Saturday 5 March 2011

Thrutu - In-Call and awesome - but not new...

Having been in the mobile industry for many years, it's interesting to see great new products come to market - things that can genuinely change the way we communicate with each other. One such product is Thrutu.
Thrutu uses the ability of the UMTS network to provide simultaneous voice and data session to provide an enriched telephony experience.

What do I mean by enriched? Well, imagine that you can see the location of the other party in the call? Or instead of reaching for a pen to jot down a phone number, this is simply shared from one phone to another.

This is Thrutu and it's available in Beta now for Android 2.1 handsets.

One of the reasons I am so enthusiastic about Thrutu is that I jointly invented and implemented an identical system in 2003/2004 when I was CEO of Bullant Software.

So why has Thrutu launched this and not Bullant? read on...

What was Bullant?

Bullant was a company based in North Sydney, Australia. We sold a product called the Bullant Remote - in essence a JavaScript VM for Symbian and MIDP handsets coupled with networked back end. The VM exposed most handset features - telephony included - to app developers. It made it possible to deploy applications on-demand to any handset equipped with the client. It was used in a variety of telco projects in Australia, Singapore, Japan and the UK. Bullant lives on as xumii.

The Call-Share Idea

In late 2003 or early 2004 I got an enquiry through our website from the Chief Trials Manager at Vodafone UK, a guy called Mike Howells. Mike and his team were looking at IMS applications and ideas around sharing content between handsets using a generic client.

After a face-to-face meeting in early 2004, I proposed a solution that would work on the existing Bullant client that leveraged the features of the soon to be launched UMTS network (but not SIP/IMS).

The product was called 'Call-Share'.

Vodafone Call-Share

Call-Share used the Bullant Remote for Series 60 handsets. The client started and connected to the network and then became dormant until a call was detected. If the call was on the 3G network, the client initiated a data-connection and passed the B-Party call details to a server. The server then attempted to determine whether there was a 'call rendezvous' between two handsets with the client.

On detection of a rendezvous, pCode was sent to each handset that was immediately executed and caused a UI to popup over the system Telephone application. This UI facilitated the sharing of a range of content types:

  • My business card (with a nod to Palm OS).
  • Any contact from my Address Book.
  • A map showing my location.
  • A picture (either captured live from the camera or selected from the handset Image Gallery.
  • A multi-player game (Connect4 actually).
The app also worked on 2G. We had no 3G network when building and so, perforce, one of our developers simply created the data session as soon as the GPRS network was available on call-hangup. I'm not sure if Thrutu do this, but it turned out to be quite a neat trick.

Building the App

The app was designed by myself. The JavaScript components were built by Felix Castanar-Perez and Tom Hall. Brendan Babb and Matt Powell made sure all of the Symbian Client telephony interfaces were in-place and working properly. I built the Java based network back end to detect calls between handsets. The initial prototype was the result of eight weeks development.

Our biggest headache was detecting the handset MSISDN. We did not have an SMS gateway so we had to use the MCC/MNC from the IMSI to arrive at the canonical international form of the number in order to match calls. Thrutu seem to have the same issue which they solve by sending an SMS at install time.

Our second problem was location and mapping in the dark ages before GPS and Google Maps.

For location, Vodafone provided us with the UK Cell ID database. Since this was not exact, we allowed a user to navigate a cross-hair over a map of their locality. Once selected, the map popped up on the B-party handset UI. For maps, we did a selective trawl from Map24 in the UK. We got just enough map-tiles to cover the Newbury environs for demo purposes. Mapping worked surprisingly well...

The final issue was that we built this before the 3G network was launched in the UK. Luckily, we had Hutch in Australia to test with - although we never really got simultaneous voice and data working properly before we flew back to the UK.

Testing without 3G

I flew to the UK to present the finished app to the brass at Vodafone. I think this was about May 2005. At this stage, Vodafone had a barely functional 3G network. By a happy coincidence, they did have excellent coverage in Liverpool where I worked out some last minute niggles.

The app worked brilliantly on two shiny new Nokia 6630s provided by Vodafone. I flew to London (trains were not working) and presented the app at Symbian HQ and then to Newbury. I think they were impressed.

The app was subsequently demoed in Spain and, I think, Germany. Sadly, we could not get an OpCo to run with any further funding and so we shelved the app while we delivered Optus MyZooNow and SingTel's IdeasLive.

