Contact, where the
Location information group would be its own service, able to be appended to
any procedure (e.g. Cataloging, Intake, Acquisition, etc.).

On the plus side, this approach would provide a clearer link between an
object (or procedure) and its location. We would likely need to couple this
approach with a new UI for batch processing, so that we didn't lose the
ability to make changes in current location to more than one object at a
time. The sub-record model may cause some difficulties with report writing
or import/export, as the information group is not part of the record but
related to it.

The current wireframe for Acquisition shows how this approach would look,
with location as a repeatable group within the procedural record:
http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisi
tion.

Additional Ideas

It has been suggested that making the location information group - whether
as a separate procedure or as a separate service - "write once" would help
to mitigate the potential for data or audit trail losses that are possible
with the current implementation and both new options. In other words, once
that data is filled out and the record is saved, it can no longer be edited.
To add a new location, update the current location, fix a typo, etc., a new
location record or group of fields would need to be created. We are
interested in hearing your thoughts on adding this piece of functionality to
either of the new options or to the existing functionality.

Feedback

Please submit your comments, critiques, questions, or alternative solutions
to the talk list by Friday, January 27th. I created a wiki page that
includes a description of the current implementation, both options, and a
few sample workflows to show how these new approaches differ:
http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+a
nd+Movement+Upgrade. We do also have a wiki page for location and movement
use cases - if you have a use case that isn't helped by either the current
or proposed options, please post it!
http://wiki.collectionspace.org/display/collectionspace/Location+and+Movemen
t+Control+Use+Cases.

Thanks everyone,
Megan

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us http://movingimage.us/  718 777 6800
tel:718%20777%206800
Direct 718 777  tel:718%20777%206834 6834


Work mailing list
Work@lists.collectionspace.org
http://lists.collectionspace.org/mailman/listinfo/work_lists.collectionspace
.org

Note: we have the notion in all our authorities of items that are inactive, and these (I think) are already being filtered in calls for term-completion. In addition, we can fairly easily define a new set of states for any document type (including the Location records). We currently have most of the records in a simple set that includes active and "deleted", and we filter the deleted ones in the normal UI (but allow access with reports, as you suggest). We just need some UI to mark the Location record as "inactive". Or, you can just delete them since we only soft-delete them anyway. Patrick _____ From: work-bounces@lists.collectionspace.org [mailto:work-bounces@lists.collectionspace.org] On Behalf Of Michael T. Black Sent: Monday, January 23, 2012 5:59 PM To: Megan Forbes Cc: CollectionSpace Talk List; CollectionSpace Work list Subject: Re: [Work] Upgrading Location and Movement Hi Megan, Option 2 sounds like the better/safer of the two options for us. PAHMA would need to be able to support multiple current locations (the wireframe at http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisi tion makes it appear that it could do so by having Location be a repeatable section), and to see (at least the most recent) previous locations. Presumably this information would be in the right-hand column under related procedures? The "write once" functionality for locations sounded great to our registrars. Also, a request from our registrars and collections managers: They wonder if it would be possible to implement the concept of "active" and "inactive" location records? The point is to hide inactive locations and to only show active locations. In general, current locations would be considered active and previous locations (and erroneously created location records, as in your "write once" example) would be considered inactive. Without this, some objects (i.e., frequently moved/exhibited/lent objects) would be drowning in a sea of visible location and/or movement records. Ideally, the current location(s) and perhaps the immediately previous location(s) would be active/visible, and others/older location records would require more work (e.g., running a report) to get at. Michael On Jan 19, 2012, at 7:31 AM, Megan Forbes wrote: CSpace Implementers: We've gotten a lot of feedback about the current implementation of location and movement, so have made upgrading this functionality a priority for the next release. Based on feedback, user stories, and use cases, Angela and I developed two alternative implementation options for your consideration and comments. You are also welcome to respond to this email with new suggestions for implementation that aren't covered by our proposed solutions. Option I Separate the current location and movement tab into two separate tabs, one called Location and one called Movement. The new tab would have the same functionality as all current tabs. There are both pros and cons to this approach; for example, it would get us out of the current situation where an object or group of objects must be "moved" in order to be given an updated location. It would unfortunately not solve the problem of a user being able to inadvertently change the current location of a group of objects, if s/he did not know that the location record were tied to multiple objects. Option 2 Separate the current location and movement tab into two separate services. Movement would remain a top-level service with its own tab. Location would be modeled on the sub-record approach we have taken with Contact, where the Location information group would be its own service, able to be appended to any procedure (e.g. Cataloging, Intake, Acquisition, etc.). On the plus side, this approach would provide a clearer link between an object (or procedure) and its location. We would likely need to couple this approach with a new UI for batch processing, so that we didn't lose the ability to make changes in current location to more than one object at a time. The sub-record model may cause some difficulties with report writing or import/export, as the information group is not part of the record but related to it. The current wireframe for Acquisition shows how this approach would look, with location as a repeatable group within the procedural record: http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisi tion. Additional Ideas It has been suggested that making the location information group - whether as a separate procedure or as a separate service - "write once" would help to mitigate the potential for data or audit trail losses that are possible with the current implementation and both new options. In other words, once that data is filled out and the record is saved, it can no longer be edited. To add a new location, update the current location, fix a typo, etc., a new location record or group of fields would need to be created. We are interested in hearing your thoughts on adding this piece of functionality to either of the new options or to the existing functionality. Feedback Please submit your comments, critiques, questions, or alternative solutions to the talk list by Friday, January 27th. I created a wiki page that includes a description of the current implementation, both options, and a few sample workflows to show how these new approaches differ: http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+a nd+Movement+Upgrade. We do also have a wiki page for location and movement use cases - if you have a use case that isn't helped by either the current or proposed options, please post it! http://wiki.collectionspace.org/display/collectionspace/Location+and+Movemen t+Control+Use+Cases. Thanks everyone, Megan -- Megan Forbes Collection Manager Museum of the Moving Image 36-01 35 Avenue Astoria, NY 11106 movingimage.us <http://movingimage.us/> 718 777 6800 <tel:718%20777%206800> Direct 718 777 <tel:718%20777%206834> 6834 _______________________________________________ Work mailing list Work@lists.collectionspace.org http://lists.collectionspace.org/mailman/listinfo/work_lists.collectionspace .org
MF
Megan Forbes
Tue, Jan 24, 2012 7:21 PM

Thanks to those who have responded to this email. If you haven't yet,
please read the email, visit the wiki page, and let me know your feedback
by the end of this week. If you have any questions, let me know and we can
set up a time to chat via IRC, Skype, Connect, or whatever works for you!

Thanks,
Megan

---------- Forwarded message ----------
From: Megan Forbes mforbes@movingimage.us
Date: Thu, Jan 19, 2012 at 10:31 AM
Subject: Upgrading Location and Movement
To: CollectionSpace Talk List talk@lists.collectionspace.org
Cc: CollectionSpace Work list work@lists.collectionspace.org

CSpace Implementers:

We've gotten a lot of feedback about the current implementation of location
and movement, so have made upgrading this functionality a priority for the
next release. Based on feedback, user stories, and use cases, Angela and I
developed two alternative implementation options for your consideration and
comments. You are also welcome to respond to this email with new
suggestions for implementation that aren't covered by our proposed
solutions.

Option I

Separate the current location and movement tab into two separate tabs, one
called Location and one called Movement. The new tab would have the same
functionality as all current tabs.

There are both pros and cons to this approach; for example, it would get us
out of the current situation where an object or group of objects must be
"moved" in order to be given an updated location. It would unfortunately
not solve the problem of a user being able to inadvertently change the
current location of a group of objects, if s/he did not know that the
location record were tied to multiple objects.

Option 2

Separate the current location and movement tab into two separate services.
Movement would remain a top-level service with its own tab. Location would
be modeled on the sub-record approach we have taken with Contact, where the
Location information group would be its own service, able to be appended to
any procedure (e.g. Cataloging, Intake, Acquisition, etc.).

On the plus side, this approach would provide a clearer link between an
object (or procedure) and its location. We would likely need to couple this
approach with a new UI for batch processing, so that we didn't lose the
ability to make changes in current location to more than one object at a
time. The sub-record model may cause some difficulties with report writing
or import/export, as the information group is not part of the record but
related to it.

The current wireframe for Acquisition shows how this approach would look,
with location as a repeatable group within the procedural record:
http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition
.

Additional Ideas

It has been suggested that making the location information group - whether
as a separate procedure or as a separate service - "write once" would help
to mitigate the potential for data or audit trail losses that are possible
with the current implementation and both new options. In other words, once
that data is filled out and the record is saved, it can no longer be
edited. To add a new location, update the current location, fix a typo,
etc., a new location record or group of fields would need to be created. We
are interested in hearing your thoughts on adding this piece of
functionality to either of the new options or to the existing functionality.

Feedback
**
Please submit your comments, critiques, questions, or alternative solutions
to the talk list by Friday, January 27th. I created a wiki page that
includes a description of the current implementation, both options, and a
few sample workflows to show how these new approaches differ:
http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade.
We do also have a wiki page for location and movement use cases - if you
have a use case that isn't helped by either the current or proposed
options, please post it!
http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases
.

Thanks everyone,
Megan

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800
Direct 718 777 6834

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800
Direct 718 777 6834

Thanks to those who have responded to this email. If you haven't yet, please read the email, visit the wiki page, and let me know your feedback by the end of this week. If you have any questions, let me know and we can set up a time to chat via IRC, Skype, Connect, or whatever works for you! Thanks, Megan ---------- Forwarded message ---------- From: Megan Forbes <mforbes@movingimage.us> Date: Thu, Jan 19, 2012 at 10:31 AM Subject: Upgrading Location and Movement To: CollectionSpace Talk List <talk@lists.collectionspace.org> Cc: CollectionSpace Work list <work@lists.collectionspace.org> CSpace Implementers: We've gotten a lot of feedback about the current implementation of location and movement, so have made upgrading this functionality a priority for the next release. Based on feedback, user stories, and use cases, Angela and I developed two alternative implementation options for your consideration and comments. You are also welcome to respond to this email with new suggestions for implementation that aren't covered by our proposed solutions. *Option I* Separate the current location and movement tab into two separate tabs, one called Location and one called Movement. The new tab would have the same functionality as all current tabs. There are both pros and cons to this approach; for example, it would get us out of the current situation where an object or group of objects must be "moved" in order to be given an updated location. It would unfortunately not solve the problem of a user being able to inadvertently change the current location of a group of objects, if s/he did not know that the location record were tied to multiple objects. *Option 2* Separate the current location and movement tab into two separate services. Movement would remain a top-level service with its own tab. Location would be modeled on the sub-record approach we have taken with Contact, where the Location information group would be its own service, able to be appended to any procedure (e.g. Cataloging, Intake, Acquisition, etc.). On the plus side, this approach would provide a clearer link between an object (or procedure) and its location. We would likely need to couple this approach with a new UI for batch processing, so that we didn't lose the ability to make changes in current location to more than one object at a time. The sub-record model may cause some difficulties with report writing or import/export, as the information group is not part of the record but related to it. The current wireframe for Acquisition shows how this approach would look, with location as a repeatable group within the procedural record: http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition . *Additional Ideas* It has been suggested that making the location information group - whether as a separate procedure or as a separate service - "write once" would help to mitigate the potential for data or audit trail losses that are possible with the current implementation and both new options. In other words, once that data is filled out and the record is saved, it can no longer be edited. To add a new location, update the current location, fix a typo, etc., a new location record or group of fields would need to be created. We are interested in hearing your thoughts on adding this piece of functionality to either of the new options or to the existing functionality. *Feedback* ** Please submit your comments, critiques, questions, or alternative solutions to the talk list by Friday, January 27th. I created a wiki page that includes a description of the current implementation, both options, and a few sample workflows to show how these new approaches differ: http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade. We do also have a wiki page for location and movement use cases - if you have a use case that isn't helped by either the current or proposed options, please post it! http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases . Thanks everyone, Megan -- Megan Forbes Collection Manager Museum of the Moving Image 36-01 35 Avenue Astoria, NY 11106 movingimage.us 718 777 6800 Direct 718 777 6834 -- Megan Forbes Collection Manager Museum of the Moving Image 36-01 35 Avenue Astoria, NY 11106 movingimage.us 718 777 6800 Direct 718 777 6834
KB
Kim Brasen
Thu, Jan 26, 2012 8:52 AM

Hi Megan

As Michael T. Black and PAHMA we would probably prefer option 2. And as PAHMA we need several current locations i.e. future locations. We also support the suggestion of active/inactive locations, and the notion of inactive locations being previous locations (i.e. locations with an end date).

The "write once" solution seems mandatory to us in order to avoid any confusion.

