Suruk gets more love

We didn’t stop with a Hello Android! for Suruk. We wanted the shiny new Android version to be as cool and useful as its award-winning JavaME sister. So, the Android version now has the route-finder as well as community feedback for drivers. With route finder, one can find the approximate fare for the trip in advance as well as check the route on the map. You can rate the driver after your ride as helpful, honest, or rude depending on your experience. That’ll help the community to identify which auto or taxi to trust for their rides. You can benefit from others’ feedback in the same way. We’re planning to use the aggregate data creatively so as to reward the highest-ranked drivers as well as identify the top offenders. Do try the updated Suruk and let us know how you liked it, please.

Where do our Nokia downloads come from?

After launching Suruk on Android, we had a look into our download stats on Ovi to understand the geographic spread of our Suruk users and compare against that of SOS. Suruk Distribution Following is the distribution of SOS (includes India downloads on Aircel PocketApps): SOS Distribution* colour intensity corresponds to the number of downloads, grey regions do not have downloads We noticed a couple of interesting things. While there are no downloads for our free app from China, the country is among the top downloaders of our paid app. Perhaps it’s among the ‘unresolved’ countries in the former list. Also, Germany and Switzerland top both downloads and revenue for SOS, but not as much for Suruk. Is this because of SOS is also available in Deutsch? Or is it because they’re relatively likely to buy paid apps than other developed nations? Please share with us in the comments, if you know why. Maps courtesy Gunn Map (political boundaries are indicative, not accurate). During our chat with Albert Ching, he suggested doing a density map of routes searched for in Suruk. Does anybody know of a tool that helps plot routes? Update: While digging up our route finder request logs, we found that there are indeed requests from China. So, it appears that, for some reason, the Nokia Ovi store counts the downloads under ‘unresolved’ country. Either that or they’re not counted.

Wanna go Kopa?

Our motto is to help people travel cheaper, safer, together, and better informed. The together piece was missing for a while. Not any more. The latest cookie from our bakery, Kopa, is a ride-sharing app for people going to events. Isn’t it fun to have interesting companions to ride together to and from events? Just go get Kopa! Anenth and Sandeep built the alpha version of Kopa during Yahoo! Open Hack Day. Thanks to Amarinder of Mobility Ninja, we launched the beta for the NASSCOM Product Conclave attendees. Today, we launch an improved version of Kopa for Droidcon and Wiki Conference attendees. We’ll keep adding interesting events to Kopa, so stay tuned. Also, a hat tip to Indus and Rasmus for the inspiration behind Kopa.  Now coming to the obligatory naming story. So, what’s this word Kopa? It’s Latvian for ‘together’. Is that an ideophone? Not sure. But, there’s a hack to make it one. We make this word acquire the idea of sharing rides.  Try saying the following aloud, please. Wanna kopa with me?Shall we do a kopa on the way back?Kopas in rickshaws make rides fun.By the way, Kopa can be a contrived contraction of co–passengers as well. 

Autofare Reloaded

As early stage users of various products ourselves, we always value and look up to companies which engage in continuous customer engagement and work on suggestions and improvements suggested by users. The same appreciation makes listening to our customers and working on their feedback an integral part of our product development. Autofare, which started of as a quick hack, an extension of existing feature of Route Finder in Suruk, quickly became a favourite on the txtWeb platform. It also helped us realise how having an SMS extension can help expand the reach of a valuable service to huge set of users. It also went on to win App2Fame 1st prize and fetched us many laurels. Recently txtWeb team collected a lot of feedback from users and passed on a few valuable suggestions for @autofare. When we revisited the product, we realised there were many things we could improve and we worked on all of them and are extremely pleased to launch Autofare 2.0 today. We hope the experience for the end user will improve with the new changes and hope to keep continuously improving the product. We will appreciate any feedback you might have on the product. Enjoy a hassle free ride by being better informed!

