Open ZachBeck opened 11 months ago
It seems a little odd to me to use the same table name in the deprecation and addition since any existing references will still be correct but the schema will be completely incompatible. I'm not sure what the best advice is on the naming since the current name is ideal but it should change.
Definitely not ideal but we have done it in the past with Soils. I haven't been able to dream up a name that's more descriptive than HealthCareFacilities.
We had a similar conversion with soils and the situation was a little different. This is worth a group discussion to come up with a consistent path forward for this occurrence and all future ones.
I added a discussion item in tomorrow's team meeting.
@steveoh do we have any metrics on open-sgid tracking what gets used and how often? In deprecation/addition cases where the schema changes but the name stays the same we decided to add _deprecated
to the feature class name as part of the soft delete process.
do we have any metrics on open-sgid tracking what gets used and how often?
We sorta do but no there isn't. Everyone has access to view all the data available for the cloudsql instance. To read between the lines, there should be a consistent process independent of popularity.
In deprecation/addition cases where the schema changes but the name stays the same we decided to add
_deprecated
to the feature class name as part of the soft delete process.
I thought the soft delete unshared the item/deleted the table? What does renaming the table accomplish differently? Would that only work for open data/arcgis online?
The open sgid connection will break during the soft delete, like it normally would, and will look correct after the same table name is added again. How does _deprecated
address this issue?
Any api search requests or open sgid layer in a project that goes unnoticed, and let's be honest, our reach for these deprecations is not great, will query the same table the entire time. The error message in the api will change from this table doesn't exist, which is correct, to this attribute doesn't exist on this table, which is confusing. I'm not sure what pro/map/qgis/etc would do. This seems like a poor solution that I can't support since it doesn't cover all of the sgid.
Suffixing the table with v2
seems like a better alternative. Or keeping the original schema and adding the new columns to not create a breaking change is an option. I really don't think there should ever be a deprecation and addition with the same name. It's not consistent enough.
I think a deprecation/addition with the same name, though not ideal, is a fact of life. To me having intuitive naming conventions that make sense is better in the long run and make up for whatever short term disruption that a user experiences.
Adding _deprecated
to the name in open-sgid is a way of letting the user realize that there has been a change in the data and discover the new version when they repair the link. Not ideal, but at least they are presented with the new vs old feature class.
I think appending new fields on to the existing schema isn't a viable solution. That leaves us either trying to maintain fields that no longer exist in the stewards data or having columns in the table that only have null values.
It sounds like this needs to be discussed at the next dev/data meeting.
check | status |
---|---|
@ZachBeck has completed 0 out of 6 tasks | :no_entry: |
@steveoh has completed 0 out of 4 tasks | :no_entry: |
@stdavis has completed 0 out of 1 tasks | :no_entry: |
@jacobdadams has completed 0 out of 1 tasks | :no_entry: |
@gregbunce has completed 0 out of 1 tasks | :no_entry: |
@rkelson has completed 0 out of 1 tasks | :no_entry: |
check | status |
---|---|
@ZachBeck has completed 0 out of 6 tasks | :no_entry: |
@steveoh has completed 0 out of 4 tasks | :no_entry: |
@stdavis has completed 0 out of 1 tasks | :no_entry: |
@jacobdadams has completed 0 out of 1 tasks | :no_entry: |
@gregbunce has completed 0 out of 1 tasks | :no_entry: |
@rkelson has completed 0 out of 1 tasks | :no_entry: |
check | status |
---|---|
@ZachBeck has completed 0 out of 6 tasks | :no_entry: |
@steveoh has completed 0 out of 4 tasks | :no_entry: |
@stdavis has completed 0 out of 1 tasks | :no_entry: |
@jacobdadams has completed 0 out of 1 tasks | :no_entry: |
@gregbunce has completed 0 out of 1 tasks | :no_entry: |
@rkelson has completed 0 out of 1 tasks | :no_entry: |
check | status |
---|---|
@ZachBeck has completed 0 out of 6 tasks | :no_entry: |
@steveoh has completed 0 out of 4 tasks | :no_entry: |
@stdavis has completed 0 out of 1 tasks | :no_entry: |
@jacobdadams has completed 0 out of 1 tasks | :no_entry: |
@gregbunce has completed 0 out of 1 tasks | :no_entry: |
@rkelson has completed 0 out of 1 tasks | :no_entry: |
check | status |
---|---|
@ZachBeck has completed 0 out of 6 tasks | :no_entry: |
@steveoh has completed 0 out of 4 tasks | :no_entry: |
@stdavis has completed 0 out of 1 tasks | :no_entry: |
@jacobdadams has completed 0 out of 1 tasks | :no_entry: |
@gregbunce has completed 0 out of 1 tasks | :no_entry: |
@rkelson has completed 0 out of 1 tasks | :no_entry: |
Adding blocked since work hasn't started on this yet.