When a work of art is on loan to another institution or organization we record the Current Location as the name of the institution or organization in question. This means that we either need two separate Current Location fields - internal/external, or the Current Location field to be fed from both the Storage Location Authority and the Organization Name Authority if possible.

We have a couple of extensions that might be of interest to other implementers. We need a Location Type (controlled list: deposit, exhibition, office, permanent display, storage, loan out, etc.). Today we also record whether a location is connected to an exhibition (whether organized by us or another institution), so we have en Exhibition field fed by an Exhibition Authority (I'm fully aware that this will probably remain a SMK specific extension).

Regarding the Movement Procedure we also have a number of additions that other implementers might like to take into consideration. We need a Destination field fed by the Storage Location Authority. In order to plan future movements we suggest addition of a field for Movement Order Date (calendar date), and the Contact field as a repeatable group with a Contact > Role added to it.

Cheers,

Kim

Best regards,

Kim Brasen

Curator

T +45 3374 8460

Statens Museum for Kunst

Sølvgade 48-50

DK-1307 København K

T +45 3374 8494

F +45 3374 8404

smk.dk http://smk.dk/


Fra: talk-bounces@lists.collectionspace.org [mailto:talk-bounces@lists.collectionspace.org] På vegne af Megan Forbes
Sendt: 19. januar 2012 16:31
Til: CollectionSpace Talk List
Cc: CollectionSpace Work list
Emne: [Talk] Upgrading Location and Movement

CSpace Implementers:

We've gotten a lot of feedback about the current implementation of location and movement, so have made upgrading this functionality a priority for the next release. Based on feedback, user stories, and use cases, Angela and I developed two alternative implementation options for your consideration and comments. You are also welcome to respond to this email with new suggestions for implementation that aren't covered by our proposed solutions.

Option I

Separate the current location and movement tab into two separate tabs, one called Location and one called Movement. The new tab would have the same functionality as all current tabs.

There are both pros and cons to this approach; for example, it would get us out of the current situation where an object or group of objects must be "moved" in order to be given an updated location. It would unfortunately not solve the problem of a user being able to inadvertently change the current location of a group of objects, if s/he did not know that the location record were tied to multiple objects.

Option 2

Separate the current location and movement tab into two separate services. Movement would remain a top-level service with its own tab. Location would be modeled on the sub-record approach we have taken with Contact, where the Location information group would be its own service, able to be appended to any procedure (e.g. Cataloging, Intake, Acquisition, etc.).

On the plus side, this approach would provide a clearer link between an object (or procedure) and its location. We would likely need to couple this approach with a new UI for batch processing, so that we didn't lose the ability to make changes in current location to more than one object at a time. The sub-record model may cause some difficulties with report writing or import/export, as the information group is not part of the record but related to it.

The current wireframe for Acquisition shows how this approach would look, with location as a repeatable group within the procedural record: http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition.

Additional Ideas

It has been suggested that making the location information group - whether as a separate procedure or as a separate service - "write once" would help to mitigate the potential for data or audit trail losses that are possible with the current implementation and both new options. In other words, once that data is filled out and the record is saved, it can no longer be edited. To add a new location, update the current location, fix a typo, etc., a new location record or group of fields would need to be created. We are interested in hearing your thoughts on adding this piece of functionality to either of the new options or to the existing functionality.

Feedback

Please submit your comments, critiques, questions, or alternative solutions to the talk list by Friday, January 27th. I created a wiki page that includes a description of the current implementation, both options, and a few sample workflows to show how these new approaches differ: http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade. We do also have a wiki page for location and movement use cases - if you have a use case that isn't helped by either the current or proposed options, please post it! http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases.

Thanks everyone,
Megan

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800 tel:718%20777%206800
Direct 718 777 6834 tel:718%20777%206834

Hi Megan As Michael T. Black and PAHMA we would probably prefer option 2. And as PAHMA we need several current locations i.e. future locations. We also support the suggestion of active/inactive locations, and the notion of inactive locations being previous locations (i.e. locations with an end date). The "write once" solution seems mandatory to us in order to avoid any confusion. When a work of art is on loan to another institution or organization we record the Current Location as the name of the institution or organization in question. This means that we either need two separate Current Location fields - internal/external, or the Current Location field to be fed from both the Storage Location Authority and the Organization Name Authority if possible. We have a couple of extensions that might be of interest to other implementers. We need a Location Type (controlled list: deposit, exhibition, office, permanent display, storage, loan out, etc.). Today we also record whether a location is connected to an exhibition (whether organized by us or another institution), so we have en Exhibition field fed by an Exhibition Authority (I'm fully aware that this will probably remain a SMK specific extension). Regarding the Movement Procedure we also have a number of additions that other implementers might like to take into consideration. We need a Destination field fed by the Storage Location Authority. In order to plan future movements we suggest addition of a field for Movement Order Date (calendar date), and the Contact field as a repeatable group with a Contact > Role added to it. Cheers, Kim Best regards, Kim Brasen Curator T +45 3374 8460 Statens Museum for Kunst Sølvgade 48-50 DK-1307 København K T +45 3374 8494 F +45 3374 8404 smk.dk <http://smk.dk/> ________________________________ Fra: talk-bounces@lists.collectionspace.org [mailto:talk-bounces@lists.collectionspace.org] På vegne af Megan Forbes Sendt: 19. januar 2012 16:31 Til: CollectionSpace Talk List Cc: CollectionSpace Work list Emne: [Talk] Upgrading Location and Movement CSpace Implementers: We've gotten a lot of feedback about the current implementation of location and movement, so have made upgrading this functionality a priority for the next release. Based on feedback, user stories, and use cases, Angela and I developed two alternative implementation options for your consideration and comments. You are also welcome to respond to this email with new suggestions for implementation that aren't covered by our proposed solutions. Option I Separate the current location and movement tab into two separate tabs, one called Location and one called Movement. The new tab would have the same functionality as all current tabs. There are both pros and cons to this approach; for example, it would get us out of the current situation where an object or group of objects must be "moved" in order to be given an updated location. It would unfortunately not solve the problem of a user being able to inadvertently change the current location of a group of objects, if s/he did not know that the location record were tied to multiple objects. Option 2 Separate the current location and movement tab into two separate services. Movement would remain a top-level service with its own tab. Location would be modeled on the sub-record approach we have taken with Contact, where the Location information group would be its own service, able to be appended to any procedure (e.g. Cataloging, Intake, Acquisition, etc.). On the plus side, this approach would provide a clearer link between an object (or procedure) and its location. We would likely need to couple this approach with a new UI for batch processing, so that we didn't lose the ability to make changes in current location to more than one object at a time. The sub-record model may cause some difficulties with report writing or import/export, as the information group is not part of the record but related to it. The current wireframe for Acquisition shows how this approach would look, with location as a repeatable group within the procedural record: http://wiki.collectionspace.org/display/collectionspace/Wireframes+-+Acquisition. Additional Ideas It has been suggested that making the location information group - whether as a separate procedure or as a separate service - "write once" would help to mitigate the potential for data or audit trail losses that are possible with the current implementation and both new options. In other words, once that data is filled out and the record is saved, it can no longer be edited. To add a new location, update the current location, fix a typo, etc., a new location record or group of fields would need to be created. We are interested in hearing your thoughts on adding this piece of functionality to either of the new options or to the existing functionality. Feedback Please submit your comments, critiques, questions, or alternative solutions to the talk list by Friday, January 27th. I created a wiki page that includes a description of the current implementation, both options, and a few sample workflows to show how these new approaches differ: http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade. We do also have a wiki page for location and movement use cases - if you have a use case that isn't helped by either the current or proposed options, please post it! http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases. Thanks everyone, Megan -- Megan Forbes Collection Manager Museum of the Moving Image 36-01 35 Avenue Astoria, NY 11106 movingimage.us 718 777 6800 <tel:718%20777%206800> Direct 718 777 6834 <tel:718%20777%206834>
J
jblowe@berkeley.edu
Fri, Jan 27, 2012 12:11 AM

Megan, et al.,

Thanks for the opporunity to comment on changes to this functionality.

The alas rather lengthy response here grew from recent discussions with
the UCB deployment team working on PAHMA and other CSpace folks.  While
the use cases and points made below are relevant to the PAHMA
implementation, a review of wiki pages indicates that most if not all of
these are germane to other institutions, so I apologize in advance if the
points and use cases at times redundant with remarks already made wrt
other deployments; nevertheless, I have at times mentioned other use cases
in the discussion below to butress my points.

And parenthetically, let me say that despite the fact that this is my
first month as a CSpace deployer, I fully expect to be held to answer for
any errors, omissions, and misrepresentations in these remarks. Please
point out any faults you find! And I apologize again in advance for the
length of this post. (Je n'ai fait celle-ci plus longue que parce que je
n'ai pas eu le loisir de la faire plus courte. -- Pascal)

WIKI PAGES REFERRED TO

Current proposal for changes, contains links to current and proposed
workflows

http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade

A Requirements page, perhaps now outdated, describing how locations and
and the history of an object's location are to be handled.

http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Requirements

Use cases and other "on the ground" writings about locations and movements.

http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases

http://wiki.collectionspace.org/display/deploy/PAHMA+Storage+Location+data+mapping

http://wiki.collectionspace.org/display/deploy/PAHMA+Customizations+-+Storage+Authority

https://wiki.collectionspace.org:8444/display/deploy/Crates+and+other+mobile+containers

REMARKS AND OBSERVATIONS

While it does seem clear that some revisions will be needed to the way
Locations and Movements are handled, some of the changes suggested seem
rather drastic while others seem not to go far enough. Some of the issues
identified appear to be merely semantic, conceptual, or ontological (and
therefore may simply require a change in thinking and documentation),
while others are engineering conundrums which (may) require changes at
every level of the software, schema to UI). I am not sure which is which!

The notion that "the current situation where an object or group of objects
must be 'move' in order to be given an updated location" represents a bug
to be fixed is a misconceptualization. It is in fact a desired feature
whose utility deserves clarification. Some examples:

  1. Moving an object from one location to the same location is either a
    data entry mistake or an inventory action.  That is, the Inventory
    process can viewed as a sub-case of Movement. (This is how it works at
    PAHMA in TMS now).  Therefore, all location updates are indeed moves!
    (Aside: from an implementation point of view, perhaps all movements merit
    a separate attribute on the Movement record to indicate their "moveType",
    to facilitate UI filtering and reporting.)

  2. Both Objects and Locations can move.  It is important not to conflate
    these two cases.

2a. On the one hand, moving a bunch of objects that are all in the same
location to another location could be conceptualized as just that: a bunch
of objects moving. However, if in reality it is the location that is
moving (a tray, or a crate), then perhaps this is best thought of (and
perhaps implemented as) a change to one Location and not as a change to
multiple Objects.

Indeed, it is unclear to me how CSpace implements "Movements of
Locations", if it handles them at all.

2b. Thinking about this is complicated by the fact that Movable Locations
have a number of potential important properties that Immovable Locations
do not have, and these distinctions may need to be captured.  Subcases of
Movable Locations include not only the "crate use case" and "specimen tray
use case" (not the same, I think), but alsot Transitory Locations (which
exist only for a little while, e.g. "this crate is now on Moving Van #7
and enroute", "this package was picked up by UPS, tracking number
1Z23456789") and Fuzzy Locations (which are underspecified or even
completely unspecified e.g. "all the crates in this room will be moving
tomorrow to our Storage Facility, we just don't know which room", "the
specimens on these trays on Rack 26 are gonna need to get cleared out and
go someplace refrigerated soon!").

  1. The constraints on Object Movement are of course different than those
    for Location Movement. A short list of these:

3a. Movable Locations can move to most any location (movable or not,
though real-world constraints certainly exist), but Fixed Locations cannot
move.  The case of a Movable Location moving to another Movable (or
Transitory) Location is usually either a "nesting move" (a container into
another container) or an instance of "typical move", where objects are
being moved en masse for exhibition, restoration, etc.

3b. Changing a Fixed Location (in the Storage Location Authority) may
entail a number of real-world and databases changes, ergo the idea of
having "write once" location storage records seems like quite a good one.
In PAHMA's case, a location's displayName will be barcoded so updating its
record in the database implies reprinting and distributing barcode labels.
In PAHMA's case, therefore "crates", if represented as Locations, will
have to have immutable displayNames. This in turn creates the
uncomfortable conundrum that the displayName in CSpace as shown in the UI
will not show the actual location in PAHMA's location hierarchy. Not sure
what to do about this yet!

3c. Object movements are generally simple or atomic, though in the case
where an object is at a Movable Location a UI caution about moving the
other objects is probably warranted.

