Mobile Platforms - MoMo London September notes

After a break for August MoMo London was back again last Monday with an event on "Mobile platforms - Hobson's Choice or Too Much Choice".   The sponsors were EMCC and Yahoo! - thanks guys!  Here are my notes from the evening, never fails to amaze how much you get out of taking notes and writing them up - I felt a bit disengaged during the panel (maybe sat too far back this time) but lots of insights in the notes now I look back at them.

Opening Remarks - Dan Applequest

- lots of change happening around mobile platforms, question is where to put investment? There is potential for confusion but a lot of innovation occuring. 

Panel Introductions

Marco Balabonovic - Lastminute.com labs

- this year is "Year of the Mobile" (whoop whoop! - Ed) at lastminute.com labs,

- approach has been to "build a [mobile] website and pretend it will work on all phones and largely it has" (this was the most sensible comment of the night in my view, mobile people usually insist on making things too complicated - Ed)

- the main thing they built was a mobile restaurant booking service FoneFood that lets you find a restaurant and book a table via your mobile

- they only adapt the site for three different sizes, but are going to make a separate site for the iphone, an investment they are happy to make

NB - Marco said very little after this - maybe he was a bit bemused by the other panellists and audience getting tied up in the discussion around fragmentation etc which they have basically solved by not worrying about it!  The mobile web is the platform!

Riccardo Valera - Yahoo Mobile Europe

- Lead engineer for Yahoo Europe mobile integration work with operator partners etc for services like Yahoo One Search and Yahoo Go!

- mobile is a "complete nightmare" given all the different handsets, operator restrictions, different APIs, devices, you need a Blackberry version, RAZR version, large screen version etc - "it just doesn't scale"!

- so they are trying to create a framework for developers to plug into to save them all this work - Yahoo! BluePrint

Ben Last - EMCC

- EMCC is 10 years old, does a lot of hard engineering stuff for people like Slingbox

- some people a bit ambitious for mobile widgets, they are way up the app stack, not everything is going to translate from the PC world

Nick Allot - OMTP

- has run into the same problem again and again in mobile over the last 10 years, it is hard to do anything on mobile due to platform and device fragmentation

- the only platforms with mass penetration are things like SMS, but the things with a rich experience like Symbian embedded apps have a low penetration

- mobile web is fine for broad applications but for deep mobile apps you really need access to contacts, location etc but raises security issues

- sees a collision of two worlds - Mobile OS and mobile web - risk of nightmare - neglible market, terrible development environment

- mobile widgets already fragmenting - opera, Yahoo, Google, Nokia all have own widget platforms

- BONDI initiative aims to provide consistency in access of mobile web to native device capabilities whilst maintaining security

Simon Rockman - Sony Ericsson

- NB Simon was quite a contrary speaker - couldn't work out if I violently agreed or disagreed with him or whether he was bang on the money or talking bollocks but he did make some interesting points and a great panellist

- Simon spends his time looking at "what do people want their phones to do" (Ed - question is would they be able to tell you that?  Would anyone have said they wanted an SMS facility?)

- Apple appstore, Android appstore are great - a better route to market that operator portals

- there is a trend towards people doing more than making calls, look at the iphone and the appstore success

- games developers optimise for one platform (e.g. Nintendo DS games sell millions just targeting that one system) - Sony Ericsson sell millions of each handset so surely developers can target by device - don't need to cover everything (Ed- I kind of agree with this - just aim at the main handsets and forget the rest is a good approach for mobile)

Discussion

- the last comment really kicked off the discussion proper, not going to attribute quotes but just knock out the key points

- US TV ads are now focusing on the appstore showing screen by screen how to get an app on your phone - apple is really pushing user education here - people are going to think apple invented the idea of downloading an app to your phone!

- web 1.0 worked because it took simple content using the capabilities of the basic browser then built on this - same should happen for mobile (Ed - Yes!)

- Solving the fragmentation issue is key to operators getting some of the apple appstore pie - huge numbers of people download 1 mobile game (they can do it!) but never again because the game is crap!

- phone prices keep getting cheaper for the same functionality not necessary becoming more powerful - the next 2 bn subscribers will be using basic £10 phones

- connectivity still a massive bottleneck, batteries only getting 10% better each year so processing power is not going to get mobile out of a hole

- BONDI has 8 of the big GSM operators and 5 major OEMs signed up plus vendors but not always easy to get big companies to go in same direction for obvious reasons

- they are interested because fragmentation can only be stopped by agreement on standards (allows lots of players) OR a major player wins and imposes a standard e.g. NTT DoCoMo did this in Japan - standard workflow for doing stuff on phone

- N95 can have 8 different firmware releases all with different bugs!

- massive uplift in usage 2/3/ 250% when you let people access a mobile service with no PC registration barrier (Ed - when are facebook and myspace going to let you REGISTER on their mobile sites?)

- everyone says there are too many platforms then even more get created - its started to get unmanageable

- java is an effective standard in mobile at the moment - does work across the vast majority of phones

- end-to-end platforms work well e.g. Blackberry RIM - Iphone

- no-one ever made any money out of mobile operating systems

- mobile will only work when there are two or three main platforms like the web

There was a bit more discussion i think and a demo from Shazam but my hand was getting tired.  For me the only sensible approach at the moment if you are a startup is to build for the mobile web or use java/ symbian apps to target only the major handsets.  Trying to cover all bases just isn't practical (or necessary). 

Thanks to the organisers for arranging another lively event and to the sponsors - better nibbles than usual although the catering at Thompson Reuters a few months ago has yet to be beaten.

Andrew Grill twittered the event http://twitter.com/andrewgrill

Also other people have written this up too

http://mobilephonedevelopment.com/archives/675

http://natalian.org/archives/2008/09/16/3-reasons-why-the-web-platform-will-not-fragment-on-mobiles/