Waiting on updated facilities from Health.
check | status |
---|---|
@ZachBeck has completed 0 out of 6 tasks | :no_entry: |
@brigsz has completed 0 out of 1 tasks | :no_entry: |
@steveoh has completed 0 out of 4 tasks | :no_entry: |
@stdavis has completed 0 out of 1 tasks | :no_entry: |
@jacobdadams has completed 0 out of 1 tasks | :no_entry: |
@gregbunce has completed 0 out of 1 tasks | :no_entry: |
rkelson has completed 1 out of 1 tasks | :+1: |
This does not yet have a record series, I will find out how to create one
check | status |
---|---|
@ZachBeck has completed 0 out of 6 tasks | :no_entry: |
@brigsz has completed 0 out of 1 tasks | :no_entry: |
@steveoh has completed 0 out of 4 tasks | :no_entry: |
@stdavis has completed 0 out of 1 tasks | :no_entry: |
@jacobdadams has completed 0 out of 1 tasks | :no_entry: |
@gregbunce has completed 0 out of 1 tasks | :no_entry: |
rkelson has completed 1 out of 1 tasks | :+1: |
check | status |
---|---|
@ZachBeck has completed 0 out of 6 tasks | :no_entry: |
@brigsz has completed 0 out of 1 tasks | :no_entry: |
@steveoh has completed 0 out of 4 tasks | :no_entry: |
@stdavis has completed 0 out of 1 tasks | :no_entry: |
@jacobdadams has completed 0 out of 1 tasks | :no_entry: |
@gregbunce has completed 0 out of 1 tasks | :no_entry: |
rkelson has completed 1 out of 1 tasks | :+1: |
check | status |
---|---|
@ZachBeck has completed 0 out of 6 tasks | :no_entry: |
@brigsz has completed 0 out of 1 tasks | :no_entry: |
@steveoh has completed 0 out of 4 tasks | :no_entry: |
@stdavis has completed 0 out of 1 tasks | :no_entry: |
@jacobdadams has completed 0 out of 1 tasks | :no_entry: |
@gregbunce has completed 0 out of 1 tasks | :no_entry: |
rkelson has completed 1 out of 1 tasks | :+1: |
Summary
This will replace the existing, soon to be deprecated SGID.HEALTH.HealthCareFacilities. This data has been geocoded from licensed health providers and will have the same source as the current HealthCareFacilities but a different schema that will be easier to maintain.
The data should be available in
1 Check [x] all the areas where you expect the data to show up.
The data is of high quality
2024/07/02
)2024/07/02
)2024/07/02
)Where is the data source
Choose one.
Action items
name
with their github@name
.2024/07/02
)2022/00/00
)~2022/00/00
)2024/07/02
)SGID.META.AGOLItems
record (@ZachBeck2024/07/02
)static
orshelved
item (name, completed:2022/00/00
)2023/00/00
):robot: Automation validation
name
with their github@name
.2020/01/01
when the task is verified.2022/00/00
)2022/00/00
)2022/00/00
)2022/00/00
)2022/00/00
)2022/00/00
)Notification
Group Task Assignments