  1. Parenthetically, it's been lamented that a Storage Location authority
    lacks a suitable hierarchical model. Some of the location movement issues
    mentioned above might be addressed with such a model.  A Storage
    Location Authority is a sub-type of a Concept Authority, with (it seems)
    some additional properties. At PAHMA, the 5 levels of the "location
    hierarchy" are quite parallel to the 7 ranks of Biological Classification.
    Any CSpace implementation supporting the biological or geographical
    domains would be welcome in the location domain.

(A complication is that the values used for each rank in a location
hierarchy is not guaranteed to be unique: the string "Cordata" only occurs
in the rank of phylum, under kingdom "Animalia"; but container "Vault 15"
may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12" of
site "Hearst Gym".)

UPSHOT

The upshot of this, presented tentatively and with humility, is:

  • Movement and Location Authority records deserve some set of attributes
    (term statuses?) in the core implementation to facilitate some
    as-yet-not-completely-understood-or-implemented features. Examples of
    these attributes might include (for Storage Locations) "active/inactive",
    "movable/fixed", "transitory/permanent", etc.  Patrick noted that some/all
    of this functionality exists for authorities. For movements, statuses or
    attributes might include "planned/completed", "inventory/reallymoving",
    etc. (Can a term in an authority have several statuses?)

  • Best Practices and a few "worked examples" might go a long way towards
    clarifying what can and can't be done and how. Wiki pages covering cases
    like "How to Move Containers like Crates and Trays", "Moving Multiple
    Objects", "Moving Objects When You Don't Know Where They Are Going",
    "Updating a Location Authority Record: Making Sure It Works", etc. remain
    to be defined.  The workflows proposed may in fact cover the required
    cases, but a warm and fuzzy sense of completeness is lacking.

  • At PAHMA, we are considering whether a generalization of one of the
    "regular" Movement workflows (i.e. "location update via barcode scanner")
    might not be the way to actually accomplish the migration of existing
    object location data to CSpace.  Reactions from other users about the
    efficacy and wisdom of this would be most welcome!

Hope this helps!

John B. Lowe, PhD

"Wherever you go, there you are" -- DR Davenport

Megan, et al., Thanks for the opporunity to comment on changes to this functionality. The alas rather lengthy response here grew from recent discussions with the UCB deployment team working on PAHMA and other CSpace folks. While the use cases and points made below are relevant to the PAHMA implementation, a review of wiki pages indicates that most if not all of these are germane to other institutions, so I apologize in advance if the points and use cases at times redundant with remarks already made wrt other deployments; nevertheless, I have at times mentioned other use cases in the discussion below to butress my points. And parenthetically, let me say that despite the fact that this is my first month as a CSpace deployer, I fully expect to be held to answer for any errors, omissions, and misrepresentations in these remarks. Please point out any faults you find! And I apologize again in advance for the length of this post. (Je n'ai fait celle-ci plus longue que parce que je n'ai pas eu le loisir de la faire plus courte. -- Pascal) WIKI PAGES REFERRED TO Current proposal for changes, contains links to current and proposed workflows http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade A Requirements page, perhaps now outdated, describing how locations and and the history of an object's location are to be handled. http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Requirements Use cases and other "on the ground" writings about locations and movements. http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases http://wiki.collectionspace.org/display/deploy/PAHMA+Storage+Location+data+mapping http://wiki.collectionspace.org/display/deploy/PAHMA+Customizations+-+Storage+Authority https://wiki.collectionspace.org:8444/display/deploy/Crates+and+other+mobile+containers REMARKS AND OBSERVATIONS While it does seem clear that some revisions will be needed to the way Locations and Movements are handled, some of the changes suggested seem rather drastic while others seem not to go far enough. Some of the issues identified appear to be *merely* semantic, conceptual, or ontological (and therefore may *simply* require a change in thinking and documentation), while others are engineering conundrums which (may) require changes at every level of the software, schema to UI). I am not sure which is which! The notion that "the current situation where an object or group of objects must be 'move' in order to be given an updated location" represents a bug to be fixed is a misconceptualization. It is in fact a desired feature whose utility deserves clarification. Some examples: 1. Moving an object from one location to the same location is either a data entry mistake *or* an inventory action. That is, the Inventory process can viewed as a sub-case of Movement. (This is how it works at PAHMA in TMS now). Therefore, all location updates are indeed moves! (Aside: from an implementation point of view, perhaps all movements merit a separate attribute on the Movement record to indicate their "moveType", to facilitate UI filtering and reporting.) 2. Both Objects and Locations can move. It is important not to conflate these two cases. 2a. On the one hand, moving a bunch of objects that are all in the same location to another location could be conceptualized as just that: a bunch of objects moving. However, if in reality it is the *location* that is moving (a tray, or a crate), then perhaps this is best thought of (and perhaps implemented as) a change to one Location and not as a change to multiple Objects. Indeed, it is unclear to me how CSpace implements "Movements of Locations", if it handles them at all. 2b. Thinking about this is complicated by the fact that Movable Locations have a number of potential important properties that Immovable Locations do not have, and these distinctions may need to be captured. Subcases of Movable Locations include not only the "crate use case" and "specimen tray use case" (not the same, I think), but alsot Transitory Locations (which exist only for a little while, e.g. "this crate is now on Moving Van #7 and enroute", "this package was picked up by UPS, tracking number 1Z23456789") and Fuzzy Locations (which are underspecified or even completely unspecified e.g. "all the crates in this room will be moving tomorrow to our Storage Facility, we just don't know which room", "the specimens on these trays on Rack 26 are gonna need to get cleared out and go someplace refrigerated soon!"). 3. The constraints on Object Movement are of course different than those for Location Movement. A short list of these: 3a. Movable Locations can move to most any location (movable or not, though real-world constraints certainly exist), but Fixed Locations cannot move. The case of a Movable Location moving to another Movable (or Transitory) Location is usually either a "nesting move" (a container into another container) or an instance of "typical move", where objects are being moved en masse for exhibition, restoration, etc. 3b. Changing a Fixed Location (in the Storage Location Authority) may entail a number of real-world and databases changes, ergo the idea of having "write once" location storage records seems like quite a good one. In PAHMA's case, a location's displayName will be barcoded so updating its record in the database implies reprinting and distributing barcode labels. In PAHMA's case, therefore "crates", if represented as Locations, will have to have immutable displayNames. This in turn creates the uncomfortable conundrum that the displayName in CSpace as shown in the UI will not show the actual location in PAHMA's location hierarchy. Not sure what to do about this yet! 3c. Object movements are generally simple or atomic, though in the case where an object is at a Movable Location a UI caution about moving the other objects is probably warranted. 4. Parenthetically, it's been lamented that a Storage Location authority lacks a suitable hierarchical model. Some of the location movement issues mentioned above *might* be addressed with such a model. A Storage Location Authority is a sub-type of a Concept Authority, with (it seems) some additional properties. At PAHMA, the 5 levels of the "location hierarchy" are quite parallel to the 7 ranks of Biological Classification. Any CSpace implementation supporting the biological or geographical domains would be welcome in the location domain. (A complication is that the values used for each rank in a location hierarchy is not guaranteed to be unique: the string "Cordata" only occurs in the rank of phylum, under kingdom "Animalia"; but container "Vault 15" may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12" of site "Hearst Gym".) UPSHOT The upshot of this, presented tentatively and with humility, is: * Movement and Location Authority records deserve some set of attributes (term statuses?) in the core implementation to facilitate some as-yet-not-completely-understood-or-implemented features. Examples of these attributes might include (for Storage Locations) "active/inactive", "movable/fixed", "transitory/permanent", etc. Patrick noted that some/all of this functionality exists for authorities. For movements, statuses or attributes might include "planned/completed", "inventory/reallymoving", etc. (Can a term in an authority have several statuses?) * Best Practices and a few "worked examples" might go a long way towards clarifying what can and can't be done and how. Wiki pages covering cases like "How to Move Containers like Crates and Trays", "Moving Multiple Objects", "Moving Objects When You Don't Know Where They Are Going", "Updating a Location Authority Record: Making Sure It Works", etc. remain to be defined. The workflows proposed may in fact cover the required cases, but a warm and fuzzy sense of completeness is lacking. * At PAHMA, we are considering whether a generalization of one of the "regular" Movement workflows (i.e. "location update via barcode scanner") might not be the way to actually accomplish the migration of existing object location data to CSpace. Reactions from other users about the efficacy and wisdom of this would be most welcome! Hope this helps! John B. Lowe, PhD "Wherever you go, there you are" -- DR Davenport
MF
Megan Forbes
Fri, Jan 27, 2012 5:12 PM

Thank you for the feedback, John, and welcome to the team!

I think there are more issues in the below than can be comfortably dealt
with over email - how about an open Skype or Connect session with any
interested parties? I could do anytime in the afternoon on either Monday or
Tuesday next week. If anyone from SMK would like to attend, we'd have to go
earlier, more like 8PST/11EST, which I could do on Tuesday or Wednesday of
next week.

Please let me know what works for you.

Best regards,
Megan

On Thu, Jan 26, 2012 at 7:11 PM, jblowe@berkeley.edu wrote:

Megan, et al.,

Thanks for the opporunity to comment on changes to this functionality.

The alas rather lengthy response here grew from recent discussions with
the UCB deployment team working on PAHMA and other CSpace folks.  While
the use cases and points made below are relevant to the PAHMA
implementation, a review of wiki pages indicates that most if not all of
these are germane to other institutions, so I apologize in advance if the
points and use cases at times redundant with remarks already made wrt
other deployments; nevertheless, I have at times mentioned other use cases
in the discussion below to butress my points.

And parenthetically, let me say that despite the fact that this is my
first month as a CSpace deployer, I fully expect to be held to answer for
any errors, omissions, and misrepresentations in these remarks. Please
point out any faults you find! And I apologize again in advance for the
length of this post. (Je n'ai fait celle-ci plus longue que parce que je
n'ai pas eu le loisir de la faire plus courte. -- Pascal)

WIKI PAGES REFERRED TO

Current proposal for changes, contains links to current and proposed
workflows

http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade

A Requirements page, perhaps now outdated, describing how locations and
and the history of an object's location are to be handled.

http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Requirements

Use cases and other "on the ground" writings about locations and movements.

http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases

http://wiki.collectionspace.org/display/deploy/PAHMA+Storage+Location+data+mapping

http://wiki.collectionspace.org/display/deploy/PAHMA+Customizations+-+Storage+Authority

https://wiki.collectionspace.org:8444/display/deploy/Crates+and+other+mobile+containers

REMARKS AND OBSERVATIONS

While it does seem clear that some revisions will be needed to the way
Locations and Movements are handled, some of the changes suggested seem
rather drastic while others seem not to go far enough. Some of the issues
identified appear to be merely semantic, conceptual, or ontological (and
therefore may simply require a change in thinking and documentation),
while others are engineering conundrums which (may) require changes at
every level of the software, schema to UI). I am not sure which is which!

The notion that "the current situation where an object or group of objects
must be 'move' in order to be given an updated location" represents a bug
to be fixed is a misconceptualization. It is in fact a desired feature
whose utility deserves clarification. Some examples:

  1. Moving an object from one location to the same location is either a
    data entry mistake or an inventory action.  That is, the Inventory
    process can viewed as a sub-case of Movement. (This is how it works at
    PAHMA in TMS now).  Therefore, all location updates are indeed moves!
    (Aside: from an implementation point of view, perhaps all movements merit
    a separate attribute on the Movement record to indicate their "moveType",
    to facilitate UI filtering and reporting.)

  2. Both Objects and Locations can move.  It is important not to conflate
    these two cases.

2a. On the one hand, moving a bunch of objects that are all in the same
location to another location could be conceptualized as just that: a bunch
of objects moving. However, if in reality it is the location that is
moving (a tray, or a crate), then perhaps this is best thought of (and
perhaps implemented as) a change to one Location and not as a change to
multiple Objects.

Indeed, it is unclear to me how CSpace implements "Movements of
Locations", if it handles them at all.

2b. Thinking about this is complicated by the fact that Movable Locations
have a number of potential important properties that Immovable Locations
do not have, and these distinctions may need to be captured.  Subcases of
Movable Locations include not only the "crate use case" and "specimen tray
use case" (not the same, I think), but alsot Transitory Locations (which
exist only for a little while, e.g. "this crate is now on Moving Van #7
and enroute", "this package was picked up by UPS, tracking number
1Z23456789") and Fuzzy Locations (which are underspecified or even
completely unspecified e.g. "all the crates in this room will be moving
tomorrow to our Storage Facility, we just don't know which room", "the
specimens on these trays on Rack 26 are gonna need to get cleared out and
go someplace refrigerated soon!").

  1. The constraints on Object Movement are of course different than those
    for Location Movement. A short list of these:

3a. Movable Locations can move to most any location (movable or not,
though real-world constraints certainly exist), but Fixed Locations cannot
move.  The case of a Movable Location moving to another Movable (or
Transitory) Location is usually either a "nesting move" (a container into
another container) or an instance of "typical move", where objects are
being moved en masse for exhibition, restoration, etc.

