The WDN Symbian DevZone Looks at FastMobile's fastxt Instant Messaging

by Richard Bloor, January 20, 2003

FastMobile Inc. is a private company with headquarters in Chicago, IL USA. The company has ambitious plans to provide users with a range of text, voice and picture based products built around Instant Messaging functionality. This week we are discussing their business strategy and choice of Symbian OS with James Tagg, European Office M.D, and John O’Boyle, Marketing Director. - Web Version


SMS has been a huge hit and is being used as a form of defacto instant messaging by many users. It is perhaps not surprising therefore that a number of companies are building IM products for the new generation of smartphones. One such company is FastMobile who recently launched their fastxt service using the Nokia 7650 in the UK.



WDN: Why did FastMobile perceive a need for a mobile IM product given that SMS already provides users with very similar capabilities?

John: We see Mobile Instant Messaging as being a natural progression for the large number of people in the UK and Europe who are, to put it bluntly, text addicts. It also extends the PC owners use of Instant Messaging into their mobile. As such we believe that Instant Messaging is going to be of significant benefit to people on their mobile in the same way as it has been on their PC. So we have developed what we believe to be the best system, one that can be used across all 4 networks in the UK, with plans to expand internationally.

WDN: However there are already clients for existing web IM services starting to emerge for mobile phones so why do you believe there is a place for a new service?

John: Firstly we have it working and working very well. Secondly, while there are a number of Internet companies providing these services we are fundamentally different because we are not only an Instant Messaging company. fastxt is part of a family of products we will be launching under the ‘fast’ brand that will appeal to smartphone users including a picture-based product, a chat based product and others. We are expecting to launch additional services in the first half of this year.

WDN: You have chosen the Nokia 7650 as your initial launch phone, what were the commercial or technical reasons for using that phone?

John: We chose the Nokia 7650 because it was the first major commercial camera and MMS phone. In addition Nokia has dominance in the UK and Europe, so it was the obvious one to go for because of the sales volumes. Very much the right phone at the right time, it was extensively advertised in the last quarter of 2002 by both Nokia and the Operators so it made sense to target a phone which was getting maximum exposure. We will also be moving to phones like the Nokia 3650 and Sony Ericsson P800. If its GPRS and Symbian we will look to launch for it. We are also using Java to cover other non-Symbian phones such as the Motorola T720.

WDN: So is it fair to say that you chose the phone rather than the operating system?

John: Not entirely. We look for technology that will be on handsets in the markets that are of commercial interest to us and likely to have good market penetration. Symbian OS and Java are technologies that fulfil these criteria.

WDN: In developing the Symbian based application have you used any of the programs offered by Symbian or its partners?

John: We are members of the Symbian Platinum program and have been working well with them for some time now. It is a very good relationship. Most of the support they provide is in the technical area but we also attend their conferences and meet regularly with them to discuss marketing opportunities.

WDN: What about the Symbian OS licensees, Nokia for example, given you launched on the Nokia 7650?

John: We certainly have had involvement with Nokia. From my perspective their strength is they treat everyone in an evenhanded way. I don’t believe there are special deals for developers, product companies or operators and I say that with 8 years experience with mobile operators. They have certainly given us everything we have asked for, to the extent that they would, I believe, provide the same information or assistance to anyone. From a marketing perspective the fact that they will invest in marketing support is great but again no more or less for us than anyone else.

WDN: Your strategy appears to be to sell fastxt through individual retailers in parallel with their phone sales. Many developers seem to feel that getting a direct relationship with operators is a key sales tool. Why have you not taken this approach?

John: I would like to emphasize that this is not a strategy that excludes operators. There are two good reasons for the approach we have taken. Firstly we want to build FastMobile as a brand. Secondly we want the products and services we are offering to be available across all the 4 mobile networks here in the UK and an early tie-up with an Operator may have been a barrier to that. Also in part it is because there are hundreds of other companies knocking on the operator’s door. So we are targeting quality retailers and selling through our own on-line shop, we are supporting this with targeted advertising in publications such as T3, BoysToys and style magazines such as GQ and Esquire. If we prove that we have a strong product, a sustainable business model and proven customers then that makes us a force to be reckoned with. The other reality is operators tend to work on a large scale and prefer their own brands. We expect that our products will make some noise in the market and as they do they will come to be of interest to operators. Certainly we hope to get involved with the operators, ultimately we won't rule out discussing our products with anyone who is interested and fits with our business plan.

WDN: We are already seeing companies working on Voice IM, given these emerging technologies how much life do you see in text based IM?

John: If you look at SMS it is not an advanced technology by any means yet for most operators it is a very, if not the, most significant revenue stream. Mobile IM improves dramatically on SMS and fits well with what customers have shown that they want. In the studies we have done text based IM has some advantages over voice based products, the fact that it is silent, you can not be overheard and there is a significant portion of the market to whom that is important. I suspect that it is corporate users are more likely to be more interested in voice, though we have had several requests from corporates about our fastxt service.

