The software and hardware behind parking tickets with Jim Suttles of Republic Parking Systems

Transcript

Ledge: Jim, thanks for being here. Good to have you.

Jim: Well, thanks for having me. Thanks for the invitation.

Ledge: Awesome. Fantastic. Can you tell the listeners your two, three minute story of yourself and your work over the course of your career?

Jim: I’ve been in IT and software development for a little more than 25 years now. Started out a long time ago with mainframe work, COBOL and CICS and all that kind of stuff. Migrated as things changed over time, I changed with it and went into HTML and ASP, and then to .NET and SQL Server and all that good stuff.

Now I lead a small development team. We build parking related apps, public parking system, mainly with enforcement. Writing parking citations doesn’t make us popular. We don’t try to. Not too many people know that’s what we do when we’re out on the street, but it’s a good industry to be in.

When I first started with Republic, back in 2005/2006, I wasn’t sure that it was going to be very interesting work because I just didn’t know what kind of technology could go into the parking business. But we’ve seen over the last 10/12 years a lot of changes in the industry. A lot of automation and a lot of really cool technology with cameras and phones and geofencing and all that good stuff.

Ledge: I’m sure nobody jumps into their career and goes, “Hey, I’m really hoping I can work on parking enforcement technology.”

I thought that was a great part of your story when we talked before the interview. That you have all these really excellent opportunities to apply cutting edge technology, and doing just all kinds of interesting and innovative things with license plates and the virtual chalking and geofencing. Just the different ways that you’ve modernized some of the pieces of parking enforcement and put them in the background, that nobody really realizes that they’re there because they work so well.

I’d love if you’d walk through some of the product development that you guys have done.

Jim: We started out a few years ago just building an enforcement system because we, as a parking management company, had to buy those systems from other vendors. We made a decision to write it ourselves. We could build it the way we wanted it to be built and not have to pay somebody else for it.

So we started out writing the system with Windows tablets as the hand-held, and then a back-office site to do all the management of it. Over time we’ve migrated that to Windows Phone, and then when Microsoft lost interest in Windows Phone then we moved it on to iOS and Android – and Xamarin now with the citation writing.

That’s turned out to be a neat project. We built the hand-held up with Xamarin so we could have one code base for iOS and Android, and that’s worked out really well.

We use Bluetooth Low Energy printers so we don’t have to the printers and we save some battery on the printer and on the phone, so that really helps.

We go through a lot of volume. Many of the parking lots we just scan all the plates in the lots with LPR, License Plate Recognition. We just drive the streets and drive the lots looking for plates that haven’t paid and we, just an automatic process, issue the citations. It get us through a lot of volume.

The main goal for our managed lots is really just to get people to pay to park. Not to try to trick them into some kind of citation, but they know we’re there, they know we’re enforcing, then they pay to park.

Ledge: Talk to me about the driving around with the cameras and stuff. That’s got to be an interesting setup. How does that all get processed? It sounds like it would be a tremendous amount of video or still shots.

How are you chunking through that? There’s got to be some kind of wireless component. How does that all fit together?

Jim: We have mountable cameras on the vehicles. Many times two, sometimes as many as four cameras. That we can drive the streets 30/40 miles an hour if we wanted and capture every plate that we pass.

We have zones geofenced so we know if a plate belongs in that zone or not. If it’s a permit type situation where they pay for the day or pay for a few hours, or they pay by plate and we know if they belong in that zone. So we know they can’t pay for one lot and then park in a better lot somewhere close to an event or something.

We have a laptop in the car that captures every photo as it’s taken – and it is just still photos, not video. We capture every photo. Then if we get a hit that this plate does not belong there or has been there too long, it’s been there overtime or whatever, then we stop, issue a citation with the iPhone.

That is automatically uploaded to our SQL databases in Azure. At the same time, we go to a little process on the laptop in the car, in the OPR vehicle, that will look for that plate in the local database on the laptop and find any photos that the LPR cameras took of that vehicle. Then upload those with the citation up to the server. Those are all gathered all together automatically.