3b. Changing a Fixed Location (in the Storage Location Authority) may
entail a number of real-world and databases changes, ergo the idea of
having "write once" location storage records seems like quite a good one.
In PAHMA's case, a location's displayName will be barcoded so updating its
record in the database implies reprinting and distributing barcode labels.
In PAHMA's case, therefore "crates", if represented as Locations, will
have to have immutable displayNames. This in turn creates the
uncomfortable conundrum that the displayName in CSpace as shown in the UI
will not show the actual location in PAHMA's location hierarchy. Not sure
what to do about this yet!

3c. Object movements are generally simple or atomic, though in the case
where an object is at a Movable Location a UI caution about moving the
other objects is probably warranted.

  1. Parenthetically, it's been lamented that a Storage Location authority
    lacks a suitable hierarchical model. Some of the location movement issues
    mentioned above might be addressed with such a model.  A Storage
    Location Authority is a sub-type of a Concept Authority, with (it seems)
    some additional properties. At PAHMA, the 5 levels of the "location
    hierarchy" are quite parallel to the 7 ranks of Biological Classification.
    Any CSpace implementation supporting the biological or geographical
    domains would be welcome in the location domain.

(A complication is that the values used for each rank in a location
hierarchy is not guaranteed to be unique: the string "Cordata" only occurs
in the rank of phylum, under kingdom "Animalia"; but container "Vault 15"
may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12" of
site "Hearst Gym".)

UPSHOT

The upshot of this, presented tentatively and with humility, is:

  • Movement and Location Authority records deserve some set of attributes
    (term statuses?) in the core implementation to facilitate some
    as-yet-not-completely-understood-or-implemented features. Examples of
    these attributes might include (for Storage Locations) "active/inactive",
    "movable/fixed", "transitory/permanent", etc.  Patrick noted that some/all
    of this functionality exists for authorities. For movements, statuses or
    attributes might include "planned/completed", "inventory/reallymoving",
    etc. (Can a term in an authority have several statuses?)

  • Best Practices and a few "worked examples" might go a long way towards
    clarifying what can and can't be done and how. Wiki pages covering cases
    like "How to Move Containers like Crates and Trays", "Moving Multiple
    Objects", "Moving Objects When You Don't Know Where They Are Going",
    "Updating a Location Authority Record: Making Sure It Works", etc. remain
    to be defined.  The workflows proposed may in fact cover the required
    cases, but a warm and fuzzy sense of completeness is lacking.

  • At PAHMA, we are considering whether a generalization of one of the
    "regular" Movement workflows (i.e. "location update via barcode scanner")
    might not be the way to actually accomplish the migration of existing
    object location data to CSpace.  Reactions from other users about the
    efficacy and wisdom of this would be most welcome!

Hope this helps!

John B. Lowe, PhD

"Wherever you go, there you are" -- DR Davenport


Talk mailing list
Talk@lists.collectionspace.org

http://lists.collectionspace.org/mailman/listinfo/talk_lists.collectionspace.org

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800
Direct 718 777 6834

Thank you for the feedback, John, and welcome to the team! I think there are more issues in the below than can be comfortably dealt with over email - how about an open Skype or Connect session with any interested parties? I could do anytime in the afternoon on either Monday or Tuesday next week. If anyone from SMK would like to attend, we'd have to go earlier, more like 8PST/11EST, which I could do on Tuesday or Wednesday of next week. Please let me know what works for you. Best regards, Megan On Thu, Jan 26, 2012 at 7:11 PM, <jblowe@berkeley.edu> wrote: > Megan, et al., > > Thanks for the opporunity to comment on changes to this functionality. > > The alas rather lengthy response here grew from recent discussions with > the UCB deployment team working on PAHMA and other CSpace folks. While > the use cases and points made below are relevant to the PAHMA > implementation, a review of wiki pages indicates that most if not all of > these are germane to other institutions, so I apologize in advance if the > points and use cases at times redundant with remarks already made wrt > other deployments; nevertheless, I have at times mentioned other use cases > in the discussion below to butress my points. > > And parenthetically, let me say that despite the fact that this is my > first month as a CSpace deployer, I fully expect to be held to answer for > any errors, omissions, and misrepresentations in these remarks. Please > point out any faults you find! And I apologize again in advance for the > length of this post. (Je n'ai fait celle-ci plus longue que parce que je > n'ai pas eu le loisir de la faire plus courte. -- Pascal) > > WIKI PAGES REFERRED TO > > Current proposal for changes, contains links to current and proposed > workflows > > > http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade > > A Requirements page, perhaps now outdated, describing how locations and > and the history of an object's location are to be handled. > > > http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Requirements > > Use cases and other "on the ground" writings about locations and movements. > > > http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases > > > http://wiki.collectionspace.org/display/deploy/PAHMA+Storage+Location+data+mapping > > > http://wiki.collectionspace.org/display/deploy/PAHMA+Customizations+-+Storage+Authority > > > https://wiki.collectionspace.org:8444/display/deploy/Crates+and+other+mobile+containers > > REMARKS AND OBSERVATIONS > > While it does seem clear that some revisions will be needed to the way > Locations and Movements are handled, some of the changes suggested seem > rather drastic while others seem not to go far enough. Some of the issues > identified appear to be *merely* semantic, conceptual, or ontological (and > therefore may *simply* require a change in thinking and documentation), > while others are engineering conundrums which (may) require changes at > every level of the software, schema to UI). I am not sure which is which! > > The notion that "the current situation where an object or group of objects > must be 'move' in order to be given an updated location" represents a bug > to be fixed is a misconceptualization. It is in fact a desired feature > whose utility deserves clarification. Some examples: > > 1. Moving an object from one location to the same location is either a > data entry mistake *or* an inventory action. That is, the Inventory > process can viewed as a sub-case of Movement. (This is how it works at > PAHMA in TMS now). Therefore, all location updates are indeed moves! > (Aside: from an implementation point of view, perhaps all movements merit > a separate attribute on the Movement record to indicate their "moveType", > to facilitate UI filtering and reporting.) > > 2. Both Objects and Locations can move. It is important not to conflate > these two cases. > > 2a. On the one hand, moving a bunch of objects that are all in the same > location to another location could be conceptualized as just that: a bunch > of objects moving. However, if in reality it is the *location* that is > moving (a tray, or a crate), then perhaps this is best thought of (and > perhaps implemented as) a change to one Location and not as a change to > multiple Objects. > > Indeed, it is unclear to me how CSpace implements "Movements of > Locations", if it handles them at all. > > 2b. Thinking about this is complicated by the fact that Movable Locations > have a number of potential important properties that Immovable Locations > do not have, and these distinctions may need to be captured. Subcases of > Movable Locations include not only the "crate use case" and "specimen tray > use case" (not the same, I think), but alsot Transitory Locations (which > exist only for a little while, e.g. "this crate is now on Moving Van #7 > and enroute", "this package was picked up by UPS, tracking number > 1Z23456789") and Fuzzy Locations (which are underspecified or even > completely unspecified e.g. "all the crates in this room will be moving > tomorrow to our Storage Facility, we just don't know which room", "the > specimens on these trays on Rack 26 are gonna need to get cleared out and > go someplace refrigerated soon!"). > > 3. The constraints on Object Movement are of course different than those > for Location Movement. A short list of these: > > 3a. Movable Locations can move to most any location (movable or not, > though real-world constraints certainly exist), but Fixed Locations cannot > move. The case of a Movable Location moving to another Movable (or > Transitory) Location is usually either a "nesting move" (a container into > another container) or an instance of "typical move", where objects are > being moved en masse for exhibition, restoration, etc. > > 3b. Changing a Fixed Location (in the Storage Location Authority) may > entail a number of real-world and databases changes, ergo the idea of > having "write once" location storage records seems like quite a good one. > In PAHMA's case, a location's displayName will be barcoded so updating its > record in the database implies reprinting and distributing barcode labels. > In PAHMA's case, therefore "crates", if represented as Locations, will > have to have immutable displayNames. This in turn creates the > uncomfortable conundrum that the displayName in CSpace as shown in the UI > will not show the actual location in PAHMA's location hierarchy. Not sure > what to do about this yet! > > 3c. Object movements are generally simple or atomic, though in the case > where an object is at a Movable Location a UI caution about moving the > other objects is probably warranted. > > 4. Parenthetically, it's been lamented that a Storage Location authority > lacks a suitable hierarchical model. Some of the location movement issues > mentioned above *might* be addressed with such a model. A Storage > Location Authority is a sub-type of a Concept Authority, with (it seems) > some additional properties. At PAHMA, the 5 levels of the "location > hierarchy" are quite parallel to the 7 ranks of Biological Classification. > Any CSpace implementation supporting the biological or geographical > domains would be welcome in the location domain. > > (A complication is that the values used for each rank in a location > hierarchy is not guaranteed to be unique: the string "Cordata" only occurs > in the rank of phylum, under kingdom "Animalia"; but container "Vault 15" > may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12" of > site "Hearst Gym".) > > UPSHOT > > The upshot of this, presented tentatively and with humility, is: > > * Movement and Location Authority records deserve some set of attributes > (term statuses?) in the core implementation to facilitate some > as-yet-not-completely-understood-or-implemented features. Examples of > these attributes might include (for Storage Locations) "active/inactive", > "movable/fixed", "transitory/permanent", etc. Patrick noted that some/all > of this functionality exists for authorities. For movements, statuses or > attributes might include "planned/completed", "inventory/reallymoving", > etc. (Can a term in an authority have several statuses?) > > * Best Practices and a few "worked examples" might go a long way towards > clarifying what can and can't be done and how. Wiki pages covering cases > like "How to Move Containers like Crates and Trays", "Moving Multiple > Objects", "Moving Objects When You Don't Know Where They Are Going", > "Updating a Location Authority Record: Making Sure It Works", etc. remain > to be defined. The workflows proposed may in fact cover the required > cases, but a warm and fuzzy sense of completeness is lacking. > > * At PAHMA, we are considering whether a generalization of one of the > "regular" Movement workflows (i.e. "location update via barcode scanner") > might not be the way to actually accomplish the migration of existing > object location data to CSpace. Reactions from other users about the > efficacy and wisdom of this would be most welcome! > > Hope this helps! > > John B. Lowe, PhD > > "Wherever you go, there you are" -- DR Davenport > > > _______________________________________________ > Talk mailing list > Talk@lists.collectionspace.org > > http://lists.collectionspace.org/mailman/listinfo/talk_lists.collectionspace.org > -- Megan Forbes Collection Manager Museum of the Moving Image 36-01 35 Avenue Astoria, NY 11106 movingimage.us 718 777 6800 Direct 718 777 6834
J
jblowe@berkeley.edu
Fri, Jan 27, 2012 5:56 PM

Megan,

Yes, there's a lot there!

I could do any of those time, with a slight preference for Monday.

May I tentatively suggest Monday 10AM PST / 1PM EST? We can wait to here
from other interested parties before confirming.

Regards,

John

Thank you for the feedback, John, and welcome to the team!

I think there are more issues in the below than can be comfortably dealt
with over email - how about an open Skype or Connect session with any
interested parties? I could do anytime in the afternoon on either Monday
or
Tuesday next week. If anyone from SMK would like to attend, we'd have to
go
earlier, more like 8PST/11EST, which I could do on Tuesday or Wednesday of
next week.

Please let me know what works for you.

Best regards,
Megan

On Thu, Jan 26, 2012 at 7:11 PM, jblowe@berkeley.edu wrote:

Megan, et al.,

Thanks for the opporunity to comment on changes to this functionality.

The alas rather lengthy response here grew from recent discussions with
the UCB deployment team working on PAHMA and other CSpace folks.  While
the use cases and points made below are relevant to the PAHMA
implementation, a review of wiki pages indicates that most if not all of
these are germane to other institutions, so I apologize in advance if
the
points and use cases at times redundant with remarks already made wrt
other deployments; nevertheless, I have at times mentioned other use
cases
in the discussion below to butress my points.

And parenthetically, let me say that despite the fact that this is my
first month as a CSpace deployer, I fully expect to be held to answer
for
any errors, omissions, and misrepresentations in these remarks. Please
point out any faults you find! And I apologize again in advance for the
length of this post. (Je n'ai fait celle-ci plus longue que parce que je
n'ai pas eu le loisir de la faire plus courte. -- Pascal)

WIKI PAGES REFERRED TO

Current proposal for changes, contains links to current and proposed
workflows

http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade

A Requirements page, perhaps now outdated, describing how locations and
and the history of an object's location are to be handled.

http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Requirements

Use cases and other "on the ground" writings about locations and
movements.

http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases

http://wiki.collectionspace.org/display/deploy/PAHMA+Storage+Location+data+mapping

http://wiki.collectionspace.org/display/deploy/PAHMA+Customizations+-+Storage+Authority

https://wiki.collectionspace.org:8444/display/deploy/Crates+and+other+mobile+containers

REMARKS AND OBSERVATIONS

While it does seem clear that some revisions will be needed to the way
Locations and Movements are handled, some of the changes suggested seem
rather drastic while others seem not to go far enough. Some of the
issues
identified appear to be merely semantic, conceptual, or ontological
(and
therefore may simply require a change in thinking and documentation),
while others are engineering conundrums which (may) require changes at
every level of the software, schema to UI). I am not sure which is
which!

