Paul Golding's Weblog on Wireless  
The future’s bright, the future’s ubiquity
 
             

Visit www.paulgolding.info















Subscribe to "Paul Golding's Weblog on Wireless" in Radio UserLand.

Click to see the XML version of this web page.

Click here to send an email to the editor of this weblog.

Blogroll Me!

Join Paul's mail list

 

Thursday, June 03, 2004
 

:: Phew! That's a lot of trivia ::

In the MDA forecasters’ event, one of the analysts mentioned, with all seriousness, a category of mobile service called "Trivia", which accounted for 18% of the non-messaging mobile data market. Strange that someone would use such a label. I could see what they meant; horoscopes, gossip and all that, well, trivial stuff. Oddly enough, it did not include text messaging, as this is messaging revenue. In that category, I did not see a segment called Messaging Trivia, even though many studies (e.g. Katz et al) have shown that huge quantities of messages are demonstrably trivia.

I remember sitting recently with an RFID "guru", interviewing him for my new book. One of the problems he cited was the incredible volumes of data that RFID scanning could potentially produce, were it used for small inventory items. This is a difficult problem to cope with and in some cases renders an RFID project untenable.

To back this up, at the MDA meeting, an emerging-technology expert from Deloittes produced a figure from Walmart that the expected daily data flow from RFID tags will be 2.6 Petabytes.

Naturally, none of us has heard of a Petabyte, but it sounds like a big number. It is actually 2 to the 50th power (1,125,899,906,842,624) bytes. A Petabyte is equal to 1,024 terabytes, which itself is still a big number. The library of congress contains seven Petabytes of information, which, to use the presenter’s words "puts it into perspective".

I know what he meant. However, Petabytes of inventory data about sugar bags and potatoes hardly compares with the loftier knowledge housed by the library. One is tempted to confer the inventory data to the trivia bin, although that is apparently now a respectable category for information.

 


3:22:47 PM      

:: "This is the year of mobile data (again)" ::

Someone kindly invited me to attend the Mobile Data Association's annual gathering of mobile market forecasting analysts. It was instructive.

The consensus was that things are at last beginning to move up the growth curve for mobile data. If this is true, it is promising.

 

However, no one was quite sure why things had begun to move, which was not that comforting. There was a lot of talk about re-branding of portals, and one analyst used the word renaissance to describe the current phase of their evolution. All the analysts mentioned Vodafone Live (VL) repeatedly. Vodafone are reporting over 4 million users for their service, although a user is apparently anyone with a VL capable phone, not a real user in the normal sense of the word. The MDA themselves are reporting large increase in the number of “WAP page impressions”.

 

Analysts questioned the widely held assumption that high usage of texting would lead automatically to an uptake of picture messaging and games downloads. So far, this is not true.

 

It is clear that no one really knows what’s going to happen in this market. It seemed that there was a feeling of forecasting a mobile future instead of innovating one. The signposts have become more important than where they are pointing.

 

No need to forecast what we already know

 

The most instructive thing was when one of the analysts talked about his experience of using CDMA1x in the US versus using GPRS in the UK for laptop connectivity: essentially, a case of usable versus unusable. The surprising thing was how he announced this as if no one had ever realised how unusable GPRS actually is in the real world.

 

Has anyone thought to test the sensitivity of WAP page impressions to network speed? Ah, yes! I remember Jakob Nielsen doing this several years ago and announcing that is was a problem. At the time, the “WAP industry” denounced his findings as bunk.

 

Now, that would be a useful forecast. Number of page impressions versus network speed (or latency, more usefully). Forecasters love to use “buzznyms” like KPI (Key Performance Indicators), yet I didn’t see any analysis of Key Performance “Affectors”.

 

Defeated by low tech problems

 

There were two high points for me. The first was when the most slickly presented analyst couldn't open the window to get some fresh air, no matter how hard he tried to pull the lower sash upwards. After nearly busting a blood vessel, he gave up. Someone who looked like an engineer casually stood up and pulled down the upper sash effortlessly. Lateral thinking! I struggled hard not to laugh too loudly.

 

The second high point was when the presenters' laptop battery died just when a "future trends" guy from Deloittes started talking about fuel cells as battery replacements. A fortuitous moment perhaps, as he was able to say how much longer the laptop would have kept going with a fuel cell. However, I felt that plugging it in to the one of the abundant mains sockets would also have done the trick.


1:15:58 AM      


Click here to visit the Radio UserLand website. © Copyright 2004 Paul Golding.
Last update: 7/7/2004; 5:22:12 PM.

June 2004
Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30      
May   Jul

Sign up to my email list

Click here to send an email to the editor of this weblog. - Send me email

Next Generation Wireless Applications

Visit my personal website to find out more about me.

Consult my booklist on Amazon

Blogroll: