Open louh opened 10 years ago
Looks good. I didn't notice this until yesterday, but the dummy data is good for now. So what do we need to do in order to get non-dummy data. We have a naics code, all we need to do is match it to a business classification id and a planning use code... right?
Yup.
We have (somewhere - I need to find where I put it) NAICS matched with business licensing already, which Richa wrote a scraper for and have it formatted. The planning data set, unfortunately, is completely unusable. In which case I'm just probably going to build a slightly better dummy set myself, and use it to lead Doug in the right direction. I will follow up soon with what I've got.
That sounds like a good start. That'll be one of the next priorities... matching logic for NAICS -> Business Licensing.
@rclosner Now that we are back up and running in a stable state I think this is my next priority.
Awesome. Totally agree. Then the first few steps will be totally done.
Ryan — Sent from Mailbox for iPhone
On Wed, Jan 8, 2014 at 4:44 PM, Lou Huang notifications@github.com wrote:
@rclosner Now that we are back up and running in a stable state I think this is my next priority.
Reply to this email directly or view it on GitHub: https://github.com/codeforamerica/fast_pass/issues/79#issuecomment-31892445
The initial NAICS to business licenses matching - which @richaagarwal created by scraping the latest business licensing CSV that we have from Las Vegas - is here: https://github.com/codeforamerica/naics-api/blob/master/data/business-cats.json
It's rough and represents the snapshot of Las Vegas data - and not a "one to one" match of NAICS to license codes:
Sample data
"444130": {
"business_categories": {
"M50": 1,
"A02": 1,
"H02": 1,
"B50": 427,
"M05": 1,
"G50": 4,
"L06": 1,
"R50": 1
},
This means, if someone selects NAICS Code 444130 (Hardware stores) from Step 10, we can look this up and see the history of business licenses assigned by the city. The vast majority of them (427 licenses out of 437 total) are assigned to B50 (Building, Plant Nursery, and Hardware Supplies), which seems correct to me. For the purpose of seeding our initial data set, this seems to be a fairly good way of capturing what's likely to be a one-to-one relationship.
@rclosner Using placeholder text for now, so not linked up to the controller. Thoughts?