We subsequently hired Mike Howells from Vodafone UK to become our CTO in Australia. In a dog-and-pony show in October 2005 we took the Call-Share app to Helsinki to show Nokia, Stockholm to show the Ericsson IMS team and to various UK operators as well as one in Hong Kong and one in Thailand.

[Edit]What was Nokia's response? That it was very difficult to do anything with the Series 60 Telephone app.

In 2006 we took the app to Cingular in Atlanta and, rather embarrassingly, Verizon (more on that in a moment).

Of all the operators I demoed to, only Three UK expressed a strong-interest in the app. However, I don't know where that account got to.

Realistically, this is an off-deck app - we should never have tried to sell this directly to an operator.

Enriching Communication Sessions Anyone?

Mike and I took the IP we had created and he quickly wrote up a patent application entitled 'A SYSTEM FOR CONDUCTING MULTI-MEDIA COMMUNICATION SESSIONS'.

This was quite a departure from the original app and emphasised SIP and XMPP. We filed a claims with WIPO and PCT claim 12/187,378.

Share the Mobile Future?

Bullant raised a significant amount of money from CM Capital and Southern Cross Ventures in order to commercialise Call-Share and other apps in our portfolio. We merged with a US company and formed uiActive, whose "Share the Mobile Future" tag line was inspired by Call Share.

You can see our quote from the uiActive website via the Internet Wayback machine:

Enriched 3G communication sessions Intuitive user interface makes it easy to share applications and data (images, appointments, contacts, games, ring tones and locations) while talking.

So What Happened to the App?

Actually, I do not know for certain.

I stepped down as CEO in June 2006 when uiActive was formed and then left the company in January 2007. uiActive was subsequently renamed xumii.

At a guess, xumii did not pursue the patent application. (I assume this because after assigning the rights I was never contacted again in relation to this, one assumes the application was allowed to lapse).

xumii as a company took on and expanded greatly on ideas of mobile social networking - something they are doing very successfully to this day.

[Edit] Actually, on reflection, I suspect that the 2008 release of the iPhone 3G, SDK and App Store was the reason they dropped the app.

[Edit] The patent claim was filed in Australia in February 2006. A check of the patent database shows that the claim was allowed to lapse.

[Edit] I incorrectly stated that the US patent had been allowed to lapse. This is not the case. Apparently, it is still being argued, though the examiner has, thus far, rejected all of the claims due to prior art.

Do I use any of these Ideas?

I designed the PACE client in early 2008. It makes full use of simultaneous voice and data in order to minimise the battery overhead of moving campaigns to the handset. It also uses overlay windows drawn on the Telephony app to provide a range of value added telephony services such as roaming and voicemail support.

What was that about Verizon?

Note to self: It's best not to demo an app that relies on the simultaneous voice and data capabilities of UMTS to an operator who uses CDMA - which does not support simulaneous voice and data...

[Edit] I have been reminded that it was not actually me who attempted the Verizon demo. Fair cop - I used some artistic license - it was actually one of my colleagues.

I did make the same mistake with AIS in Thailand, but they are not quite the same as Verizon.

In the end...

The Bullant client was great because it was generic. What actually happened when content was shared was that an application was dynamically shared from the network and this app dynamically created the UI to view and interact with the shared content.

So, to engage in a shared game of Connect4, one user suggested the sharing and when the second user accepted the request, both handsets received the app from the network which then created the UI within the existing application window.

This is apropos of nothing, however, because the app was never launched to a mass market. Thrutu have done this. It may not be new, but it is brilliant. IMHO, it is the killer telephony app on 3G. I wish them all the best for this and I hope they get it pre-installed on every Android handset. I also wish the team good luck with iPhone - frankly, I will be amazed if they are able to put this into the Telephone screen although it is an obvious and valuable enhancement to FaceTime.

[Edit] I just got asked why I didn't build this myself. The answer is I could not due to my severance agreement with uiActive.

[Edit] I just got asked why try and sell to an Operator. There were no app stores in 2007 - Operators were our best chance at hitting the mass market. As I said, we did take this to Nokia, but did not get a bite...

[Edit] Why no screenshots?
I'll see what I can dig up from the information that was public.