The notion that "the current situation where an object or group of
objects
must be 'move' in order to be given an updated location" represents a
bug
to be fixed is a misconceptualization. It is in fact a desired feature
whose utility deserves clarification. Some examples:

  1. Moving an object from one location to the same location is either a
    data entry mistake or an inventory action.  That is, the Inventory
    process can viewed as a sub-case of Movement. (This is how it works at
    PAHMA in TMS now).  Therefore, all location updates are indeed moves!
    (Aside: from an implementation point of view, perhaps all movements
    merit
    a separate attribute on the Movement record to indicate their
    "moveType",
    to facilitate UI filtering and reporting.)

  2. Both Objects and Locations can move.  It is important not to conflate
    these two cases.

2a. On the one hand, moving a bunch of objects that are all in the same
location to another location could be conceptualized as just that: a
bunch
of objects moving. However, if in reality it is the location that is
moving (a tray, or a crate), then perhaps this is best thought of (and
perhaps implemented as) a change to one Location and not as a change to
multiple Objects.

Indeed, it is unclear to me how CSpace implements "Movements of
Locations", if it handles them at all.

2b. Thinking about this is complicated by the fact that Movable
Locations
have a number of potential important properties that Immovable Locations
do not have, and these distinctions may need to be captured.  Subcases
of
Movable Locations include not only the "crate use case" and "specimen
tray
use case" (not the same, I think), but alsot Transitory Locations (which
exist only for a little while, e.g. "this crate is now on Moving Van #7
and enroute", "this package was picked up by UPS, tracking number
1Z23456789") and Fuzzy Locations (which are underspecified or even
completely unspecified e.g. "all the crates in this room will be moving
tomorrow to our Storage Facility, we just don't know which room", "the
specimens on these trays on Rack 26 are gonna need to get cleared out
and
go someplace refrigerated soon!").

  1. The constraints on Object Movement are of course different than those
    for Location Movement. A short list of these:

3a. Movable Locations can move to most any location (movable or not,
though real-world constraints certainly exist), but Fixed Locations
cannot
move.  The case of a Movable Location moving to another Movable (or
Transitory) Location is usually either a "nesting move" (a container
into
another container) or an instance of "typical move", where objects are
being moved en masse for exhibition, restoration, etc.

3b. Changing a Fixed Location (in the Storage Location Authority) may
entail a number of real-world and databases changes, ergo the idea of
having "write once" location storage records seems like quite a good
one.
In PAHMA's case, a location's displayName will be barcoded so updating
its
record in the database implies reprinting and distributing barcode
labels.
In PAHMA's case, therefore "crates", if represented as Locations, will
have to have immutable displayNames. This in turn creates the
uncomfortable conundrum that the displayName in CSpace as shown in the
UI
will not show the actual location in PAHMA's location hierarchy. Not
sure
what to do about this yet!

3c. Object movements are generally simple or atomic, though in the case
where an object is at a Movable Location a UI caution about moving the
other objects is probably warranted.

  1. Parenthetically, it's been lamented that a Storage Location authority
    lacks a suitable hierarchical model. Some of the location movement
    issues
    mentioned above might be addressed with such a model.  A Storage
    Location Authority is a sub-type of a Concept Authority, with (it seems)
    some additional properties. At PAHMA, the 5 levels of the "location
    hierarchy" are quite parallel to the 7 ranks of Biological
    Classification.
    Any CSpace implementation supporting the biological or geographical
    domains would be welcome in the location domain.

(A complication is that the values used for each rank in a location
hierarchy is not guaranteed to be unique: the string "Cordata" only
occurs
in the rank of phylum, under kingdom "Animalia"; but container "Vault
15"
may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12" of
site "Hearst Gym".)

UPSHOT

The upshot of this, presented tentatively and with humility, is:

  • Movement and Location Authority records deserve some set of attributes
    (term statuses?) in the core implementation to facilitate some
    as-yet-not-completely-understood-or-implemented features. Examples of
    these attributes might include (for Storage Locations)
    "active/inactive",
    "movable/fixed", "transitory/permanent", etc.  Patrick noted that
    some/all
    of this functionality exists for authorities. For movements, statuses or
    attributes might include "planned/completed", "inventory/reallymoving",
    etc. (Can a term in an authority have several statuses?)

  • Best Practices and a few "worked examples" might go a long way towards
    clarifying what can and can't be done and how. Wiki pages covering cases
    like "How to Move Containers like Crates and Trays", "Moving Multiple
    Objects", "Moving Objects When You Don't Know Where They Are Going",
    "Updating a Location Authority Record: Making Sure It Works", etc.
    remain
    to be defined.  The workflows proposed may in fact cover the required
    cases, but a warm and fuzzy sense of completeness is lacking.

  • At PAHMA, we are considering whether a generalization of one of the
    "regular" Movement workflows (i.e. "location update via barcode
    scanner")
    might not be the way to actually accomplish the migration of existing
    object location data to CSpace.  Reactions from other users about the
    efficacy and wisdom of this would be most welcome!

Hope this helps!

John B. Lowe, PhD

"Wherever you go, there you are" -- DR Davenport


Talk mailing list
Talk@lists.collectionspace.org

http://lists.collectionspace.org/mailman/listinfo/talk_lists.collectionspace.org

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800
Direct 718 777 6834

Megan, Yes, there's a lot there! I could do any of those time, with a slight preference for Monday. May I tentatively suggest Monday 10AM PST / 1PM EST? We can wait to here from other interested parties before confirming. Regards, John > Thank you for the feedback, John, and welcome to the team! > > I think there are more issues in the below than can be comfortably dealt > with over email - how about an open Skype or Connect session with any > interested parties? I could do anytime in the afternoon on either Monday > or > Tuesday next week. If anyone from SMK would like to attend, we'd have to > go > earlier, more like 8PST/11EST, which I could do on Tuesday or Wednesday of > next week. > > Please let me know what works for you. > > Best regards, > Megan > > > On Thu, Jan 26, 2012 at 7:11 PM, <jblowe@berkeley.edu> wrote: > >> Megan, et al., >> >> Thanks for the opporunity to comment on changes to this functionality. >> >> The alas rather lengthy response here grew from recent discussions with >> the UCB deployment team working on PAHMA and other CSpace folks. While >> the use cases and points made below are relevant to the PAHMA >> implementation, a review of wiki pages indicates that most if not all of >> these are germane to other institutions, so I apologize in advance if >> the >> points and use cases at times redundant with remarks already made wrt >> other deployments; nevertheless, I have at times mentioned other use >> cases >> in the discussion below to butress my points. >> >> And parenthetically, let me say that despite the fact that this is my >> first month as a CSpace deployer, I fully expect to be held to answer >> for >> any errors, omissions, and misrepresentations in these remarks. Please >> point out any faults you find! And I apologize again in advance for the >> length of this post. (Je n'ai fait celle-ci plus longue que parce que je >> n'ai pas eu le loisir de la faire plus courte. -- Pascal) >> >> WIKI PAGES REFERRED TO >> >> Current proposal for changes, contains links to current and proposed >> workflows >> >> >> http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade >> >> A Requirements page, perhaps now outdated, describing how locations and >> and the history of an object's location are to be handled. >> >> >> http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Requirements >> >> Use cases and other "on the ground" writings about locations and >> movements. >> >> >> http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases >> >> >> http://wiki.collectionspace.org/display/deploy/PAHMA+Storage+Location+data+mapping >> >> >> http://wiki.collectionspace.org/display/deploy/PAHMA+Customizations+-+Storage+Authority >> >> >> https://wiki.collectionspace.org:8444/display/deploy/Crates+and+other+mobile+containers >> >> REMARKS AND OBSERVATIONS >> >> While it does seem clear that some revisions will be needed to the way >> Locations and Movements are handled, some of the changes suggested seem >> rather drastic while others seem not to go far enough. Some of the >> issues >> identified appear to be *merely* semantic, conceptual, or ontological >> (and >> therefore may *simply* require a change in thinking and documentation), >> while others are engineering conundrums which (may) require changes at >> every level of the software, schema to UI). I am not sure which is >> which! >> >> The notion that "the current situation where an object or group of >> objects >> must be 'move' in order to be given an updated location" represents a >> bug >> to be fixed is a misconceptualization. It is in fact a desired feature >> whose utility deserves clarification. Some examples: >> >> 1. Moving an object from one location to the same location is either a >> data entry mistake *or* an inventory action. That is, the Inventory >> process can viewed as a sub-case of Movement. (This is how it works at >> PAHMA in TMS now). Therefore, all location updates are indeed moves! >> (Aside: from an implementation point of view, perhaps all movements >> merit >> a separate attribute on the Movement record to indicate their >> "moveType", >> to facilitate UI filtering and reporting.) >> >> 2. Both Objects and Locations can move. It is important not to conflate >> these two cases. >> >> 2a. On the one hand, moving a bunch of objects that are all in the same >> location to another location could be conceptualized as just that: a >> bunch >> of objects moving. However, if in reality it is the *location* that is >> moving (a tray, or a crate), then perhaps this is best thought of (and >> perhaps implemented as) a change to one Location and not as a change to >> multiple Objects. >> >> Indeed, it is unclear to me how CSpace implements "Movements of >> Locations", if it handles them at all. >> >> 2b. Thinking about this is complicated by the fact that Movable >> Locations >> have a number of potential important properties that Immovable Locations >> do not have, and these distinctions may need to be captured. Subcases >> of >> Movable Locations include not only the "crate use case" and "specimen >> tray >> use case" (not the same, I think), but alsot Transitory Locations (which >> exist only for a little while, e.g. "this crate is now on Moving Van #7 >> and enroute", "this package was picked up by UPS, tracking number >> 1Z23456789") and Fuzzy Locations (which are underspecified or even >> completely unspecified e.g. "all the crates in this room will be moving >> tomorrow to our Storage Facility, we just don't know which room", "the >> specimens on these trays on Rack 26 are gonna need to get cleared out >> and >> go someplace refrigerated soon!"). >> >> 3. The constraints on Object Movement are of course different than those >> for Location Movement. A short list of these: >> >> 3a. Movable Locations can move to most any location (movable or not, >> though real-world constraints certainly exist), but Fixed Locations >> cannot >> move. The case of a Movable Location moving to another Movable (or >> Transitory) Location is usually either a "nesting move" (a container >> into >> another container) or an instance of "typical move", where objects are >> being moved en masse for exhibition, restoration, etc. >> >> 3b. Changing a Fixed Location (in the Storage Location Authority) may >> entail a number of real-world and databases changes, ergo the idea of >> having "write once" location storage records seems like quite a good >> one. >> In PAHMA's case, a location's displayName will be barcoded so updating >> its >> record in the database implies reprinting and distributing barcode >> labels. >> In PAHMA's case, therefore "crates", if represented as Locations, will >> have to have immutable displayNames. This in turn creates the >> uncomfortable conundrum that the displayName in CSpace as shown in the >> UI >> will not show the actual location in PAHMA's location hierarchy. Not >> sure >> what to do about this yet! >> >> 3c. Object movements are generally simple or atomic, though in the case >> where an object is at a Movable Location a UI caution about moving the >> other objects is probably warranted. >> >> 4. Parenthetically, it's been lamented that a Storage Location authority >> lacks a suitable hierarchical model. Some of the location movement >> issues >> mentioned above *might* be addressed with such a model. A Storage >> Location Authority is a sub-type of a Concept Authority, with (it seems) >> some additional properties. At PAHMA, the 5 levels of the "location >> hierarchy" are quite parallel to the 7 ranks of Biological >> Classification. >> Any CSpace implementation supporting the biological or geographical >> domains would be welcome in the location domain. >> >> (A complication is that the values used for each rank in a location >> hierarchy is not guaranteed to be unique: the string "Cordata" only >> occurs >> in the rank of phylum, under kingdom "Animalia"; but container "Vault >> 15" >> may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12" of >> site "Hearst Gym".) >> >> UPSHOT >> >> The upshot of this, presented tentatively and with humility, is: >> >> * Movement and Location Authority records deserve some set of attributes >> (term statuses?) in the core implementation to facilitate some >> as-yet-not-completely-understood-or-implemented features. Examples of >> these attributes might include (for Storage Locations) >> "active/inactive", >> "movable/fixed", "transitory/permanent", etc. Patrick noted that >> some/all >> of this functionality exists for authorities. For movements, statuses or >> attributes might include "planned/completed", "inventory/reallymoving", >> etc. (Can a term in an authority have several statuses?) >> >> * Best Practices and a few "worked examples" might go a long way towards >> clarifying what can and can't be done and how. Wiki pages covering cases >> like "How to Move Containers like Crates and Trays", "Moving Multiple >> Objects", "Moving Objects When You Don't Know Where They Are Going", >> "Updating a Location Authority Record: Making Sure It Works", etc. >> remain >> to be defined. The workflows proposed may in fact cover the required >> cases, but a warm and fuzzy sense of completeness is lacking. >> >> * At PAHMA, we are considering whether a generalization of one of the >> "regular" Movement workflows (i.e. "location update via barcode >> scanner") >> might not be the way to actually accomplish the migration of existing >> object location data to CSpace. Reactions from other users about the >> efficacy and wisdom of this would be most welcome! >> >> Hope this helps! >> >> John B. Lowe, PhD >> >> "Wherever you go, there you are" -- DR Davenport >> >> >> _______________________________________________ >> Talk mailing list >> Talk@lists.collectionspace.org >> >> http://lists.collectionspace.org/mailman/listinfo/talk_lists.collectionspace.org >> > > > > -- > Megan Forbes > Collection Manager > > Museum of the Moving Image > 36-01 35 Avenue Astoria, NY 11106 > movingimage.us 718 777 6800 > Direct 718 777 6834 >
MT
Michael T. Black
Fri, Jan 27, 2012 8:21 PM

Hi John,

I'd be interested in joining the conversation, but I won't be able to meet on Monday from 9:20 to 11:20.  Any other time on Monday or Tuesday (or even today) would be fine.