APIfy It

We started developing the very first version of Suruk just after a small discussion in a coffee shop way back in May ’09 (That does seem like ages ago). Anenth and myself were just starting to learn J2ME and PHP, we completed the product within a month. We do often laugh over the fact that a Java class in client app ran over 2000+ LOC and server side code well over 1000+. But, what was really motivating was the fact that it worked, and people loved it. Suddenly we were like this bunch of guys in college who had real customer issues to tackle. In the midst of our assignments, we would set aside a couple of nights for dealing with “real” customer feedback. We’ve come a long way since May ’09, we iterated tons of times over Suruk codebase and it’s quite stable on all platforms today. With every version, iteration and product, the learning we all acquired was really valuable. Since all our apps were related to travel and commute, having  a consolidated data model was perhaps a natural progression. We started writing reusable components and modules which were used across products, and this made life much easier. The iteration time reduced immensely and we were launching newer stuff in rapid speed. Hence Wander (we code-named it so, as it was dealing with objects in travel and commute) was born.. and Cosec joins us in IIM-A! We started evaluating various platforms like Sinatra, Zend, Django, Flask, wsloader and so on. Being Python lovers we decided to stick to Python platforms and ruled out likes of RoR and PHP. Keeping in mind our requirement of APIifying modules, we finally nailed it down to wsloader. A big shout goes out to Tahir Hashmi, developer of wsloader, who was more than happy to guide us through some initial queries we had and we were good to go. Cosec also did some performance testing comparing wsloader with other frameworks and we were really happy with the findings. In simple words, wsloader allows exposing native Python interfaces as WebServices (REST+JSON) without having to write any HTTP/Web service specific code. In some ways it’s a parallel to, but makes it insanely easy to make any python module a webservice. It’s a WSGI module which runs on top of Apache with error handling integrated with Apache error logs. Consider an example python module as given below:class greeter:def say_hello():return “Hello Ideophone!” The say_hello() method can be accessed via http://{hostname}/services/greeter/say_hello Super simple, no? To get started, simply visit Do keep in mind that wsloader is a minimalistic framework and does not provide in-depth support for session and user management as the likes of Django would give. More on Django (our latest crush) will come soon to a blogpost near you. Till then, APIfy your code 

The story of Droid Juice!

Last year while researching network coverage during train travel for Pyka, I did a small hack to log signal strength for a given location. I used it during my trip to Chennai and data was logged using a background process in my Android phone for every kilometer. Incidentally, my onward and return journeys were on different routes. When we visualized using Google Fusion Charts, it was quite telling. There were many dead spots that would be impossible to detect otherwise. During this exercise, my droid had to be fed with constant power to keep it running.  We build products for travel and recently did several interviews with frequent travelers. We repeatedly heard people saying they usually turn off data just to conserve battery life. They were also concerned about running out of juice and becoming unreachable. This sparked an idea to build a public battery profile for people displaying the last known battery status. So, if someone is not reachable, one can find out if it is because of running out of juice. During last week’s Droidcon India Hack Night, I teamed up with Sandeep and Santhosh to build the basic version. Thanks to Muthuraj for suggesting that we fall back to SMS when the battery is really low, which we’ll add in a future release. In the past few days, the hack got spit and polish from Team Ideophone and it’s now available for download from the Play Store.  The heart of the app is a shareable battery profile. However, as pointed out in this article, when more and more people start using it, our data will show which Android phones have better battery life in real-life scenarios. While price and feature comparisons of Android devices abound, it’s high time we had battery performance halls of fame and shame! The high point was to know that Sundar‘s Ace had a really low juice retention factor and he wasn’t complaining about anything all these days.

Develop Your Idea Into an Android App in 24 Hours

Over the last 4 years, we’ve developed many Android Apps, which have been downloaded over 200,000 times. In that time, we’ve realized that our best work is done in quick, short bursts, with 80% of features built within the first 24 hours. Many startup founders following the lean startup methodology have told us that this is something that they want – to get to a functional app (minimum viable product) in almost no time, instead of floundering to build a full product over months. So today we’re announcing our new Android app development service. We will only take on short, well-scoped projects at the idea & sketch stage. We will turn your idea into an app in 24-72 hours (we can mutually agree on project size first). You will get access to full source code and assets once we’re done. Since we have both developers & UX designers in-house, you’ll get something more than a functional replica of your idea. The project size restrictions will also help you focus and scope out a truly minimal product. We’ve decided to be very transparent & simple with the pricing as well. We’ll charge $250 (Rs. 15,000) per calendar day, and won’t take on projects for longer than 3 days. If we’re unable to finish your product in the time agreed, you get 100% of your money back, simple.