WDN: Given that you have got a Java implementation of your IM client and Symbian can run Java why not simply use a Java Client? James: There are actually several reasons. Java MIDP is heavily sandboxed, so for example in building an instant messaging application it would be advantageous to use the contacts list that is already on someone’s phone. In Symbian OS there is an API we can use to read, manipulate and write details to the contacts list. In Java, because of the security model, you can not look at other data on the phone so you are not able to access the existing phone features, which can hamper usability. The second thing is that Java on a mobile is a bit like the old DOS computers, you start a program do something, quit and move onto the next program as opposed to Windows were the program hangs around forever until you want to use it. So in Java the application will terminate when a phone call comes in and when you hang up you have to restart the application. To be honest most people forget meaning that the default for IM on a Java phone is off-line with the user periodically connects to see if any of their buddies are on-line. Now obviously if everyone in the community is in this mode, mainly off-line, there is not going to be a community to talk to and you don’t have an Instant Messaging product. So Java only really works with pre-existing communities of users. Now it turns out that IM only works for pre-existing communities of members anyway, because no one is going to create a community from scratch. So Java has a lot of promise for people with a Java smartphone who want to check their instant messaging but it not as rich an experience as can be provided on Symbian OS. There are a number of other reasons the usability of the Series 60 and other Symbian OS UI’s, that Java currently blocks you on all the really clever things you could want to do. There are plans for all of that to change with Java, but the process could take 2 to 3 years to work through standardization and get built into phones. Finally there is the issue that TCP/IP and UDP are not supported on many Java phones, just HTTP. Our experience is that customers don’t like the HTTP implementation, when we show them it they are not very receptive, it’s the difference between a 10 second response and 0.5 second response when we use TCP/IP. After all it is instant messaging! That’s why the first Java phone we will launch on is the Motorola T720 because it’s one of the first to support TCP/IP.

WDN: So how have you addressed the issue of using existing communities? James: By interoperating with MSN Messenger and SMS, so we are not building a whole community from scratch, although it is practical to use our product as a stand alone Instant Messenger. We have to piggyback on the network effect created by existing communities.

WDN: Perhaps I can move on to ask a little about the architecture, are you running the whole service from the phone or using a intermediate server? James: We have our own server. There are actually a number of ways we could have implemented the service, but using a server has several advantages. Say we want to interoperate with AOL, now AOL change their protocols regularly, so if we had 10,000 client applications out there on phones it will take forever to get everyone upgraded and working again. However having our own server means we simply change the protocols once on the server and we are back in service. The other significant advantage is that it keeps the client small, as it does not have to implement multiple protocols. Our client is about 61K, whereas I have seen clients where no server is employed getting to upwards of 800K and that makes it very hard to do over the air downloads. It does mean that you have to maintain a server but the costs are fairly small per customer when you get to scale. Another good reason for having our own server is to help with future products and services, such as our forthcoming Voice Messaging product.

WDN: You mentioned AOL just now, but not as a service you interconnect with currently, presumably your server approach means you can turn AOL on easily in the future? James: Yes, we certainly can. The choice at the moment is driven by the fact that in the UK MSN pretty much has the market, so after you have implemented MSN you are really scratching your head about whether it is worth doing anything else. Of course in America it will be AOL and when we are in 2 or 3 international markets we will probably be supporting just about everything.

WDN: There are a number of emerging standards for IM and presence, have you used these in the applications? James: The big problem at the moment is that there are, I think, 4 standards bodies all working on IM and presence and the standards are not finalized. So we have used a hybrid combining our own proprietary protocols and the developing standards for communication between the phones and our server. But again as we use our own server we can easily work with any of these standards when they are finalized.

WDN: Why did you choose Symbian OS over the other smartphone operating systems? James: Well it’s a good operating system. We actually implemented the application on a Microsoft PDA initially. The primary reason to go to Symbian and the Nokia 7650 was market share. If you look at the adoption of new technology 85% of early adopters are Nokia phone users. So launching on any other phone would have meant chasing a fraction of 15%.

WDN: Was it the same team that did the Symbian OS development? James: Half and half, we hired 4 ex-Symbian people and created a team with 4 of our existing developers.

WDN: For the developers who were not familiar with Symbian OS how did they find the transition? James: We ported the whole thing to Symbian OS and launched it to the market in 2 and an half months, obviously the ex-Symbian people helped greatly, but within about 2 to 3 weeks you probably would not have known the difference between them.

About the WDN Symbian Editor, Richard Bloor:
Richard Bloor has 16 years experience in the IT industry. His earlier work was largely in design and development of commercial and manufacturing systems but more recently has focused on development and test management of government systems.

Richard Bloor is the Mobile Applications champion at System Architecture consultancy Equinox of Wellington, New Zealand.

Richard can be reached at bloor@wirelessdevnet.com.

Symbian DevZone Home

Nokia DevZone Home