I attended Barcamp Bangalore 3 today, held at IIM Bangalore. I reached early and saw the place getting organized in a ad-hoc manner. Initially there was chaos, and gradually order emerged. It was very interesting to see how the whole thing got organized with just a handful of volunteers. The event was well attended, with I think an attendance of well over 300 people. Kudos to the organizers!
There were a number of interesting sessions arranged around the broad themes of Mobiles, Society, Internet and Demo/Training. I felt that the mobile room by far attracted the largest crowds. Randy Wang kicked off the society room with a talk about Digital Study Hall. Thereafter, I spent most of my time in the mobile room. some of those that I attended and liked:
- Impact of Camera phones by Wwigo people — The key takeaway was that mobiles with cameras are becoming ubiquitous and there are interesting applications going to become prevalent. Some pointers are in the area of (1) scanning barcodes and finding more information about products, (2) Citizenship Journalism and many others. The form factor of the UI is a severe limitation. Privacy concerns abound and also to whom does the rights of a picture belong (subject or shooter). They showed a demo of their product Wwigo which lets you use your camera phone as a web cam with your PC.
- Activ Mobs — This was by far the most exciting demo I attended. The idea is Yahoo! groups ported to mobiles. They let users create small groups (which they call mobs) using mobile phones and let people message all their friends at once with a single SMS. The utility is obvious and the product is already a hit with the target crowd (18-25) with more than 10k users and 25k SMSs sent every day. They are still trying to figure out their revenue story and find investors, but the concept is cool, and can spread like a virus. I tried it out and they have some kinks due to overloading, but hopefully the service will only get better with time.
They talked about their learning in the last 4 months that the service is up. For one, they add Activ-Tip when the message is less than 100 chars, and have found that people actually read them. That could help them monetize it by replacing the Activ Tip by an (context-sensitive) advertisement. Secondly, it is difficult to enforce a format/grammar on the users since it is difficult to remember commands and syntax. Thirdly, which is a really interesting concept — instead of providing a web-interface which is completely different from their mobile interface, they are building a command-builder UI which lets the users easy build commands on the screen and submit online (with suitalbe help/directions). This will not only help in usability, but also help the users to remember the commands.
They also described their stack – Linux, Kennel, MySQL, Ruby, and a web server.
Good luck to these guys (Akshat, Sidu and Vidit)!
- mChek – mChek is a mobile payment solution which is currently live in Delhi. Airtel customers can pay their bills through their credit card using mChek as a gateway. It is good to see such gateways being set up since that is going to spur mobile commerce and more services being sold through mobiles. They use SMS + USSD (Unstructured Supplementary Data) to register and make payments. They claim that the USSD provides them an advantage over the other players such as Pay Mate and is also more secure (since USSD is not saved in sent items). They are certified by VISA. However, there were some issues raised about security in the light of the PIN and credit card number being sent in clear text.
- Zook – This is a mobile search solution on both SMS and GPRS. They have tried to keep extremely structured data in their databases which the query against the users question and display results, unlike Google/Yahoo!/MSN mobile search which essentiallly try to mimic their usage of unstructured data as on the web. They currently only focus on a few categories (ringtones/wallpapers/flights/restaurants/events etc.) due to the restriction of having structured data. Another choice they have made is to prompt the user with more users in case they can not find exact solution. For instance, if you such for Pizza, they might come back to ask you if you are looking for Pizza in Bangalore? and even names of localities such as Koramangla. They feel that this interaction differentiates them from other providers. They also have the option of the users contributing to their knowledge base, but I would not buy this poing until they can demonstrate its efficacy with large number of users.
- Socio-Net – This was about Social Networks becoming pervasive in future. Social Networks will evolve to become intelligent with personalization, intelligent minig of information, and closer integration with other applications, unlock collaboration and become drivers of many of our current applications. I feel that social networks have been in existence long before (LDAP, IM, Amazon, mailing lists) they became branded so, and what we are going to see is the defintion of Social-Networking applied to it. The reason I call Amazon a social network is because they are implicitly adding ‘friend‘-edges between people who bought the same book and doing collaborative filtering on it. There is need for more research in the area with many interesting applications possible (Mechanical Turk, Community Customer Support). There was a (rather) long discussion about entreprises having social networks so that employees spend time on their own social networks rather than external sites like Orkut. While I am all for leveraging social networks in an enterprise setting, I would not subscribe to providing a company social network like infosys.orkut.com on which employees can spend time. That is legalizing time pass :-) There can be better models.
Overall, some great ideas and it was great fun!