At the end of the day, all the photos that we didn’t need, which is the vast majority of them on the laptop, are discarded. If they didn’t result in a citation, we don’t need them.

Ledge: So the local laptop does all the image LPR processing. How heavy is that a load on the local setup?

Jim: It’s really not bad. It’s just your mid-range laptop. Not a super high-power laptop. It has cell connectivity built in so we’re getting the images up real time to the servers, but it’s not really a lot of processing.

The LPR system has kind of a black box to do some of the heavy lifting and take if off the laptop for the image processing. So that helps a lot.

Ledge: Is the LPR a separate unit from the processing and laptop itself, when you say black box?

Jim: Yeah. There’s cameras and there’s, like a I said, a black box and networking and that kind of thing that we get all together from a Canadian company, Genetec, provides those LPR solutions.

Ledge: You didn’t have to home-grow the LPR.

Jim: That’s right. We just integrated with it.

Ledge: So it’s a virtual cloud, if you will. Driving around with a private cloud in each car.

Jim: That’s exactly right. Yeah.

Ledge: That’s interesting.

Jim: You can go through a lot of scans.

Ledge: Did you guys have to intuit all these… I’m thinking, edge case upon edge case upon edge case to solve for all these different dimensions. Did that exist before and you just did it better? Or were you able to think of new solutions to the same enforcement problems?

Jim: Well, for the most part, it’s things that the parking industry has done manually forever and we’ve just brought it up to a new technology.

One of the things we talked about in our chat the other day was chalking. In the old days, they would literally take a piece of chalk and swipe a tire that was parked in a time-limited parking spot. If it’s two hours, they would come back more than two or more hours later and see if that chalk mark was at the same spot. If it was then they issued a citation.

We do the same thing now. They call it digital chalking. If it’s not an LPR situation but it’s a small lot that guys are just walking around on foot, they record the plate and the valve position – the tire valve position like on a clock. 1 o’clock, 2 o’clock, 9 o’clock. They just record the valve position, come back a couple of hours later. If that plate is still in the same spot with the same valve position, it’s not moved, they get a citation.

Ledge: That’s amazing. You just had to think of it sort of, what marks the right spot. Just to take yourself a snapshot of time based on the position of the vehicle.

Jim: That’s right. The geofence thing is an integral part of it because there are some neighborhoods that people can park on the street. They have neighborhood parking and only the plates that live in that neighborhood can park there. They can buy visitors passes if they have visitors come over and they’re allowed to park. But we geofence that area so that only the plates that should be in that geofenced area are there.

There’s a lot of good applications for that. I saw something on the internet just this morning about Burger King selling Whoppers for a penny if you bought the Whopper through the BK app and you were in a McDonalds parking lot when you bought it.

They said they had geofenced all the McDonalds, and if you opened up the BK app in a McDonald’s parking lot you would be presented with an offer to buy a Whopper for a penny, and then you’d be navigated with the to the nearest Burger King.

Ledge: To go pick it up. That fun good old marketing.

Jim: So that’s the kind of thing that we chatted about a little bit the other day. There’s a lot of these customer facing apps that there’s a lot of possibilities we’ve not thought about yet, of using all the data that we have.

Ledge: Right. You have to deal with this proliferation of parking and payment apps. There’s no standard bearer in this industry. Every city I go to has a different, “Download our app and pay for the parking.”

How do you create that massive ecosystem? Do they all have APIs? How are you dealing with that?

Jim: They do. They have APIs that we have to integrate with but, like you said, every time we go to a new city as a parking management company, almost invariably there are new apps that we haven’t integrated with yet because they’re just popping up everywhere.

One of the things that we are working on is an app to gather all those different APIs into one tunnel, that we can just query that one place instead of 50 different APIs that we have to query every time.

Ledge: You could almost have like the global standard API for parking payment apps, then all all other business that you could sell.