Michael

On Jan 27, 2012, at 9:56 AM, jblowe@berkeley.edu wrote:

Megan,

Yes, there's a lot there!

I could do any of those time, with a slight preference for Monday.

May I tentatively suggest Monday 10AM PST / 1PM EST? We can wait to here
from other interested parties before confirming.

Regards,

John

Thank you for the feedback, John, and welcome to the team!

I think there are more issues in the below than can be comfortably dealt
with over email - how about an open Skype or Connect session with any
interested parties? I could do anytime in the afternoon on either Monday
or
Tuesday next week. If anyone from SMK would like to attend, we'd have to
go
earlier, more like 8PST/11EST, which I could do on Tuesday or Wednesday of
next week.

Please let me know what works for you.

Best regards,
Megan

On Thu, Jan 26, 2012 at 7:11 PM, jblowe@berkeley.edu wrote:

Megan, et al.,

Thanks for the opporunity to comment on changes to this functionality.

The alas rather lengthy response here grew from recent discussions with
the UCB deployment team working on PAHMA and other CSpace folks.  While
the use cases and points made below are relevant to the PAHMA
implementation, a review of wiki pages indicates that most if not all of
these are germane to other institutions, so I apologize in advance if
the
points and use cases at times redundant with remarks already made wrt
other deployments; nevertheless, I have at times mentioned other use
cases
in the discussion below to butress my points.

And parenthetically, let me say that despite the fact that this is my
first month as a CSpace deployer, I fully expect to be held to answer
for
any errors, omissions, and misrepresentations in these remarks. Please
point out any faults you find! And I apologize again in advance for the
length of this post. (Je n'ai fait celle-ci plus longue que parce que je
n'ai pas eu le loisir de la faire plus courte. -- Pascal)

WIKI PAGES REFERRED TO

Current proposal for changes, contains links to current and proposed
workflows

http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade

A Requirements page, perhaps now outdated, describing how locations and
and the history of an object's location are to be handled.

http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Requirements

Use cases and other "on the ground" writings about locations and
movements.

http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases

http://wiki.collectionspace.org/display/deploy/PAHMA+Storage+Location+data+mapping

http://wiki.collectionspace.org/display/deploy/PAHMA+Customizations+-+Storage+Authority

https://wiki.collectionspace.org:8444/display/deploy/Crates+and+other+mobile+containers

REMARKS AND OBSERVATIONS

While it does seem clear that some revisions will be needed to the way
Locations and Movements are handled, some of the changes suggested seem
rather drastic while others seem not to go far enough. Some of the
issues
identified appear to be merely semantic, conceptual, or ontological
(and
therefore may simply require a change in thinking and documentation),
while others are engineering conundrums which (may) require changes at
every level of the software, schema to UI). I am not sure which is
which!

The notion that "the current situation where an object or group of
objects
must be 'move' in order to be given an updated location" represents a
bug
to be fixed is a misconceptualization. It is in fact a desired feature
whose utility deserves clarification. Some examples:

  1. Moving an object from one location to the same location is either a
    data entry mistake or an inventory action.  That is, the Inventory
    process can viewed as a sub-case of Movement. (This is how it works at
    PAHMA in TMS now).  Therefore, all location updates are indeed moves!
    (Aside: from an implementation point of view, perhaps all movements
    merit
    a separate attribute on the Movement record to indicate their
    "moveType",
    to facilitate UI filtering and reporting.)

  2. Both Objects and Locations can move.  It is important not to conflate
    these two cases.

2a. On the one hand, moving a bunch of objects that are all in the same
location to another location could be conceptualized as just that: a
bunch
of objects moving. However, if in reality it is the location that is
moving (a tray, or a crate), then perhaps this is best thought of (and
perhaps implemented as) a change to one Location and not as a change to
multiple Objects.

Indeed, it is unclear to me how CSpace implements "Movements of
Locations", if it handles them at all.

2b. Thinking about this is complicated by the fact that Movable
Locations
have a number of potential important properties that Immovable Locations
do not have, and these distinctions may need to be captured.  Subcases
of
Movable Locations include not only the "crate use case" and "specimen
tray
use case" (not the same, I think), but alsot Transitory Locations (which
exist only for a little while, e.g. "this crate is now on Moving Van #7
and enroute", "this package was picked up by UPS, tracking number
1Z23456789") and Fuzzy Locations (which are underspecified or even
completely unspecified e.g. "all the crates in this room will be moving
tomorrow to our Storage Facility, we just don't know which room", "the
specimens on these trays on Rack 26 are gonna need to get cleared out
and
go someplace refrigerated soon!").

  1. The constraints on Object Movement are of course different than those
    for Location Movement. A short list of these:

3a. Movable Locations can move to most any location (movable or not,
though real-world constraints certainly exist), but Fixed Locations
cannot
move.  The case of a Movable Location moving to another Movable (or
Transitory) Location is usually either a "nesting move" (a container
into
another container) or an instance of "typical move", where objects are
being moved en masse for exhibition, restoration, etc.

3b. Changing a Fixed Location (in the Storage Location Authority) may
entail a number of real-world and databases changes, ergo the idea of
having "write once" location storage records seems like quite a good
one.
In PAHMA's case, a location's displayName will be barcoded so updating
its
record in the database implies reprinting and distributing barcode
labels.
In PAHMA's case, therefore "crates", if represented as Locations, will
have to have immutable displayNames. This in turn creates the
uncomfortable conundrum that the displayName in CSpace as shown in the
UI
will not show the actual location in PAHMA's location hierarchy. Not
sure
what to do about this yet!

3c. Object movements are generally simple or atomic, though in the case
where an object is at a Movable Location a UI caution about moving the
other objects is probably warranted.

  1. Parenthetically, it's been lamented that a Storage Location authority
    lacks a suitable hierarchical model. Some of the location movement
    issues
    mentioned above might be addressed with such a model.  A Storage
    Location Authority is a sub-type of a Concept Authority, with (it seems)
    some additional properties. At PAHMA, the 5 levels of the "location
    hierarchy" are quite parallel to the 7 ranks of Biological
    Classification.
    Any CSpace implementation supporting the biological or geographical
    domains would be welcome in the location domain.

(A complication is that the values used for each rank in a location
hierarchy is not guaranteed to be unique: the string "Cordata" only
occurs
in the rank of phylum, under kingdom "Animalia"; but container "Vault
15"
may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12" of
site "Hearst Gym".)

UPSHOT

The upshot of this, presented tentatively and with humility, is:

  • Movement and Location Authority records deserve some set of attributes
    (term statuses?) in the core implementation to facilitate some
    as-yet-not-completely-understood-or-implemented features. Examples of
    these attributes might include (for Storage Locations)
    "active/inactive",
    "movable/fixed", "transitory/permanent", etc.  Patrick noted that
    some/all
    of this functionality exists for authorities. For movements, statuses or
    attributes might include "planned/completed", "inventory/reallymoving",
    etc. (Can a term in an authority have several statuses?)

  • Best Practices and a few "worked examples" might go a long way towards
    clarifying what can and can't be done and how. Wiki pages covering cases
    like "How to Move Containers like Crates and Trays", "Moving Multiple
    Objects", "Moving Objects When You Don't Know Where They Are Going",
    "Updating a Location Authority Record: Making Sure It Works", etc.
    remain
    to be defined.  The workflows proposed may in fact cover the required
    cases, but a warm and fuzzy sense of completeness is lacking.

  • At PAHMA, we are considering whether a generalization of one of the
    "regular" Movement workflows (i.e. "location update via barcode
    scanner")
    might not be the way to actually accomplish the migration of existing
    object location data to CSpace.  Reactions from other users about the
    efficacy and wisdom of this would be most welcome!

Hope this helps!

