Open louh opened 10 years ago
This looks good. So, to make sure that I understand correctly, a user would upload a file for each dataset they want to add to parcels... right?
An example dataset might look like:
Name of data set: Zone Information Description: This data set contains the zoning information for each parcel. Sponsoring department: Planning Department
File: Zones.json
{
"parcels": [
{
"parcel_id": 1234,
"attributes": {
"zone_id": "C-2"
}
}
]
}
Yup, I think that's what we're going with.
I do wonder about the meta data on the dataset itself - should it be part of the JSON? and meta data on the attributes. Should there be a user-friendly name for zone_id
(e.g. just 'Zone')? Should there be something that relates 'C-2' to more relevant information (e.g. a user-friendly name, descriptions, external links to more info, etc.)?
Yeah, I've been thinking about that (re: our discussion with Jack about a spec). Specifically with things like 'building height' we may need additional data on the object like 'units'. This is a rough pass, but maybe something like the following.
{
"id": 1234,
"attributes": [
{ "id": 'zone_id', "value": 'C-2', "type": 'String', "name": 'Zone', "description": 'What is this?' }
]
}
That may be overcomplexifying this, though. I'd be interested to see what @migurski has to say about this. For reference, Mike, basically what we're trying to do is create a standardized way for end users to upload parcel data (which could include a wide array of data sets: e.g. business_classification, zone_id, has_grease_trap, building_height, year_built, etc.) Not sure how introspective we should force users to be about the data they're uploading, but it seems that some introspection (building_height -> units) is necessary. Thoughts?
@rclosner Take a look at this: http://louhuang.com/lv-fastpass-admin/add-data.html