191–200 of 248 posts
Earlier quoted context omitted.
Could also signal the area is starved for parking, so the lane should be removed and bicyclists should just ride in the road for those stretches. Small inconvenience to the bicyclists who still get to use the road, big win for the drivers.
It's not a small inconvenience for cyclists, it's a large decrease in safety
Earlier quoted context omitted.
Did the bars close when the new signs went up? Coincidentally or otherwise?
The new signs caused the rate of tickets at the spot to slow down, not stop. According to the charts, there are still 300 tickets per year being written at that one spot. That is huge. It's not a coincidence that that huge number of tickets is being written in a busy bar area.
Earlier quoted context omitted.
This problem would solve itself if vandalism laws were changed slightly.
Wouldn't it be more sensible to deputise willing cyclists for a limited range of offenses?
Wow, great stuff! Did you give more thought into the address cleaning bit? Or does anyone have an idea how to go about transforming mangled addresses into coordinates? I have a problem that's been bothering me for months, similar to what you have here: people from an emergency service call-center are inputting the addresses of the emergencies. For emergencies that happen on the public domain, there is often not a spe…
My current stack is:
1. Send addresses to https://smartystreets.com/ - They gave me a year's worth of unlimited geocoding for free. They also tokenize the addresses, but I had about a 50% success rate with them.
2. Tokenization raw addresses with https://github.com/datamade/usaddress.
3. Use a normalized levenstein distance algo to get ratio of difference.
4. Compare all of the addresses' levenstein distances with each other.
5. Apply logistical regression/gradient ascent algo to tickets by chaining heavilytypo'd addresses to less-typo'd and eventually to a static list of verified-correct addresses.
It works surprisingly well, but there are still a lot of problems that can't easily be solved:
1. Street types (st/ave/blvd/etc) are missing. So, when two addresses have the same street name, it's difficult to pair the two. It's still possible with some probability stuffs and matching the ticketers' paths to the nearest street.
2. Addresses have a LOT of one-off situations. For example, there's a street name called "Avenue A". The street name here is "Avenue", and the street type (usually st/ave/etc) is "A".
3. Lots of four letter streets make levenstein distance very difficult.
Glad you enjoyed it!
Earlier quoted context omitted.
Sure that's a bad example, but he does kind of have a point. In the blog post he estimated it's saved $60,000 in fines. That's a drop in the bucket at the scale we're talking about (a quick Google says the proposed 2018 budget for Chicago was $10.1 billion), but still a decrease. Not only would it cost the government money to employ someone to go through this data and find hotspots like this, someone to go out and ev…
The government is the only entity that does not have to (and never should) follow capitalistic rules. The government is for the people, and capitalism, by definition, is not.
I’m sorry for the confusion, I thought that was blatantly self-obvious.
Earlier quoted context omitted.
Hey! Thanks for posting this, and the kind words. Interesting note about getting data like this - Illinois FOIA allows a requester to submit a SQL as part of their request.. so long as they know the tables and columns within the database ;)
Do you have any specific information about this or the relevant regulations? Perhaps we need a meta-site with best practices and how best to attack open records requests in general.
This is awesome. Question though: how is producing license plate data like this not a disallowed privacy invasion? It seems like you could totally track who's parking where and potentially do nasty stuff, if you know (say) someone well-off whom you don't like and who doesn't seem to mind getting tickets on a regular basis.
Earlier quoted context omitted.
What are "capitalistic rules"? Governments absolutely should be looking to make profits, it's just that they don't necessarily have to make them in dollars. So say you were able to objectively measure the value a program produced (costs are usually already known). If a program costs 10 units for every unit of value it produces, maybe it isn't a good program. If it produces modestly more value than it costs, it's maki…
What are "capitalistic rules"? In this case, the dogmatic assumption that every endeavour should be financially profitable.
I was in no way implying the government was trying to make money off increased fines (quite the opposite with the last paragraph), simply that it would very likely end up costing them more than it saved the taxpayers to support such an initiative at a larger scale, and that would have a very nebulous gain.
Can you imagine being the Mayor of your department at work and proposing to your board of directors a multi-million dollar budget for next year that includes a huge carve out for evaluating all the petty fines and late fees you collected from customers because it’ll make them happier?
Goodwill is one thing, but who is ever going to approve that?
Wow, great stuff! Did you give more thought into the address cleaning bit? Or does anyone have an idea how to go about transforming mangled addresses into coordinates? I have a problem that's been bothering me for months, similar to what you have here: people from an emergency service call-center are inputting the addresses of the emergencies. For emergencies that happen on the public domain, there is often not a spe…
I've put a LOT of thought into address cleaning! And yep - levenstein distance seems to be the way to go. My current stack is: 1. Send addresses to https://smartystreets.com/ - They gave me a year's worth of unlimited geocoding for free. They also tokenize the addresses, but I had about a 50% success rate with them. 2. Tokenization raw addresses with https://github.com/datamade/usaddress . 3. Use a normalized levenst…
I already have some preliminary data - in a city with 350k inhabitants, they gave 150k fines last year, totaling 2.5 mil EUR. I can't wait to search for the hotspots