Wireless Utopia

This event of MoMo was definitely one of the more interesting ones. Our speaker was Naveen R from Tonse Telecom and the topic, WiFi Market in India, generated some interesting comments. Nobody it seems was convinced of the potential of the WiFi in India. The numbers (around $1B) were unimpressive and the growth nothing compared to mobile phones. Unfortunately the mobile aspect completely overshadowed the discussion. IMHO there's too much focus on mobiles in India. But thats a separate post.

Demo's were the real show stealer this time around. In MoMo Bangalore's own characteristic way, it was chaotic. The audience gave a hard time to Akshat and Rajan. Akshat demoed the Active Mobs product which is a group messaging solution built on SMS. Rajan showed us Motvik's WWIGO (Wireless Webcam Wherever I Go) product which turns your camera phone into a live webcam. Understandably both of them were a bit upset about rough treatment.

But I kind of liked the way it all turned out in the end. Real world users are pretty much like this. Real users hate to learn a new things. Real users tend to oppose change. Real users have to be marketed to, won over, wooed. If anything, it was a preview of things to come and unbiased first hand opinion of their targeted customer base. This was as close they can get to their real world users. And to be fair, quite a few people were impressed.

The one common aspect among them is that their focus is to first build the community first. These are free services which do not generate any revenue right now. In fact the SMS volume of Inactiv costs them Rs. 200 a day. Its the mantra of Web 2.0 being applied to Mobile VAS; community first, monetization later. And both of them are looking at advertising as a possible revenue stream. Which does not again have many fans in India. But monetization is really an afterthought here. Which I think is the right way to go with these kind of services.

What is needed right now is strategization. They have to pick their next step very carefully and figure out a way to move forward. As Sramana points out, they need experienced hands to guide them. Hopefully they'll find some. Entrepreneurship needs courage. Kudos to them for doing it. Here's wishing them good luck and success. We really need more such examples.

We have announced the Feb Event of MoMo Bangalore. We ran into a bit of trouble when our scheduled speaker backed out due to personal commitments. After frantically looking for another one we found Naveen R of Tonse Telecom. Tonse Telecom is one of the few Market Research firms specializing in India. They recently released a report on WiFi Market in India. Content Sutra had covered it earlier this month. Naveen's talk will cover this report.

We had kept Demo's as a backup if we do not find a replacement speaker. But we have had an enthusiastic response. So we are thinking of regularizing this in our events. An open door policy for anyone who wants to demo his product/service. It only has to related to Mobility and/or Wireless.

This time around its Motivk and Inactiv. Rajan has recently launched his Alpha product called WWIGO. He is going to Demo it this event. Inactiv is the other company doing the Demo. I ran into them in the Bangalore Barcamp and they have a unique mobile messaging solution. Next time it'll be Sloka Telecom and Mango Technologies.

So register yourself and come down and enjoy the show.

Tags: Mobile Monday Bangalore, MoMo, Bangalore, India.

We have just launched our product Blog. Here is the URL.

http://www.thinfone.com/blog/

The idea is to document the process which leads to the Pilot Program launch. It's going to be a very exciting next few months for us. Join us in this journey. :)

And if you dont know what ThinFone is, dont worry. Once we are ready we'll tell you all about it. For starters, check out the website for an intro.

Tags: ThinFone, Startup, Entrepreneurship, India

A quick post to point to this great program. If you have a device which you want to supported in Linux, the Linux Kernel developers will do it for free for you. And push it into the mainline. It means a great deal in the embedded world where to get the devices working you need a working driver from the manufacturer. Having the kernel developers do it is a much better way of getting it done though. The only issue is the NDA and to cover that there is a program with the FSF.

Linux is a very powerful alternative to all the embedded OS's these days. It already supports a large variety of interfaces and devices. And nowadays most of the manufacturers by default provide support for Linux for their products. There are companies which support these Linux Distributions. They generally provide a basic customized kernel but for the full thing you have to pay. For an early stage venture this is not an option. Therefore you end up writing the device drivers yourself or integrating from some other source. That, although is fun, takes a lot of time and effort.

If this program succeeds, it will become a trivial task to port linux on any embedded solution.

Via: Slashdot.

Tags: Linux Kernel, Device Drivers, Embedded, Open Source.

Tremor is a fixed point version of the Ogg Vorbis decoder meant for embedded systems which do not have a Floating Point Unit. Its an excellent piece of software developed by the Xiph foundation. Unfortunately its not maintained properly. It took me some time to get it all together. So here is a guide for those who are interested.

The trunk version does not work for me; autogen fails. But the low mem branch wroks fine. Download it straight from the svn here.

This version has a big memory hole. The patch for that is available from the mailing list. Download it here.

There is a performance improvement patch which uses IMDCT by Sandvall. That patch can be found here (thesis is here). Many thanks to Frantisek Dufka for sharing it. I had to make some changes in the makefile to get this working properly.

There is another patch which increases the performance by 15 %. I havent tried it out but its available here. Dunno if its already there in the low mem branch.

Thats all that is needed to make it work. Also check out the Rockbox's version of Tremor. They seem to have made a lot of changes, which again I havent tried out.

The one thing that is missing from the Ogg Vorbis specification is redundancy. The primary focus there seems to be encoding/decoding from/to files. I have been using it for streaming and packet losses lead to some nasty breaks. I am entertaining thoughts of adding that capability to Vorbis. Sounds like a good challenge! I am hard pressed for time right now but. :(