John B. Lowe, PhD

"Wherever you go, there you are" -- DR Davenport


Talk mailing list
Talk@lists.collectionspace.org

http://lists.collectionspace.org/mailman/listinfo/talk_lists.collectionspace.org

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800
Direct 718 777 6834

Hi John, I'd be interested in joining the conversation, but I won't be able to meet on Monday from 9:20 to 11:20. Any other time on Monday or Tuesday (or even today) would be fine. Michael On Jan 27, 2012, at 9:56 AM, jblowe@berkeley.edu wrote: > Megan, > > Yes, there's a lot there! > > I could do any of those time, with a slight preference for Monday. > > May I tentatively suggest Monday 10AM PST / 1PM EST? We can wait to here > from other interested parties before confirming. > > Regards, > > John > >> Thank you for the feedback, John, and welcome to the team! >> >> I think there are more issues in the below than can be comfortably dealt >> with over email - how about an open Skype or Connect session with any >> interested parties? I could do anytime in the afternoon on either Monday >> or >> Tuesday next week. If anyone from SMK would like to attend, we'd have to >> go >> earlier, more like 8PST/11EST, which I could do on Tuesday or Wednesday of >> next week. >> >> Please let me know what works for you. >> >> Best regards, >> Megan >> >> >> On Thu, Jan 26, 2012 at 7:11 PM, <jblowe@berkeley.edu> wrote: >> >>> Megan, et al., >>> >>> Thanks for the opporunity to comment on changes to this functionality. >>> >>> The alas rather lengthy response here grew from recent discussions with >>> the UCB deployment team working on PAHMA and other CSpace folks. While >>> the use cases and points made below are relevant to the PAHMA >>> implementation, a review of wiki pages indicates that most if not all of >>> these are germane to other institutions, so I apologize in advance if >>> the >>> points and use cases at times redundant with remarks already made wrt >>> other deployments; nevertheless, I have at times mentioned other use >>> cases >>> in the discussion below to butress my points. >>> >>> And parenthetically, let me say that despite the fact that this is my >>> first month as a CSpace deployer, I fully expect to be held to answer >>> for >>> any errors, omissions, and misrepresentations in these remarks. Please >>> point out any faults you find! And I apologize again in advance for the >>> length of this post. (Je n'ai fait celle-ci plus longue que parce que je >>> n'ai pas eu le loisir de la faire plus courte. -- Pascal) >>> >>> WIKI PAGES REFERRED TO >>> >>> Current proposal for changes, contains links to current and proposed >>> workflows >>> >>> >>> http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade >>> >>> A Requirements page, perhaps now outdated, describing how locations and >>> and the history of an object's location are to be handled. >>> >>> >>> http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Requirements >>> >>> Use cases and other "on the ground" writings about locations and >>> movements. >>> >>> >>> http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases >>> >>> >>> http://wiki.collectionspace.org/display/deploy/PAHMA+Storage+Location+data+mapping >>> >>> >>> http://wiki.collectionspace.org/display/deploy/PAHMA+Customizations+-+Storage+Authority >>> >>> >>> https://wiki.collectionspace.org:8444/display/deploy/Crates+and+other+mobile+containers >>> >>> REMARKS AND OBSERVATIONS >>> >>> While it does seem clear that some revisions will be needed to the way >>> Locations and Movements are handled, some of the changes suggested seem >>> rather drastic while others seem not to go far enough. Some of the >>> issues >>> identified appear to be *merely* semantic, conceptual, or ontological >>> (and >>> therefore may *simply* require a change in thinking and documentation), >>> while others are engineering conundrums which (may) require changes at >>> every level of the software, schema to UI). I am not sure which is >>> which! >>> >>> The notion that "the current situation where an object or group of >>> objects >>> must be 'move' in order to be given an updated location" represents a >>> bug >>> to be fixed is a misconceptualization. It is in fact a desired feature >>> whose utility deserves clarification. Some examples: >>> >>> 1. Moving an object from one location to the same location is either a >>> data entry mistake *or* an inventory action. That is, the Inventory >>> process can viewed as a sub-case of Movement. (This is how it works at >>> PAHMA in TMS now). Therefore, all location updates are indeed moves! >>> (Aside: from an implementation point of view, perhaps all movements >>> merit >>> a separate attribute on the Movement record to indicate their >>> "moveType", >>> to facilitate UI filtering and reporting.) >>> >>> 2. Both Objects and Locations can move. It is important not to conflate >>> these two cases. >>> >>> 2a. On the one hand, moving a bunch of objects that are all in the same >>> location to another location could be conceptualized as just that: a >>> bunch >>> of objects moving. However, if in reality it is the *location* that is >>> moving (a tray, or a crate), then perhaps this is best thought of (and >>> perhaps implemented as) a change to one Location and not as a change to >>> multiple Objects. >>> >>> Indeed, it is unclear to me how CSpace implements "Movements of >>> Locations", if it handles them at all. >>> >>> 2b. Thinking about this is complicated by the fact that Movable >>> Locations >>> have a number of potential important properties that Immovable Locations >>> do not have, and these distinctions may need to be captured. Subcases >>> of >>> Movable Locations include not only the "crate use case" and "specimen >>> tray >>> use case" (not the same, I think), but alsot Transitory Locations (which >>> exist only for a little while, e.g. "this crate is now on Moving Van #7 >>> and enroute", "this package was picked up by UPS, tracking number >>> 1Z23456789") and Fuzzy Locations (which are underspecified or even >>> completely unspecified e.g. "all the crates in this room will be moving >>> tomorrow to our Storage Facility, we just don't know which room", "the >>> specimens on these trays on Rack 26 are gonna need to get cleared out >>> and >>> go someplace refrigerated soon!"). >>> >>> 3. The constraints on Object Movement are of course different than those >>> for Location Movement. A short list of these: >>> >>> 3a. Movable Locations can move to most any location (movable or not, >>> though real-world constraints certainly exist), but Fixed Locations >>> cannot >>> move. The case of a Movable Location moving to another Movable (or >>> Transitory) Location is usually either a "nesting move" (a container >>> into >>> another container) or an instance of "typical move", where objects are >>> being moved en masse for exhibition, restoration, etc. >>> >>> 3b. Changing a Fixed Location (in the Storage Location Authority) may >>> entail a number of real-world and databases changes, ergo the idea of >>> having "write once" location storage records seems like quite a good >>> one. >>> In PAHMA's case, a location's displayName will be barcoded so updating >>> its >>> record in the database implies reprinting and distributing barcode >>> labels. >>> In PAHMA's case, therefore "crates", if represented as Locations, will >>> have to have immutable displayNames. This in turn creates the >>> uncomfortable conundrum that the displayName in CSpace as shown in the >>> UI >>> will not show the actual location in PAHMA's location hierarchy. Not >>> sure >>> what to do about this yet! >>> >>> 3c. Object movements are generally simple or atomic, though in the case >>> where an object is at a Movable Location a UI caution about moving the >>> other objects is probably warranted. >>> >>> 4. Parenthetically, it's been lamented that a Storage Location authority >>> lacks a suitable hierarchical model. Some of the location movement >>> issues >>> mentioned above *might* be addressed with such a model. A Storage >>> Location Authority is a sub-type of a Concept Authority, with (it seems) >>> some additional properties. At PAHMA, the 5 levels of the "location >>> hierarchy" are quite parallel to the 7 ranks of Biological >>> Classification. >>> Any CSpace implementation supporting the biological or geographical >>> domains would be welcome in the location domain. >>> >>> (A complication is that the values used for each rank in a location >>> hierarchy is not guaranteed to be unique: the string "Cordata" only >>> occurs >>> in the rank of phylum, under kingdom "Animalia"; but container "Vault >>> 15" >>> may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12" of >>> site "Hearst Gym".) >>> >>> UPSHOT >>> >>> The upshot of this, presented tentatively and with humility, is: >>> >>> * Movement and Location Authority records deserve some set of attributes >>> (term statuses?) in the core implementation to facilitate some >>> as-yet-not-completely-understood-or-implemented features. Examples of >>> these attributes might include (for Storage Locations) >>> "active/inactive", >>> "movable/fixed", "transitory/permanent", etc. Patrick noted that >>> some/all >>> of this functionality exists for authorities. For movements, statuses or >>> attributes might include "planned/completed", "inventory/reallymoving", >>> etc. (Can a term in an authority have several statuses?) >>> >>> * Best Practices and a few "worked examples" might go a long way towards >>> clarifying what can and can't be done and how. Wiki pages covering cases >>> like "How to Move Containers like Crates and Trays", "Moving Multiple >>> Objects", "Moving Objects When You Don't Know Where They Are Going", >>> "Updating a Location Authority Record: Making Sure It Works", etc. >>> remain >>> to be defined. The workflows proposed may in fact cover the required >>> cases, but a warm and fuzzy sense of completeness is lacking. >>> >>> * At PAHMA, we are considering whether a generalization of one of the >>> "regular" Movement workflows (i.e. "location update via barcode >>> scanner") >>> might not be the way to actually accomplish the migration of existing >>> object location data to CSpace. Reactions from other users about the >>> efficacy and wisdom of this would be most welcome! >>> >>> Hope this helps! >>> >>> John B. Lowe, PhD >>> >>> "Wherever you go, there you are" -- DR Davenport >>> >>> >>> _______________________________________________ >>> Talk mailing list >>> Talk@lists.collectionspace.org >>> >>> http://lists.collectionspace.org/mailman/listinfo/talk_lists.collectionspace.org >>> >> >> >> >> -- >> Megan Forbes >> Collection Manager >> >> Museum of the Moving Image >> 36-01 35 Avenue Astoria, NY 11106 >> movingimage.us 718 777 6800 >> Direct 718 777 6834 >> > > > > _______________________________________________ > Talk mailing list > Talk@lists.collectionspace.org > http://lists.collectionspace.org/mailman/listinfo/talk_lists.collectionspace.org
MF
Megan Forbes
Fri, Jan 27, 2012 8:59 PM

How about 12PST/3EST on Monday?

On Fri, Jan 27, 2012 at 3:21 PM, Michael T. Black mtblack@berkeley.eduwrote:

Hi John,

    I'd be interested in joining the conversation, but I won't be able

to meet on Monday from 9:20 to 11:20.  Any other time on Monday or Tuesday
(or even today) would be fine.

Michael

On Jan 27, 2012, at 9:56 AM, jblowe@berkeley.edu wrote:

Megan,

Yes, there's a lot there!

I could do any of those time, with a slight preference for Monday.

May I tentatively suggest Monday 10AM PST / 1PM EST? We can wait to here
from other interested parties before confirming.

Regards,

John

Thank you for the feedback, John, and welcome to the team!

I think there are more issues in the below than can be comfortably dealt
with over email - how about an open Skype or Connect session with any
interested parties? I could do anytime in the afternoon on either Monday
or
Tuesday next week. If anyone from SMK would like to attend, we'd have to
go
earlier, more like 8PST/11EST, which I could do on Tuesday or Wednesday

of

next week.

Please let me know what works for you.

Best regards,
Megan

On Thu, Jan 26, 2012 at 7:11 PM, jblowe@berkeley.edu wrote:

Megan, et al.,

Thanks for the opporunity to comment on changes to this functionality.

The alas rather lengthy response here grew from recent discussions with
the UCB deployment team working on PAHMA and other CSpace folks.  While
the use cases and points made below are relevant to the PAHMA
implementation, a review of wiki pages indicates that most if not all

of

these are germane to other institutions, so I apologize in advance if
the
points and use cases at times redundant with remarks already made wrt
other deployments; nevertheless, I have at times mentioned other use
cases
in the discussion below to butress my points.

And parenthetically, let me say that despite the fact that this is my
first month as a CSpace deployer, I fully expect to be held to answer
for
any errors, omissions, and misrepresentations in these remarks. Please
point out any faults you find! And I apologize again in advance for the
length of this post. (Je n'ai fait celle-ci plus longue que parce que

je

n'ai pas eu le loisir de la faire plus courte. -- Pascal)

WIKI PAGES REFERRED TO

Current proposal for changes, contains links to current and proposed
workflows

A Requirements page, perhaps now outdated, describing how locations and
and the history of an object's location are to be handled.

Use cases and other "on the ground" writings about locations and
movements.

REMARKS AND OBSERVATIONS

While it does seem clear that some revisions will be needed to the way
Locations and Movements are handled, some of the changes suggested seem
rather drastic while others seem not to go far enough. Some of the
issues
identified appear to be merely semantic, conceptual, or ontological
(and
therefore may simply require a change in thinking and documentation),
while others are engineering conundrums which (may) require changes at
every level of the software, schema to UI). I am not sure which is
which!

The notion that "the current situation where an object or group of
objects
must be 'move' in order to be given an updated location" represents a
bug
to be fixed is a misconceptualization. It is in fact a desired feature
whose utility deserves clarification. Some examples:

  1. Moving an object from one location to the same location is either a
    data entry mistake or an inventory action.  That is, the Inventory
    process can viewed as a sub-case of Movement. (This is how it works at
    PAHMA in TMS now).  Therefore, all location updates are indeed moves!
    (Aside: from an implementation point of view, perhaps all movements
    merit
    a separate attribute on the Movement record to indicate their
    "moveType",
    to facilitate UI filtering and reporting.)

  2. Both Objects and Locations can move.  It is important not to

conflate

these two cases.

2a. On the one hand, moving a bunch of objects that are all in the same
location to another location could be conceptualized as just that: a
bunch
of objects moving. However, if in reality it is the location that is
moving (a tray, or a crate), then perhaps this is best thought of (and
perhaps implemented as) a change to one Location and not as a change to
multiple Objects.

Indeed, it is unclear to me how CSpace implements "Movements of
Locations", if it handles them at all.

2b. Thinking about this is complicated by the fact that Movable
Locations
have a number of potential important properties that Immovable

Locations

do not have, and these distinctions may need to be captured.  Subcases
of
Movable Locations include not only the "crate use case" and "specimen
tray
use case" (not the same, I think), but alsot Transitory Locations

(which

exist only for a little while, e.g. "this crate is now on Moving Van #7
and enroute", "this package was picked up by UPS, tracking number
1Z23456789") and Fuzzy Locations (which are underspecified or even
completely unspecified e.g. "all the crates in this room will be moving
tomorrow to our Storage Facility, we just don't know which room", "the
specimens on these trays on Rack 26 are gonna need to get cleared out
and
go someplace refrigerated soon!").

  1. The constraints on Object Movement are of course different than

those

for Location Movement. A short list of these:

3a. Movable Locations can move to most any location (movable or not,
though real-world constraints certainly exist), but Fixed Locations
cannot
move.  The case of a Movable Location moving to another Movable (or
Transitory) Location is usually either a "nesting move" (a container
into
another container) or an instance of "typical move", where objects are
being moved en masse for exhibition, restoration, etc.

3b. Changing a Fixed Location (in the Storage Location Authority) may
entail a number of real-world and databases changes, ergo the idea of
having "write once" location storage records seems like quite a good
one.
In PAHMA's case, a location's displayName will be barcoded so updating
its
record in the database implies reprinting and distributing barcode
labels.
In PAHMA's case, therefore "crates", if represented as Locations, will
have to have immutable displayNames. This in turn creates the
uncomfortable conundrum that the displayName in CSpace as shown in the
UI
will not show the actual location in PAHMA's location hierarchy. Not
sure
what to do about this yet!

3c. Object movements are generally simple or atomic, though in the case
where an object is at a Movable Location a UI caution about moving the
other objects is probably warranted.

  1. Parenthetically, it's been lamented that a Storage Location

authority

lacks a suitable hierarchical model. Some of the location movement
issues
mentioned above might be addressed with such a model.  A Storage
Location Authority is a sub-type of a Concept Authority, with (it

seems)

some additional properties. At PAHMA, the 5 levels of the "location
hierarchy" are quite parallel to the 7 ranks of Biological
Classification.
Any CSpace implementation supporting the biological or geographical
domains would be welcome in the location domain.

(A complication is that the values used for each rank in a location
hierarchy is not guaranteed to be unique: the string "Cordata" only
occurs
in the rank of phylum, under kingdom "Animalia"; but container "Vault
15"
may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12"

of

site "Hearst Gym".)

UPSHOT

The upshot of this, presented tentatively and with humility, is:

  • Movement and Location Authority records deserve some set of

attributes

(term statuses?) in the core implementation to facilitate some
as-yet-not-completely-understood-or-implemented features. Examples of
these attributes might include (for Storage Locations)
"active/inactive",
"movable/fixed", "transitory/permanent", etc.  Patrick noted that
some/all
of this functionality exists for authorities. For movements, statuses

or

attributes might include "planned/completed", "inventory/reallymoving",
etc. (Can a term in an authority have several statuses?)

  • Best Practices and a few "worked examples" might go a long way

towards

clarifying what can and can't be done and how. Wiki pages covering

cases

like "How to Move Containers like Crates and Trays", "Moving Multiple
Objects", "Moving Objects When You Don't Know Where They Are Going",
"Updating a Location Authority Record: Making Sure It Works", etc.
remain
to be defined.  The workflows proposed may in fact cover the required
cases, but a warm and fuzzy sense of completeness is lacking.

  • At PAHMA, we are considering whether a generalization of one of the
    "regular" Movement workflows (i.e. "location update via barcode
    scanner")
    might not be the way to actually accomplish the migration of existing
    object location data to CSpace.  Reactions from other users about the
    efficacy and wisdom of this would be most welcome!

Hope this helps!

John B. Lowe, PhD

"Wherever you go, there you are" -- DR Davenport


Talk mailing list
Talk@lists.collectionspace.org

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800
Direct 718 777 6834


Talk mailing list
Talk@lists.collectionspace.org

--
Megan Forbes
Collection Manager

Museum of the Moving Image
36-01 35 Avenue  Astoria, NY 11106
movingimage.us  718 777 6800
Direct 718 777 6834

How about 12PST/3EST on Monday? On Fri, Jan 27, 2012 at 3:21 PM, Michael T. Black <mtblack@berkeley.edu>wrote: > Hi John, > > I'd be interested in joining the conversation, but I won't be able > to meet on Monday from 9:20 to 11:20. Any other time on Monday or Tuesday > (or even today) would be fine. > > Michael > > On Jan 27, 2012, at 9:56 AM, jblowe@berkeley.edu wrote: > > > Megan, > > > > Yes, there's a lot there! > > > > I could do any of those time, with a slight preference for Monday. > > > > May I tentatively suggest Monday 10AM PST / 1PM EST? We can wait to here > > from other interested parties before confirming. > > > > Regards, > > > > John > > > >> Thank you for the feedback, John, and welcome to the team! > >> > >> I think there are more issues in the below than can be comfortably dealt > >> with over email - how about an open Skype or Connect session with any > >> interested parties? I could do anytime in the afternoon on either Monday > >> or > >> Tuesday next week. If anyone from SMK would like to attend, we'd have to > >> go > >> earlier, more like 8PST/11EST, which I could do on Tuesday or Wednesday > of > >> next week. > >> > >> Please let me know what works for you. > >> > >> Best regards, > >> Megan > >> > >> > >> On Thu, Jan 26, 2012 at 7:11 PM, <jblowe@berkeley.edu> wrote: > >> > >>> Megan, et al., > >>> > >>> Thanks for the opporunity to comment on changes to this functionality. > >>> > >>> The alas rather lengthy response here grew from recent discussions with > >>> the UCB deployment team working on PAHMA and other CSpace folks. While > >>> the use cases and points made below are relevant to the PAHMA > >>> implementation, a review of wiki pages indicates that most if not all > of > >>> these are germane to other institutions, so I apologize in advance if > >>> the > >>> points and use cases at times redundant with remarks already made wrt > >>> other deployments; nevertheless, I have at times mentioned other use > >>> cases > >>> in the discussion below to butress my points. > >>> > >>> And parenthetically, let me say that despite the fact that this is my > >>> first month as a CSpace deployer, I fully expect to be held to answer > >>> for > >>> any errors, omissions, and misrepresentations in these remarks. Please > >>> point out any faults you find! And I apologize again in advance for the > >>> length of this post. (Je n'ai fait celle-ci plus longue que parce que > je > >>> n'ai pas eu le loisir de la faire plus courte. -- Pascal) > >>> > >>> WIKI PAGES REFERRED TO > >>> > >>> Current proposal for changes, contains links to current and proposed > >>> workflows > >>> > >>> > >>> > http://wiki.collectionspace.org/display/collectionspace/Phase+III+Location+and+Movement+Upgrade > >>> > >>> A Requirements page, perhaps now outdated, describing how locations and > >>> and the history of an object's location are to be handled. > >>> > >>> > >>> > http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Requirements > >>> > >>> Use cases and other "on the ground" writings about locations and > >>> movements. > >>> > >>> > >>> > http://wiki.collectionspace.org/display/collectionspace/Location+and+Movement+Control+Use+Cases > >>> > >>> > >>> > http://wiki.collectionspace.org/display/deploy/PAHMA+Storage+Location+data+mapping > >>> > >>> > >>> > http://wiki.collectionspace.org/display/deploy/PAHMA+Customizations+-+Storage+Authority > >>> > >>> > >>> > https://wiki.collectionspace.org:8444/display/deploy/Crates+and+other+mobile+containers > >>> > >>> REMARKS AND OBSERVATIONS > >>> > >>> While it does seem clear that some revisions will be needed to the way > >>> Locations and Movements are handled, some of the changes suggested seem > >>> rather drastic while others seem not to go far enough. Some of the > >>> issues > >>> identified appear to be *merely* semantic, conceptual, or ontological > >>> (and > >>> therefore may *simply* require a change in thinking and documentation), > >>> while others are engineering conundrums which (may) require changes at > >>> every level of the software, schema to UI). I am not sure which is > >>> which! > >>> > >>> The notion that "the current situation where an object or group of > >>> objects > >>> must be 'move' in order to be given an updated location" represents a > >>> bug > >>> to be fixed is a misconceptualization. It is in fact a desired feature > >>> whose utility deserves clarification. Some examples: > >>> > >>> 1. Moving an object from one location to the same location is either a > >>> data entry mistake *or* an inventory action. That is, the Inventory > >>> process can viewed as a sub-case of Movement. (This is how it works at > >>> PAHMA in TMS now). Therefore, all location updates are indeed moves! > >>> (Aside: from an implementation point of view, perhaps all movements > >>> merit > >>> a separate attribute on the Movement record to indicate their > >>> "moveType", > >>> to facilitate UI filtering and reporting.) > >>> > >>> 2. Both Objects and Locations can move. It is important not to > conflate > >>> these two cases. > >>> > >>> 2a. On the one hand, moving a bunch of objects that are all in the same > >>> location to another location could be conceptualized as just that: a > >>> bunch > >>> of objects moving. However, if in reality it is the *location* that is > >>> moving (a tray, or a crate), then perhaps this is best thought of (and > >>> perhaps implemented as) a change to one Location and not as a change to > >>> multiple Objects. > >>> > >>> Indeed, it is unclear to me how CSpace implements "Movements of > >>> Locations", if it handles them at all. > >>> > >>> 2b. Thinking about this is complicated by the fact that Movable > >>> Locations > >>> have a number of potential important properties that Immovable > Locations > >>> do not have, and these distinctions may need to be captured. Subcases > >>> of > >>> Movable Locations include not only the "crate use case" and "specimen > >>> tray > >>> use case" (not the same, I think), but alsot Transitory Locations > (which > >>> exist only for a little while, e.g. "this crate is now on Moving Van #7 > >>> and enroute", "this package was picked up by UPS, tracking number > >>> 1Z23456789") and Fuzzy Locations (which are underspecified or even > >>> completely unspecified e.g. "all the crates in this room will be moving > >>> tomorrow to our Storage Facility, we just don't know which room", "the > >>> specimens on these trays on Rack 26 are gonna need to get cleared out > >>> and > >>> go someplace refrigerated soon!"). > >>> > >>> 3. The constraints on Object Movement are of course different than > those > >>> for Location Movement. A short list of these: > >>> > >>> 3a. Movable Locations can move to most any location (movable or not, > >>> though real-world constraints certainly exist), but Fixed Locations > >>> cannot > >>> move. The case of a Movable Location moving to another Movable (or > >>> Transitory) Location is usually either a "nesting move" (a container > >>> into > >>> another container) or an instance of "typical move", where objects are > >>> being moved en masse for exhibition, restoration, etc. > >>> > >>> 3b. Changing a Fixed Location (in the Storage Location Authority) may > >>> entail a number of real-world and databases changes, ergo the idea of > >>> having "write once" location storage records seems like quite a good > >>> one. > >>> In PAHMA's case, a location's displayName will be barcoded so updating > >>> its > >>> record in the database implies reprinting and distributing barcode > >>> labels. > >>> In PAHMA's case, therefore "crates", if represented as Locations, will > >>> have to have immutable displayNames. This in turn creates the > >>> uncomfortable conundrum that the displayName in CSpace as shown in the > >>> UI > >>> will not show the actual location in PAHMA's location hierarchy. Not > >>> sure > >>> what to do about this yet! > >>> > >>> 3c. Object movements are generally simple or atomic, though in the case > >>> where an object is at a Movable Location a UI caution about moving the > >>> other objects is probably warranted. > >>> > >>> 4. Parenthetically, it's been lamented that a Storage Location > authority > >>> lacks a suitable hierarchical model. Some of the location movement > >>> issues > >>> mentioned above *might* be addressed with such a model. A Storage > >>> Location Authority is a sub-type of a Concept Authority, with (it > seems) > >>> some additional properties. At PAHMA, the 5 levels of the "location > >>> hierarchy" are quite parallel to the 7 ranks of Biological > >>> Classification. > >>> Any CSpace implementation supporting the biological or geographical > >>> domains would be welcome in the location domain. > >>> > >>> (A complication is that the values used for each rank in a location > >>> hierarchy is not guaranteed to be unique: the string "Cordata" only > >>> occurs > >>> in the rank of phylum, under kingdom "Animalia"; but container "Vault > >>> 15" > >>> may appear in room "Room 12" of site "Kroeber Hall" or room "Room 12" > of > >>> site "Hearst Gym".) > >>> > >>> UPSHOT > >>> > >>> The upshot of this, presented tentatively and with humility, is: > >>> > >>> * Movement and Location Authority records deserve some set of > attributes > >>> (term statuses?) in the core implementation to facilitate some > >>> as-yet-not-completely-understood-or-implemented features. Examples of > >>> these attributes might include (for Storage Locations) > >>> "active/inactive", > >>> "movable/fixed", "transitory/permanent", etc. Patrick noted that > >>> some/all > >>> of this functionality exists for authorities. For movements, statuses > or > >>> attributes might include "planned/completed", "inventory/reallymoving", > >>> etc. (Can a term in an authority have several statuses?) > >>> > >>> * Best Practices and a few "worked examples" might go a long way > towards > >>> clarifying what can and can't be done and how. Wiki pages covering > cases > >>> like "How to Move Containers like Crates and Trays", "Moving Multiple > >>> Objects", "Moving Objects When You Don't Know Where They Are Going", > >>> "Updating a Location Authority Record: Making Sure It Works", etc. > >>> remain > >>> to be defined. The workflows proposed may in fact cover the required > >>> cases, but a warm and fuzzy sense of completeness is lacking. > >>> > >>> * At PAHMA, we are considering whether a generalization of one of the > >>> "regular" Movement workflows (i.e. "location update via barcode > >>> scanner") > >>> might not be the way to actually accomplish the migration of existing > >>> object location data to CSpace. Reactions from other users about the > >>> efficacy and wisdom of this would be most welcome! > >>> > >>> Hope this helps! > >>> > >>> John B. Lowe, PhD > >>> > >>> "Wherever you go, there you are" -- DR Davenport > >>> > >>> > >>> _______________________________________________ > >>> Talk mailing list > >>> Talk@lists.collectionspace.org > >>> > >>> > http://lists.collectionspace.org/mailman/listinfo/talk_lists.collectionspace.org > >>> > >> > >> > >> > >> -- > >> Megan Forbes > >> Collection Manager > >> > >> Museum of the Moving Image > >> 36-01 35 Avenue Astoria, NY 11106 > >> movingimage.us 718 777 6800 > >> Direct 718 777 6834 > >> > > > > > > > > _______________________________________________ > > Talk mailing list > > Talk@lists.collectionspace.org > > > http://lists.collectionspace.org/mailman/listinfo/talk_lists.collectionspace.org > > -- Megan Forbes Collection Manager Museum of the Moving Image 36-01 35 Avenue Astoria, NY 11106 movingimage.us 718 777 6800 Direct 718 777 6834