Jim: That’s exactly right because there are so many that are trying to get into it. There are a few companies that have been in the parking management business for a long time, but there’s a lot of just tech companies that are popping up to get in on it, to build these apps, and if they sell that to a city, then we have to integrate with it.

Ledge: Do you have an opportunity to reach out preemptively to those companies and do a, hey, you need us because we handle all the real heavy lifting backend?

Jim: We have in some cases, but it’s just something that you’re always chasing because even we in the industry are surprised of how quickly new ones pop up. They get a lot of venture capital funding, and they get those apps out there quickly.

Ledge: And is there enough room for everybody, or is this like a winner-take-all type of situation where you just have to kind of watch the market evolve? It probably doesn’t cut out what you’re doing at all. It’s really just other people’s money paying for you guys to have opportunities.

Jim: That’s right. I think it can get oversaturated. I think it can be kind of hard on a customer if, like you said, every city you go to there’s a new app for you to download to pay for your parking. Then you wind up with 20 apps on your phone. I think it will shake its way out where there won’t be so many.

Ledge: Sure. Some kind of industry consolidation. So your play is that you’re going to have all the data no matter what.

Jim: That’s right.

Ledge: So you control the whole hose to get all the pipes together, and you can just let them collect data for you.

Jim: Yeah. One of the things we did we talked the other day was writing the iOS and Android app. That we wrote it with Xamarin. One of the things you asked was, was that hard to find Xamarin developers?

My view on that is, I don’t look for Xamarin developers. I feel like if I have A+ guys – which I have a small team but they are all top notch guys – they can switch to a new language very very quickly. If they know how to code and they’re not tied to a particular language and it’s all they know how to do, then it’s not a problem to switch.

To me that’s a better answer than, for myself, always trying to chase around whatever the newest technology is and find an expert in that new technology, because then I’ll always be searching for guys.

Ledge: Right. And that’s a great segue to the question I always ask last to anybody I talk to. Is, how do you evaluate engineers?

Our business is evaluating and bringing to bear just absolutely A+ engineers that are excellent at building software, hit the ground running. Check all kinds of boxes, and we have a pretty robust heuristics for that, but I like to ask engineering leaders like yourself, what are your heuristics in how you measure, when you’re bringing somebody on, choosing the right person?

Jim: I like to look for guys that they live tech. I’ve run into quite a few guys over the years that they just go into it because they think it’s a good job, they can get a good salary, it can be a good career. And they hate it. Those guys aren’t the kind of guys that I’ve needed because that’s not who they are. I need guys that tech and development is who they are.

I looked at myself back when I was writing code myself over the years, and went from COBOL to HTML to ASP to .NET and some Unix stuff mixed in. Whatever people asked of me, I just did it. That’s the same kind of guys I want.

My team that I already had in place when I made the decision to go with Xamarin for iOS/Android, I just jumped in and built it, and there was no real delay there because that’s just who they are.

I’ve had some of those APIs that we’ve talked about when we have some kind of strange device, niche device that has an API or an STK that had a problem. They’ve gone in and fixed the STKs for the vendor when we had a problem with it. They made things happen that would really hard to find a specialized skill.

One of the hand-helds we had for one of our municipal clients, just two or three days before going live we got a requirement from the client that they needed to be able to sign with their finger, to sign the hand-held and put their signature on there, and have every signature print on every citation.

That was fine except we were using Windows Phone and there was no driver for the printer that we needed to use to make that happen. But my guys still had it printing that signature in two or three days.

Ledge: That’s what you need, those problem solvers that are just going to…

Jim: That’s right. You need the problem solvers. I need the problem solvers. I need the guys that are –we joke about it – that are good at Google, and the guys that don’t pretend to know everything.

One thing that I always say in an interview is, if every single question I ask if I never get a ‘I don’t know’ then that’s a problem. Because there’s too much new technology. Nobody can know everything off the top of their head, but I want the guys that can figure it out very quickly.

Ledge: That’s great insight. Thanks Jim. It’s awesome having you on. Thanks for sharing your insights with the audience.

Jim: Yeah. Thanks.