How We Built an iOS App, an Android App and a Node.js API in 20 Hours

http://blog.semantics3.com/how-we-built-an-ios-app-an-android-app-and-a-node-js-api-in-20-hours/

by govind201 • 14 years ago

111 34 14 years ago

blog.semantics3.com

I congratulate them on acheiving your goal of writing an app in such a short time, but I whole-heartedly condemn this app's principle idea. Far from being a force for good it feeds into the incredible paranoia that seems to be prevail in the United States, that everyone is out to get them, and that there are such things as "good areas" and "bad areas".

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.

govind2013• 14 years ago

The app was born out of personal experience. Within a month of moving to Philadelphia, I was assaulted on two occasions, including once on my very first day there. I put that down to my lack of knowledge of the area; I later discovered that a more knowledgable person wouldn't have accidentally wandered off at that time of evening to that particular area. Turns out, some of my neighbors had had similar experiences in the past in the very same area. I was later warned by local authorities that the area was known for the phenomenon of "gentrification".

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.

daviddoran• 14 years ago

I see both sides here. My main concern with this approach is the false sense of security it might create. In my experience , the separation of cities into "good" and "bad" areas is only valid in aggregate. I'd rather someone took precautions and stayed alert than happily sauntered through an area that the app told them was safe. I guess it's difficult to know if this strategy works, since you'll never know if you avoided a would-be encounter with a mugger.

With that said, I can't fault your motivations, and building it so quickly was a real achievement!

kalkat• 14 years ago

Ah, the perils of "having to" listen to the feedback!

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.

crikli2• 14 years ago

Build and release the app and ignore the whiners. All that you're doing is delivering notification to your users when the statistical likelihood of sustaining harm increases. If people get offended by the macroscopic implications of mathematics that's their issue, not yours.

rwhitman• 14 years ago

Having been raised in the Philly area I've heard this story from non-natives a number of times, so its pretty common.

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.

batista• 14 years ago

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.

chris_dcosta8• 14 years ago

I made that comment, I am British I lived for many years in the poorest area of London, and I've also lived in the Banlieue de Paris. I would not class these as "dangerous".

That's an outsider's perspective, but the reality is considerably different.

Tichy2• 14 years ago

Maybe you can criticize the algorithm, but I doubt the notion that there are more and less dangerous areas is under debate.

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?

chris_dcosta8• 14 years ago

I think the point is exactly that.

caw• 14 years ago

What's the time function on your graph? Are you only showing the last X weeks or all time? Things like police patrols could affect the frequency of crime over time, and that may change your predictions and confidence intervals.

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.

vinothgopi2• 14 years ago

We managed to gather data for about a month when building this app. Some of them were as recent as a few hours and some a few days.

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 :)

crazygringo• 14 years ago

Is this available on the App Store yet? Or is there an estimated date?

vinothgopi2• 14 years ago

The designer of the iOS app is sadly a perfectionist and is refusing to put it up on the store until its pixel perfect!!(by his own standards)

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?

crikli2• 14 years ago

I travel a fair amount for business and have contacts that also travel quite a bit. Let me know if you'd like some beta testers once you get beyond the California area.

rdg• 14 years ago

Great, give clueless "project managers" another reason to think anything of value can really be accomplished in a day or two.

varelse• 14 years ago

Tip of the hat: you beat me to the punch (though I really wondered about the proper lawyerese for deploying a service like this)...

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.

Czarnian• 14 years ago

The feat of building that kind of app is impressive. I sure as heck can't do it, yet. I do have a few comments, though.

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.

donpark• 14 years ago

I had similar idea whileback so this app is interesting.

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.

How We Built an iOS App, an Android App and a Node.js API in 20 Hours

Related Stories