Impact of the Mobile Internet?
For some time, I have been pondering the question of how the "mobile internet" (a label I dislike, but by it I mean HSDPA aka Turbo 3G networks) will impact mobile phone usage in general, and specifically the company I work for?
Theoretically, the "mobile internet" has been around for years in the form of GPRS and "old" (384 kbps) 3G, but in reality things really kick off with the 1+ mbps achievable on HSDPA networks. Seems to me something similar to what happened when we went from 56kbps dial-up to broadband on the PC will happen, but what specifically?
I am in particular concerned about how fast the mobile internet will pick up and how it will affect the tradeoffs between mobile applications (native or Java ME) and web applications? We do both at OnlineCity - probably with an emphasis on mobile applications since slow networks and crappy browsers have historically rendered the mobile browser a pretty poor platform. But if mobile web application usage picks up - and what drives the uptake? - should we then shift our focus? Not only are web applications easier to develop and maintain - they are much easier to distribute. No application download or installation - both of which are user deterrents…
I ended up doing a presentation at work last week, addressing some of these questions and I would like to put forward the gist of the presentation in this post. Before I get to that, however, I should like to mention what a few other people have written on the subject.
By happy chance Michael Mace's post "Mobile Applications, RIP" from his "Mobile Opportunity" blog (highly recommended!) ticked into my feed reader a few days ago. As the title suggests, Mace proclaims mobile applications as dead - mainly because of distribution issues - and propose they will be overtaken by mobile web applications. Now, Mace's argument does not apply entirely to the situation of the company I work for - mainly because Danish carriers exert less control over customers and their handsets (Mace's post is quite understandably very US-centric); because we do vertical applications; and because we do not sell to end-users direct. Which is not to say there is not plenty of food for thought in the post - I particular like the statement concerning platform wars:
"A platform that is technically flawed but has a good business model will always beat a platform that is elegant but has a poor business model."
See also Mace's follow-up post.
Mace's suggestion for mobile developers to create "dumbed-down" mobile web versions of applications also seem pretty sound - especially considering the evidence. Off the top of my head I can think of a number of services that - more or less intentionally - pursue this strategy - like YouTube, Nokia MOSH, Yahoo!, or Google Gmail.
The post briefly mentions the quirks of mobile application development - like the problem of three million platforms with a hundred users each and the fragmentation of Java. If you like some further elaboration of these issues, go check out Israeli mobile game developer Ofir Leirner's blog - one of his posts discusses in great detail the challenge of porting mobile applications, while another post addresses the issues of distributing applications in the US. That last post is of course mainly specific to the US, but it is highly educational to read about the peculiarities (to foreigners) of the US market.
Right! Back to my own thoughts, which are somewhat specific to the Danish mobile market. In short I think the "mobile internet" is really going to take off in 2008 - it slowly started in the fall of last year, but I think a number of factors have now aligned to really drive adoption. There are four drivers:
1. The roll-out of HSDPA/Turbo 3G networks
- In Denmark there are 4 operators (US: carriers) expanding HSDPA coverage like crazy - two already cover more than 80% of the Danish population.
- The global growth rate for HSDPA networks in 2007 is estimated at 69%.
2. The availability of HSDPA-capable handsets
- The handsets announced for 2008 at GSMA WMC in Barcelona indicated (at least) two trends: the proliferation of GPS and the ubiquity of HSDPA.
- HSDPA handsets from Nokia: N96, N78, 6210 Navigator, 6220 classic.
- LG: LG-KT610, KF700
- Sony Ericsson: Xperia X1, Z770i, W980i, C702, C902
- Samsung: G810, F480, U900 (SOUL).
- Not an exhaustive list, but you get the idea…
3. The decline of data tariffs
- Competition from small MVNOs (BiBoB, Happiimobil) has driven tariffs from around €1.33 - €2 per MB down to ~€0.25 per MB.
- Introduction of flat rate data plans (more operators expected to follow suit) priced at €25 - €40 per month (10 GB fair use).
- The retail price of 1 GB data is approx. €1 in a fully built network.
- Compare flat rate monthly fee with retail data price.
4. The "iEffect"
The iPhone is probably not accountable for all of the following, but it is definitely a contributing factor…
- Apple's iPhone has made the "right" people realize mobile phones can be used for other things than speech and messaging - even the competition likes the iPhone
- The iPhone's large screen, Safari browser and UI have together pushed the envelope of what is considered feasible on a mobile phone.
- Together, the two above items have spawned a huge interest in mobile web applications: consumers are starting to become aware mobile phones have a browser; service providers and businesses exhibit a growing appetite for mobile presences; and developers have identified the mobile web as a fun and profitable platform. Expect to see sort of a gold rush to the mobile web in 2008…from everybody, advertising agencies, ISVs, you name it.
So what to make of all of this?
Right now, I think it will be most wise to defer any definitive decision between mobile applications and mobile web applications, and bet on both horses. In many cases, I think mobile applications continue to be the way forward for us. Partly because of the things mentioned above, but also because mobile applications provide access to the phone hardware like the camera, GPS-receiver, bluetooth interface, PIM data etc. Some of the stuff we are working on - or consider working on - hinges on using the hardware.
However, I think we should keep a very close eye at how the market for mobile web applications develops and I think we should ramp up our mobile web application development activities. We must present mobile web application as an option to our customers and make sure we make a compelling case by keeping track of traffic coming from mobile device browsers. Being aware of opportunities to do application/web application combos seem both wise and prudent.
If you enjoyed this post, make sure you subscribe to my RSS feed!