Speaking Engagements & Private Workshops - Get Dean Bubley to present or chair your event

Need an experienced, provocative & influential telecoms keynote speaker, moderator/chair or workshop facilitator?
To discuss Dean Bubley's appearance at a specific event, contact information AT disruptive-analysis DOT com

Tuesday, May 15, 2007

Mobile Internet access is about PCs, not phones

I thought 2007 had finally killed the rosetinted-glasses-mobile-hippy view that PC-based Internet was going to wither away against an onslaught of mobile handset-based web access.

Yet it only takes one vaguely-worded press release with few definitions to reignite the hype again. Normally I'm a fan of the MobHappy guys, but on this occasion I think Russell's mixed up what he'd like to see happening in his version of utopia, versus what actually is.

Let's start with the press release:

"In fact, 5.7 million people in the UK use the mobile web, as opposed to 30 million who access the web by PC. This means that the mobile web is already nearly one fifth the size of the PC web"

Even if you take the numbers at face value, it's still a non-sequitur. How does the number of people accessing something determine its size?

Secondly, there is the perennial issue of definition. Is the "mobile web" anything accessed via a WAP or web browser on a handset? Does it include email as well? Are laptops with 3G data cards (or WiFi for that matter) classed as mobile web devices? I can quite believe that there are now 3-5m people in the UK who connect via WiFi or 3G from a PC. And how do you define & measure "unique visitors" from mobile devices - it's much trickier than for PC-based access because of poor cookie support, differing IP addresses and so on.

Then there's the totally unreasonable assertion that:

"Accessing the web via the mobile, as any sane, thinking person must acknowledge is going to dwarf PC access. And indeed, I would argue that it’ll replace it altogether within the next 5 years"

This is wrong on so many levels it's amazing.

  • Developing countries show no signs of "skipping the PC" and going to mobile phones as an Internet platform. A combination of data-unfriendly prepaid tariffs, non-optimised handsets and operator focus on voice capacity when building their networks. Plus whenever I visit a country like Mozambique or Bolivia, the Internet cafes seem to be full of kids on MSN, the schools are getting PCs, and there's initiative like one-laptop-per-child. Not to mention that every government wants its children to be PC-literate so they can work in new businesses, learn to develop software & thus benefit the economy
  • BRIC countries (Brazil, Russia, India, China) will skip PCs. Yeah of course. That'll be why China has getting on for 100m broadband lines then.
  • There isn't a fraction of the spectrum needed to get 2bn+ people using the Internet via mobile at a decent speed, especially if they want to compete with ADSL2+ or FTTH
  • Most of the current 3G services are being held afloat by PC users with 3G data cards. This Ericsson investor presentation last week highlights the fact that Mobile PC connections will generate more traffic than anything apart from voice. And even if you assume as Ericsson does that there will be more mobile broadband subscriptions than fixed, you need to take into account that usually one fixed-sub is >1 users, while the reverse is true for mobile as a growing % of people have multiple devices & subscriptions.
  • The notion that people will use a "personal digital device" (ie a phone) that somehow docks with keyboard & screen to become a pseudo-PC, with all data stored in the network doesn't fly. First off the price curve on memory & disk space >> bandwidth, so it makes more sense to store data at the edge of the network on the device. Secondly, nobody in their right mind will want a personal "hub" which is optimised-for/locked-to a single operator rather than being guaranteed to be fully open to the Internet at large. Plus there's the huge problems of federating data between the 2,3 or 4 separate devices that most people will have.

I could go on, but I'm sure you get the point. I love using the web & email when I'm mobile (well, OK, given the apparently lousy state of T-Mobile's DNS servers, it could be a lot better). But there's no way I'd use anything other than a PC for "heavy lifting" work on the real web. Sure, mobile access to the Internet will improve, and there will be a few people who start to rely on it more or even completely. But it's a complete fallacy to suggest that people won't aspire to use a proper PC and operator-free raw Internet access.

7 comments:

raddedas said...

Hear hear!

Anonymous said...

