Improving on DIYtraffic

So, the team who pulled together DIYtraffic did a great job of it, and I think it's an awesome beta app. What should be the next steps for making it better, more useful and easier to adopt?

What should DIYtraffic 1.0 look like?

26 Feb15:42

improving on DIYtraffic

By John Geraci

Okay, I'll take first stab at this, because I've been thinking about it a lot as it came together.

I would love to see a version of DIYtraffic that:

1. allowed users to:
- specify which streets/highways they wanted to receive alerts on
- specify which time periods they wanted these for

and then:

2. pushed alerts on those specific routes to their mobile phones during the time periods specified.

That would be a really useful version of this app - would allow you to set it and forget it, and it would still alert you to the traffic issues you needed to know about.

This would be a much larger project than the current app, which is why the team opted not to do it right out of the gate. But maybe as a version 2? Or 3 or 4?

Anything else?

26 Feb15:46

Nice work

By clint

I wanted to say nice work to everyone who built this app, I don't drive so I won't be getting too much use out of it, but the idea is really great. The idea of having one data source where the general public can tweet to and follow is really picking up and I'm excited to see what comes next.

26 Feb16:10

Nice solution...

By CityofShawneeOK

Nice idea for big cities - similar to what I've done with local Police and Fire calls here. Traffic really isn't an issue in our small town but nice implementation.

City of Shawnee Oklahoma - http://www.ShawneeOK.org
Tweeting in Local Government - http://is.gd/i4Xp

27 Feb23:19

traffic.com

By dan.greenblatt

Just as a reference point, another DIYcityer (dave m.) referred me to this site as an alternative source for traffic, as yahoo doesn't serve cincinnati for some reason. traffic.com offers some of these features - it's actually quite impressive:
http://bhelp.traffic.com/getTrafficForYourDrive

BUT - I really like the idea of an open-source, extensible version of this, and there's many interesting directions it can go. For example, if a system knows where people are going and at what times, this could be an enabler to facilitate ride-sharing, or could be referenced with public-transport to help people make decisions as to which method to use on any given day.

Looking forward to working with the DIYcity community to help move DIYtraffic from a standalone app to a platform which makes getting around cities easier and safer for the commuters and less taxing on the environment.

27 Feb23:20

app -> platform

By dan.greenblatt

Just as a reference point, another DIYcityer (dave m.) referred me to this site as an alternative source for traffic, as yahoo doesn't serve cincinnati for some reason. traffic.com offers some of these features - it's actually quite impressive:
http://bhelp.traffic.com/getTrafficForYourDrive

BUT - I really like the idea of an open-source, extensible version of this, and there's many interesting directions it can go. For example, if a system knows where people are going and at what times, this could be an enabler to facilitate ride-sharing, or could be referenced with public-transport to help people make decisions as to which method to use on any given day.

Looking forward to working with the DIYcity community to help move DIYtraffic from a standalone app to a platform which makes getting around cities easier and safer for the commuters and less taxing on the environment.

11 Mar00:59

Web Front-End

By dan.greenblatt

One way I know DIYtraffic could be more useful would be if we could package it with a web-based front end that would allow people to add cities and twitter information to a running instance.

So, for example, we're running a main instance on the DIYcity apps server, it's monitoring chicago, sf and portland - but it would be great if someone could come along, type in their city info and twitter account info, and voila! now that instance is monitoring a different city.

The DIYtraffic bot is set up to handle this (it actually re-reads its config file every hour), there's just no front-end.

I haven't committed the changes yet, but now, DIYTraffic Bot is configured using an XML format, which makes it easy to dynamically set up new connections between inputs and outputs, something like:

<code>

<traffic>
<pipe>
<source type="Yahoo">
<city>New York</city>
</source>

<source type="TrafficDotCom">
<city>Cininnati</city>
</source>

<sink type="TwitterStatus">
<name>diy_twit</name>
<password>nottellingyou</password>
</sink>

</pipe>
</traffic>
</code>

So, the above example would pipe Yahoo traffic info about New York and Traffic.com traffic info about Cincinnati into the same diy_twit TWitter account. This is flexible enough to meet many different needs, the only thing now is developing a library of TrafficSource to account for cities all over the world .

I was thinking the PHP front-end can just spit out this XML file, and then the bot would pick up the changes.

So does anyone want to help build a front-end for this? Shouldn't be too complicated - just some PHP hacking, outputting to an XML format ?!?

20 Mar15:42

diytraffic web-interface

By dave_menninger

I can probably help with this. I'll try to throw together a prototype using the example xml here.

24 Mar16:03

re: diytraffic web interface

By John Geraci

Hey Dave - how this is going? We came up with a really interesting idea for a way to make DIYtraffic more useful, and would love to get that off the ground in conjunction with a web interface. If you're working on this, or still thinking about working on it, let's connect and discuss how best to do this.

If you want you can write me off-list at geraci --> gmail.dot.com.

25 Mar04:55

I'm sorry I haven't had a

By dave_menninger

I'm sorry I haven't had a chance to work on this yet. I ended up being much busier over the weekend than I expected. I have the day off work tomorrow, so I may take a crack at it then.

But if there is anyone else waiting in the wings to work on this, please do.