[RCP-51] DOM Increment

[bryanburgers]

Rendered

A couple of notes.

Field naming

I evaluated the following three sets of names.

• DaysOnMarketReplicationYN/DaysOnMarketReplication/DaysOnMarketReplicationTimestamp/.../CumulativeDaysOnMarketReplicationTimestamp
  • Pro: close to what CoreLogic uses. (Except they use DaysOnMarketReplicationDate, but it was decided in meeting that the group prefers to use timestamps.)
  • Con: this isn't just about replication.
  • Con: CumulativeDaysOnMarketReplicationTimestamp is 42 characters, which would make it the longest name in the Data Dictionary
    Media's OriginatingSystemResourceRecordSystemId is 39
    Property's longest is InternetAutomatedValuationDisplayYN at 35.
• DaysOnMarketIncrementYN/DaysOnMarketIncrementBase/DaysOnMarketIncrementTimestamp/.../CumulativeDaysOnMarketIncrementTimestamp
  • Pro: uses the name of the feature that Paul coined directly in the name
  • Con: CumulativeDaysOnMarketIncrementTimestamp is 40 characters, which would make it the longest name in the Data Dictionary
    • Media's OriginatingSystemResourceRecordSystemId is 39
    • Property's longest is InternetAutomatedValuationDisplayYN at 35.
• DomIncrementYN/DomIncrementBase/DomIncrementTimestamp/.../CdomIncrementTimestamp
  • Pro: uses the name of the feature that Paul coined directly in the name
  • Pro: not overly long for 3 to 6 fields that might show up on every property record
  • Con: abbreviation for DaysOnMarket and CumulativeDaysOnMarket used

I settled on the last set, but I'm certainly open to bikeshedding.

After discussion, option 2 was chosen because spelling out "DaysOnMarket" is more consistent with what is currently in the Data Dictionary.

Testing

Looking through the DD tests, I found a bunch of checks on metadata.

I'm wondering if this spec needs tests at the per-record level, and if so I didn't see how to do that. So I just explained the tests I wanted to see.

Additional material

At this point, there is a bunch of written material in a variety of locations – the closed discussion, some gists that were linked to from the discussion, etc.

I hope to bring that all together into an explainer at some point. I'm not blocking this proposal on the explainer because an explainer doesn't belong in the proposal anyway.

Collaboration

I definitely consider this a first draft. I'm willing to work with anyone who wants to improve it. On the issue it looked like @SergioDelRioT4Bi might be willing to help, for one?

Issue #150

[darnjo]

I hope to bring that all together into an explainer at some point. I'm not blocking this proposal on the explainer because an explainer doesn't belong in the proposal anyway.

Everything relevant to the spec should be in the specification.