Output Format Suggestion - Table Rows vs. Records

The format used for table rows differs between when a set of rows is returned (i.e. as a table) vs. when a single row is returned (i.e. as a record).

For symmetry (and so simplicity of parsing), what would be the chances of these two formats aligning, so that the same format is used for a row regardless of whether it is in a table or returned as a record?

Details

When a table is output, a JSON array is returned. Each row is represented as a JSON object, with the column names mapping to property names and column values to property values.

// #table({"Col1", "Col2"}, {{"ABC", 123}, {"DEF", 456}}): 
"Output": [
    {
        "Col1": "ABC",
        "Col2": 123
    },
    {
        "Col1": "DEF",
        "Col2": 456
    }
],
"RowCount": 2

When a single row from a table is output (a.k.a. a record), instead of a single JSON object being returned for the row (as is used above for rows), the output format changes to a JSON array with each record field as a separate key-value style JSON object:

// Table.First(#table({"Col1", "Col2"}, {{"ABC", 123}, {"DEF", 456}}))
"Output": [
    {
        "Name": "Col1",
        "Value": "ABC"
    },
    {
        "Name": "Col2",
        "Value": 123
    }
],
"RowCount": 2

No files, tests, or entry points are named. Start by tracing how the SDK serializes table results and single records, using the two JSON examples as the current behavior. Confirm the desired common row representation and document or test the agreed output before changing implementation.