"and how do you define & measure 'unique visitors' from mobile devices "

For telco's, the MSISDN can be logged against each request if the correct infrastructure is available... which makes it more reliable than web-based cookies... Dur! ;o)

Anonymous said...

"Is the "mobile web" anything accessed via a WAP or web browser on a handset? Does it include email as well? Are laptops with 3G data cards (or WiFi for that matter) classed as mobile web devices?"

Can't speak for all Telcos, but my personal experience, working in this field, is that direct email access doesn't go via the same route as general HTTP 'web' access, and therefore those users wouldn't be included.

Laptops with 3G data cards; yup, they'd be counted, although device information is passed to the logs too, and laptop users generally resolve to a Windows-type device, and so could be excluded from published figures... if desired (they are in the minority anyway... as in *small* minority).

The ability to produce 'real' figures is there, we just need a body such as the W3C, ABCe or whoever to take the opportunity to standardise reporting methods for mobile internet, and work with the correct people at the Telcos.

Dean Bubley said...

Fair point about MSISDNs but that's only useful if operators pass the logs onto every website owner that requests them, and even then it would be a nightmare to compile. Let's say I want to know who accesses www.disruptive-analysis.com via mobile devices - I'm hardly in a position to ask each of 200 carriers round the world about unique visitors.

The other "unknown" here is whether some of these stats, which to me don't "feel right" are based on log files / similar tools, or by surveying people about whether they use the "mobile web". If it's the latter, and given that there is no "mobile Internet" just "the Internet", I'm sure you'd get people using WAP, operator portals and content delivery platforms, 3G & WiFi laptops, Blackberries and maybe even fixed PCs in Internet cafes misinterpreting the question.

I could certainly believe that 5m+ people regularly use some form of Internet connectivity on a wireless device in the UK if you include all of those. But I don't buy it as a measure of handset-based access: I simply don't see that many normal people using the web "in the wild".

Separately, there's also a minor issue of double-counting where mobile users have something 1.3 devices on average, often with separate operators. Conversely I'd expect each PCs to have 1.5-2 users.

Marc said...

I enjoy reading both Russell's and your blogs. Great debates on both.

I think it's dangerous to write-off the vision that mobile internet could reach significant critical mass. I wouldn't be surprised if, ten years from now, I found myself using my portable device to 'look-up' information or 'retrieve' digital assets dozens, if not hundreds of times a day.


"In fact, 5.7 million people in the UK use the mobile web, as opposed to 30 million who access the web by PC. This means that the mobile web is already nearly one fifth the size of the PC web"

I don't want to put words into Russell's mouth, but he suggests that 20% of mobile users are having an experience of some sort. That's significant and the numbers can only go one way. The real question should be, 'can we see exponential growth yet?'

"Accessing the web via the mobile, as any sane, thinking person must acknowledge is going to dwarf PC access. And indeed, I would argue that it’ll replace it altogether within the next 5 years"


I don't know whether anyone truly predicts the death of the PC. Every home needs a workhorse and a media station. We will eventually be having internet experiences on a number of different devices. PC and mobile will be two of those devices.


Have fun,
marc

alex said...

> Every home needs a workhorse and
> a media station.

I see in long-term the PC "dropping its P", i.e. changing from a personal computer on every one's desk (Bill Gates' vision from the 70's) to a home "server" for local storage, processing power, and broadband Internet access.
Mobiles are already more personal and will become more convenient for several (not all) "computing" tasks. Mass-market usage is not there yet and requires massively better phone-PC integration - which will happen eventually as PC and phone vendors learn to do better "converged" business.

Do homes in the future really need 2 or more desktop PCs with large screens?
Wouldn't I be fine to check my RSS feeds with my mobile while my wife uses the PC for making a home video?
(This doesn't imply "one device for everything" - multiple mobile devices per user would also do it.)

Anyway I agree with Dean's conclusion that most Internet access will continue to be through stationary computing decices.

Alex

Vikas Kanungo said...

Good to know your views on mobile phones vs. PCs for mobile computing usage. Would be happy to know how would you compare public servcies on mobiles and m-democracy with PC based users.

Vikas