How two generations of internal indoor map apps created a new line of business at Sony
- Smartphone and application maturity in 2009
- SEMCMap
- SEMCMap and Indoor Finder coexist for a while
- SonyMap
- Research and Incubation
- Nimway
- References
In 2009, I was assigned to one of the Android teams at Sony Ericsson. To get up to speed I experimented with the different Android APIs. At this time I had been working at Sony Ericsson for 3 years already, and the massive footprint of the company in Lund required office landscapes the size of small towns. But also meeting rooms. Lots of meeting rooms.
This is the story of how SEMCMap (SEMC being short for Sony Ericsson Mobile Communications) was created, how SonyMap spawned from a completely different project, how they were merged, how we managed to be the first project to be accepted by the brand-new Research and Incubation initiative at Sony Mobile, how it was the first project to "exit" Research and Incubation a few months before I left Sony Mobile for Spiideo, and finally how it became a completely new line of business1 for Sony Europe.
I need to give Magnus Persson a big Thank You for helping me remember details that would have otherwise been lost to time. You were invaluable. Thank You!
Smartphone and application maturity in 2009
In 2009, there were very few smartphones: The iPhone 3Gs had just been released, and Sony Ericsson was about to release the Xperia X102. There weren't many apps, and there definitely weren't any popular indoor-navigation apps.
Smartphone screens were tiny compared to the screens we have today, and the phones weren't exactly snappy, even if it maybe felt like they were then and there. The Xperia X102 had a 4-inch screen and the iPhone 3Gs had a 3.5-inch screen. Today it's hard to find a smartphone with a screen that's smaller than 6 inches. Apps weren't very pretty, especially on Android, which honestly had a lot of catching up to do at the time.
SEMCMap
The first indoor map application I worked on was SEMCMap. I created it while I was looking into the Android APIs. My first attempts were terrible: performance-wise, design-wise, you name it. No compass, no auto-rotate, and no indoor positioning.
It was rough being a user of SEMCMap, but it quickly gained in popularity, since it filled a niche that people needed help with — finding their way to the meeting room they were supposed to be in. If you've ever worked at a large company, or in a building with a lot of meeting rooms, you probably have some sense of how disorienting it can be when every floor looks the same and there are hundreds of meeting rooms.
The main use-case was basically just typing a meeting room name into the search box, selecting the meeting room and then trying to decipher where you were going based on where you thought you were and where the map told you the meeting room was.
A lot of people were using SEMCMap at the time
SEMCMap had one redeeming feature though — the ability to send your position to someone else through the app. You could use your Google account to quickly sign in, and then use those credentials to send a message to another signed-in user along with the location. The feature was popular and well-used.
So far so good. I maintained the application myself for a few years, and a lot of people seemed to like it! Mostly I did other work: phone book, social media integrations, and so on. I worked through a lot of tasks and a lot of teams.
SEMCMap and Indoor Finder coexist for a while
Magnus Persson3 and I know each other from my early days at Sony Ericsson. He was working in some of the same teams that I did. A cool cat, and full of energy and great ideas. I think I met him the first time in 2006, I was 24 at the time feels like yesterday but almost 20 years have passed. Life comes at you fast.
While I had been working on SEMCMap on-and-off for a couple of years, Magnus was working in the Technology department. The Technology department was tasked with keeping track of new technologies and making sure those ended up in Sony Mobile's devices.
Around 2011, Magnus ran a project called Indoor Finder together with Ericsson Research, with Sony Mobile mainly contributing to the map interface and the positioning functionality. At the time, he wrote a blog post, which can still be found using The Internet Archive's wayback machine! I did a little archiving myself — have a look at the PDF.
Left is SEMCMap, right is Indoor Finder
At the end of 2011 Sony Ericsson was being transformed into Sony Mobile, after Sony bought out Ericsson to gain full ownership and control of the company.
SonyMap
Ericsson Research lost interest in the joint research project Indoor Finder at some point, and Sony Mobile fully took over the project. Magnus and I both had a lot of ideas and Magnus talked me into joining him to work full-time on SonyMap. Roger Jansson4, was my manager's manager at the time, and he was happy to give the permission to go ahead, and we went ahead... full steam.
We both agreed that Indoor Finder was a better starting point than SEMCMap, but a lot of the code and ideas from the internals of SEMCMap made it over to the new SonyMap application. We brought on more people, and we were usually at least two, but usually three, or even four people working on it at this time. One of the good people got help from was Oskar Syrén5 who implemented an improved fused indoor positioning engine that used a combination of GPS, Bluetooth Low Energy beacons, and Dead Reckoning.
Bluetooth Low-Energy Beacon
Magnus dutifully ordered parts and assembled Franken-BTLE beacons and placed them in every single meeting room — and there were a lot! Every BTLE beacon was attached to a standard Sony Mobile USB charger. You would think people would just leave them alone, but you'd be wrong.
Research and Incubation
Around 2014 things were not going great for Sony Mobile. Actually, things had not been going great for a while. Samsung and Apple are currently the biggest smartphone manufacturers, and that was true in 2014 too. There was more competition for sure: LG, Motorola, even Nokia, but they were all small too.
Patrik Olsson6 kicked off the Research and Incubation (R&I) department. Anyone that had an idea could apply, and SonyMap was the first project to get accepted. Magnus and I were very happy about that, since it showed that we had worked on something that was useful to people.
Getting accepted to R&I was great for us, but it also serves as an example of what happens when a project gets the spotlight on it in an organization starved for success. All of a sudden I wasn't spending my days planning concrete things to do with our application, or building cool features, I was in meetings with people wanting a piece of the cake.
On the other hand, we now got to staff our development team with amazing engineers and designers. It was a fun time, and the team built a lot of interesting stuff, most of which was actually usable and shippable. One of the engineers that joined at this time was Mattias Niklewski, who I consider to be a good friend, and one of the finest engineers I have worked with.
We weren't just the first project to get accepted to the R&I process. The goal for an R&I project was to "graduate" a real product, and we did that too, as the first graduate project from R&I no less.
Nimway
But, I had done my time at Sony Mobile. I was ready for a new adventure. So after a little courting I accepted an offer to join Spiideo. Magnus tried to get me to stay on, but my mind was set on moving on.
Magnus stayed for another 8 years. Mattias, and Oskar are still there, and the team did a great job acquiring clients under the Nimway1 moniker. A few years after I left, the project moved on from Sony Mobile and joined Sony Europe, and is still going strong from what I can gather!
It's satisfying to see how something with small beginnings was able to evolve and amalgamate, finally becoming a whole new business segment for a huge company like Sony. I'm proud to be one of the parents of SonyMap, now part of Nimway, and I'm glad to see that what started as our side project continues to employ people.