How We Built an iOS App, an Android App and a Node.js API in 20 Hours
blog.semantics3.com
blog.semantics3.com
There are huge problems in justifying this kind of classification, not least because ordinary law abiding citizens who are in a majority also live and work and play in so-called "bad areas" and never suffer crime. What becomes of their house values if such an irresponsible app were to become popular? Their (very) hard earned cash paid as mortgage payments gone to waste.
This app is pernicious in every sense. Even wealthy well-to-do areas can become ghetto-ized: If you tally up the number of burglaries or attempted burglaries in these areas you might find the stats show a different image.
We are not living in some post-apolcolyptic movie world, this is real life and has hard working real people in it, with families and children, who have no choice but to live where they can afford to, and I don't see how this app could ever be considered as "making the world a better place". For who exactly?
This kind of app threatens to push people who live in the targeted areas further down the food chain, and I just wish the developers would have been a bit more socially responsible, in their thought processes.
Rant over. Hopefully a lesson learnt.
Personally, I was scared out of my wits after those two incidents and landed up staying home many evenings for fear of a repeat incident. I think such an app could've really helped the general quality of my life in terms of reducing my apprehension.
I do see your point. To be honest, I didn't give the macroscopic windfalls excessive consideration. I had focused rather on the gains to individuals who stand to face the same situations that I did.
The app has not been released publicly so I think I'll do read a bit about crime patterns and their correlation to demography before taking any further step. Thanks for brining this to attention.
With that said, I can't fault your motivations, and building it so quickly was a real achievement!
If there's anyone out there with experience in sociology, I'd be glad to have a chat.
There is absolutely nothing irresponsible about your idea. You are not the one terming something good and bad. If the data says so, if people say so, if the reports and incidents so, then it is just right that somebody let that be known, so that stakeholders can make an informed decision.
A local is usually going to know what are the spots to avoid at what time of the day, it is the outsider who does not. All you are doing is giving that same information to the outsider that the local has, and I think that is completely fair. If in the process, a particular neighbourhood or area shows up as unsafe, then that is nothing but the hard truth becoming known. Hard, but the truth. People need to learn to live with it.
You wrote the program, you set the thresholds, you are responsible. You can't absolve yourself by saying "the computer did it". It wouldn't stand up in court.
In fact most of the population of the world actually live in "bad areas" and would no doubt feel pretty insulted to find an app that categorized their 'hoods as a "no-go zones".
Let me put it another way, this app would be a reminder of how poor you are. I wouldn't want that - I can't say I know anyone who lived there would either.
My mother works as a railway attendend and often has to go to different cities without guides just to get to her hotel. I'd love for her to have an app or something that guides her on the safest path possible.
If you live in a bad area you shouldn't be ashamed about it, but ignoring it just makes things worse for everybody.
The point I wanted to make was this:
If the app's aim was to draw attention to areas that needed help, for use to focus the minds of governments and aid agencies, and people with a social conscience, and they were using the data to determine the most needy areas in an aggregated fashion it could be positive thing. Where can we help next?
Do you see? This would be making the world a better place.
It is not doing this.
It is basically contributing to the view people from certain areas should be avoided, because statistically speaking they are mostly criminals.
This is not for the good of the many, it's for the good of the one.
In other words: walk the walk, don't just talk the talk.
In other words: walk the walk, don't just talk the talk.
Edit: Oh, and complaining about downvotes isn't worthwhile either.
These guys built a mobile app and node API in 20hrs based on something that - they believed - is general common-sense among any city dwellers (avoiding bad neighborhoods).
If you think you know a better way of approaching the problem, then the general response in the hacker community is: build it yourself.
Asking to change wording to something less offensive is fine... but calling them out for not being socially concious and not trying to fix a problem that has affected cities for centuries is a bit of a stretch... especially considering it was a weekend project.
Like the previous poster said, if you don't live in the area, then it's probably smart to not walk around some of these areas. Just because other people live there isn't some reason why this app shouldn't exist.
The point is, just because some people have to live in these areas isn't a reason why people who are fortunate enough to not live there should be forced to go there and put themselves in a statistically more dangerous situation.
Would you walk around most areas of Detroit or the east side of Cleveland, at night, alone?
That said I think the flaw in this from a positioning perspective is it shouldn't be an alarm that says "Danger, get out now!!!" like you're in a biohazard zone, its should be positioned more like a "Street Sense" radar.
Basically you learned urban street smarts in the hardest way possible. Make this more about softening the risks of navigating unfamiliar urban areas by giving folks the same level of street smarts as a native, and less about a shock and fear-based motivator and you'll be on the right track.
I would also consider focusing on safety by highlighting safer routes pointing out things like density of businesses and street lighting, police stations etc.
The problem is that, besides the paranoia, the problem is TRUE in many parts of the United States.
Most Europeans wouldn't understand the extent of that --well, some would, like British or French, which have similar "dangerous" areas.
That's an outsider's perspective, but the reality is considerably different.
Some poor neighborhoods really are dangerous.
What would not having such an app achieve - that people who are too naive to do any research move into crime areas, thereby smoothing out the crime stats? Is that really a viable solution?
Refine your data classification. On one of your screenshots, you are listing "Arrests" as a type of crime. That's confusing. Also, consider filtering for violent crime rather than crime in general. That would be more in line with the stated purpose of your app.
Realign the usage from a real-time application to a planning application. You don't want your users to be fiddling with a smartphone in a bad neighborhood. That really increases the odds of getting mugged and defeating the purpose of the app.
I agree with some of the other posters that you should tone down the alarmist tone of the app. If nothing else, it's more professional.
Congratulations on your experiment.
EDIT: Depending on your data freshness, you may want to do something like push notifications if a crime happened in your area in the past 4 hours. That way if someone walks through a "good area" but there's recent crime they should be more vigilant.
We implemented a weighted average algorithm to calculate the risk levels and how confident we were of the data. For the risk levels I took into account 1) How recent the data was 2) What time of the day it occurred relative to the time of query (for eg: the crime occurred at 4pm on another day and its 5pm now, it will be given a higher weightage compared to something that happened at say, 11am.) 3) The severity of the crime itself (murders >> theft) 4) Number of occurrences of crimes in the area
The assignment of weights wasn't thoroughly scientific though. It needs some tweaking.
And thanks for the suggestion on push notifications :)
What I would change:
1. Remove crime details. They are unnecessary. 2. Replace pins with transparent canvas overlay showing level of danger or safety in shades of red and green. 3. Let user set the desired 'shade of danger'. A simple horizontal or vertical color selector showing shades of red will do. App should buzz every N minutes when users is in area with deeper shade of red unless they press the override button.
That's all.
Of course, the corresponding app, badguy, which alerts would-be criminals to the likely patrol routes of cops, is likely to follow shortly thereafter from the black hats.
In all seriousness, we have a bit of work validating the ranking algorithm, pulling in more sources and general UI cleanup stuff. We are really keen on putting it on the store though. Maybe a couple of weeks to a month?