On lazy tech journalists
November 19th, 2009, 9:10am by JakeSpotted an article yesterday about AT&T’s overloaded 3G network, with the headline-grabbing claim that AT&T’s 3G traffic has increased 5,000% in 3 years. [Note that the included graphic says mobile traffic which would include 2G/EDGE as well.] There are so many nits to pick in this article. First, it would be a lot easier to understand if the article said that AT&T’s traffic has had a 50x increase in 3 years. Second, the journalist makes no attempt to understand what the baseline is. Since AT&T scarcely had a 3G network 3 years ago, of course their 3G traffic was tiny!
Worse, the journo uses this as a jumping off point to ask if Sprint or Verizon could have handled a 50x increase– which misses the point. I guarantee that Spint’s & Verizon’s 3G traffic was far higher than AT&T’s 3 years ago, and would not have had a 50x increase in traffic if they had got the iPhone. The article reaches the height of its stupidity with this rhetorical question, “Perhaps other carriers would have fared a bit better. Verizonís 3G network, even back in 2007, was much deeper and broader than AT&Tís. But could it really have supported a 5,000 percent increase in data traffic without incident?” It’s almost like the 2nd sentence didn’t read the first sentence.
November 19th, 2009 at 10:44 am
Another article from the same joker, who apparently doesn’t understand that Amazon & Wal*Mart routinely sell phones at substantial discounts to MSRP. The idiot reporter confuses these discounts as price cuts and blames the launch prices on reviews that were probably published after the pricing decisions were made.
Anyone with a decent understanding of the wireless market expected Amazon to sell the Pixi for $50 or less.
November 23rd, 2009 at 1:28 pm
Add this article to the pile of so-called news sites re-printing press releases from biased parties as if they were actually news. I have a really tough time believing that Palm Pre market share fell 50% in 1 month, and any journalist that accepts a number like that at face value should look for another line of work.
November 24th, 2009 at 6:45 pm
You’re putting words in their mouths. They never say anything about Palm dropping in market share. They are simply reporting what Admob reports as “blah percent of the smartphone traffic tracked by ad requests that use San Mateo-based AdMob.” No need to shoot the messenger here. I don’t think it’s too much to ask of readers to judge for themselves the validity of the data.
November 24th, 2009 at 7:43 pm
You see the headline? “Android, iPhone dominate US smartphone traffic.” Beyond that, this is presented as a news story, not a press release. I expect news to provide a certain amount of context, if not analysis.
November 24th, 2009 at 9:29 pm
Are you questioning the accuracy of the headline? It certainly would seem to be intuitively correct, and backed up by their data, right?
I agree with Jake that the way the data suggests that a loss of half the previous month’s smartphone requests for AdMob ads is the same as a loss of market share, which probably isn’t true. What is the statistic you’re looking for, though? What do you think would be a more true indicator of how relevant a particular smartphone is? And how different do you think that data would be from AdMob’s data?
November 25th, 2009 at 1:31 am
Admob overstates Android, iPhone and possibly Pre traffic because it includes ad units that are served up in applications on those platforms. So the data, while interesting, is only so useful in discerning trends. One case where it could be handy is in assessing devices on a platform, such as Droid vs. G1 or iPhone 3G vs. iPhone 3GS.
November 25th, 2009 at 4:03 am
Here are some other mobile web stat.
November 25th, 2009 at 7:02 am
Too bad, I used to have statcounter on pyslent long ago, would have been interesting to see graphs for The Board. We would have seen lots of PalmOS, then WinMob, and finally iPhone and webOS.
So, again, what questions are these stats trying to answer? Looking at total ads for a browser favors platforms with more AdMob apps. Total page hits shows something about mobile web use, but (a) Statcounter is slanted toward blogs, and (b) mobile web use is only one use of a smartphone. I download well over 1GB of data to my phone each month these days, but I don’t hit a lot of web pages on the phone. My mom has an Android phone, but probably has never launched the web browser (she likes getting photos by email, and using Google Maps).
I’d bet the average web use of an iPhone is less than the average web use of a Pre, since there are probably people who buy iPhones for the iPod function, or to play games. But there are a lot more iPhones out there, so the web surfing numbers are much higher for iPhone than Pre.
I feel like all these stats – ads, web hits, data transferred — are limited in some unique, but critical way. It sure would be helpful to have some hard handset numbers. In the end, is that what you’re trying to back out of these stats?
November 25th, 2009 at 4:51 pm
I’m on board if we are criticizing the data rather than the article! Admob’s survey is an indicator of how their business is going, nothing more. It might be useful in discerning trends, as Jake says, but even in this case, where the trend indicates that traffic via WebOS has been cut in half, there’s got to be something wrong, an artifact or something.
November 26th, 2009 at 4:13 am
Another great reminder why I stopped reading the Motley Fool. I’m honestly stunned to see people paid to write crap that wouldn’t withstand scrutiny on the Yahoo message boards.