Merge Points tool should have option to keep Object ID of point with lowest Object ID but other attributes and location of point with highest ObjectID

[Jessica_Watson]

Currently when the Merge Points tool merges two points together, you can select (1) the Object ID and attributes to preserve and (2) the point geometry/location to keep.

For our organization, we need the tool to (1) preserve the lowest Object ID, but (2) keep the attributes of the highest Object ID and (3) keep the geometry of either the highest or lowest Object ID (have an option but for our case we need to keep the geometry of higher Object ID).

We currently produce a product that shows change over time using unique IDs and we are working with data ingestion of points in ESRI Parcel Fabric for spatial upgrade and attribute annotation purposes. On top of that, new points with attributes are created during our parcel capture processes. So as much as possible, we are trying to not overwrite the existing fabric points. Currently if we use the tool to preserve the lower Object ID, the attributes of the higher Object ID are lost. These attributes are important for us to link up information such as reference and control mark types, status, location, and boundary point names (and other attributes) for matching up with subsequent survey plan information and preserve lineage of change over time.

In order to preserve parcel, line and point lineage over time we need a persistent point id, but we don't want to lose the new attributes. We would like for the Merge Points tool to be enhanced with an option to preserve the attributes from the higher Object ID and the location from higher Object ID while preserving the ID from the lower Object ID to protect lineage.

[AmirBar-Maor]

@Jessica_Watson 

I am curious about which organization you belong to because not many organizations manage point lineage.

Regardless:

Object IDs (or GlobalIDs) are used in relationship classes. So if your points feature class have attachments or any other relationship class, it will use the primary key of the point to establish the relationship.

That means that if you choose one point over the other, you effectively choose to maintain its related data as well(if it has a relationship class).

In general, we favor keeping the oldest point. Why? The Generate Parcel Fabric Links geoprocessing tool compares the location of the same point in 2 different moments in time. For it to be the same point it needs to be the older point that survives workflows like alignment.

The ability to preserve one point and use the attributes from a point that will not be preserved is already supported from version 3.3 and was demonstrated in this meetup.

Does this meet your requirements?

Did you consider using the geoprocessing tool Import Parcel Fabric Points?

[Jessica_Watson]

Hi @AmirBar-Maor ,

I am posting relating to the work we are doing here are Department of Natural Resources and Mines, Manufacturing and Regional and Rural Development, (formerly Department of Resources) In QLD, Australia. We are responsible for managing our states spatial cadastre and survey control and in the process of moving the maintenance of our states spatial cadastre to parcel fabric (live mid next year – 3,713,757 parcels and counting). We’re currently testing our data model and business processes around data capture and ETL.

Regarding ‘In general, we favor keeping the oldest point. Why? The Generate Parcel Fabric Links geoprocessing tool compares the location of the same point in 2 different moments in time. For it to be the same point it needs to be the older point that survives workflows like alignment.’:

Yes, this is what we need – to track the movement of the point over time. To do this we need the point to be persistent and preserved, and this is the reason for needing to keep the older ID but newer attributes (and newer location). This is so we can keep our points up to date but preserve the lineage. We currently have a service of a temporal cadastre that shows changes over time to the location and accuracy the cadastre as well changes to parcels from subdivisions.

Regarding ‘The ability to preserve one point and use the attributes from a point that will not be preserved is already supported from version 3.3 and was demonstrated in this meetup. Did this meet your requirements?’’:

While the addition of the option to select attributes from one and points location from another is great new addition, unfortunately it doesn’t meet our requirements to keep the original point Object and Global ID (plus our own managed UniqueID and FeatureID) but the remaining attributes and location of the new one. Basically, we are wanting to keep older points but update it with the location and attributes of the new one to preserve latest info about the point, such as the type of physical mark in the ground, its condition etc. Since it’s the same point we would lose lineage if we were to keep new location and attributes only. And we would lose attributes from the new point if we kept old ID and new location only.

Regarding ‘Did you consider using the geoprocessing tool Import Parcel Fabric Points?’:

For our points, many of them are boundary points for a parcel. These would have a different name for each new captured plan, so we are unable to use it based on point name. While sometimes we are importing points from ETL processes, other times we are just matching what we have captured manually from for example traversing new parcels and this is already in parcel fabric – the matching is a manual process using the ‘Merge Points’ tool but means we would lose either attributes or lineage. As ‘Import Parcel Fabric Points’ relies on data imported from another dataset, this would not work for our align processes when data is already in the same Points feature class. Also, for one-by-one merges of reference marks for example, we would need a tool where we can select 2 points and merge, keeping the oldest one.