Antennagate
First off, can we stop appending “-gate” to the end of any and every major or minor “scandal” – real or manufactured – in the US (and how many non-manufactured scandals actually use the “-gate” suffix anyway… seems like something to research someday)? How many people even remember at this point that it was named “Watergate” after the damned hotel of the same name? “Scandalgate” just seems lazy, uninformed, and in search of sensationalism. But I guess that’s American “journalism” in a nutshell these days.
Wow, anyway, moving on before I drift any further off topic…
There’s been more than enough said by the tech press about the iPhone 4′s antenna issues, but I’ll go ahead and pile on with what I hope is a more level-headed commentary than what the rest of the internet has been flinging around this past couple of weeks.
Apple’s not had a good time of things in the press the last few weeks, and the bumbling public response to the outcry by their PR department hasn’t exactly helped. I was interested to hear that Jobs was in Hawaii on vacation last week, which is when the Apple PR-signed “it’s totally a software problem that has nothing to do with the phone even though every phone has this problem” press release weirdness happened. I’m wondering if Steve has taken over the public management of this issue, including calling the press conference, as a result of that, and whether he was involved in crafting the PR response that had everyone shaking their heads.
To a certain extent, while it’s demonstrable that iPhone 4 has a reproducible signal attenuation issue when held in the lower left corner, I think the impact of the problem has been considerably overblown by the media, and the tech press/blogs in particular. It’s something of a complex problem because in areas of good signal, it’s not even an issue (even though the problem still exists), and the antenna design by itself seems to result in getting better signal in more places. Locations with already-poor signal quality seem more likely to be affected by this, and the iPhone’s wonky bar reporting algorithm made it more likely to make you think you were outside of a problem location. Given AT&T’s spotty network quality track record, it’s possible that like previous iPhones, this may disproportionately impact people in places like NY and SF, while users in other markets have no problems at all.
I think this is backed up by four bits of data that are floating around now. First, Consumer Reports’ signal strength testing does show signal loss on a consistent basis. At the same time, however, the increased rate of dropped calls compared to the 3GS and its more traditional (and oft-stated “better” by tech press) cell antenna design is marginal at best: less than 1 additional dropped call per 100 calls made, according to Apple and AT&T. Still, it is marginally higher, and if Steve is to be believed, it’s not something Apple is willing to consider acceptable. Apple’s rate of return is also 1/3rd of what it was for the 3GS, coming in at just 1.7% (though there’s no breakout for why the device was being returned to provide more granular numbers on returns based on signal issues). Finally, Apple’s own customer complaint data indicates that just over one half of one percent (0.55%) of all calls they’ve received concerning iPhone 4 are related to the antenna. Put together, these data seem to indicate that while there is signal attenuation, it has little to no impact on the actual performance of the device on the network. I think that a lot of the perceived enormity of “antennagate” is due almost entirely to the breathless coverage by the online tech press, which has used its massive echo chamber to artificially magnify the severity of the problem.
The echo chamber’s already at it again with the dropped call delta, with people using percentage magic to claim that iPhone 4, assuming a low dropped call rate for the 3GS, is dropping up to 100% more calls, because 1/100 dropped to 1.9/100 dropped is a 100% increase, you know. If only the number were 0/100, then we could have an infinitely larger dropped call rate on iPhone4! The anti-Apple folks are spinning the number Jobs cited as either a massive increase in the dropped call rate assuming a low 3GS dropped call rate, or a condemnation of Apple’s phone line as a whole assuming a high 3GS dropped call rate. Where’s this “less than one additional dropped call per 100 means .9 additional dropped calls per 100″ logic coming from, by the way? Why can’t <1 additional dropped call mean .4 additional dropped calls?
In reality land, however, an increase from 1/100 dropped calls to 1.9/100 dropped calls means you’re a whopping 9% more likely to experience a dropped call on average, not 100% (or even 50%). Still not a good number, but miles away from the backwards percentage assumptions being thrown around in the Engadget comments. Bottom line: if you rarely experience dropped calls, you’re still not likely to ever experience one. If you experience dropped calls a lot, you probably won’t notice the minor increase. I’m also inclined to think that areas with high dropped call rates like NY (which is frequently quoted as having upwards of a 30% dropped call rate on earlier iPhone models) and SF are negatively impacting AT&T’s national dropped call rates, and outside of those areas the chances of a dropped call are dramatically lower than the national rate would imply.
As far as additional data go, I’d be interested in seeing the results of a Consumer Reports test of other cell phones when held in similar positions, to see whether there is any attenuation in them as well, and how severe it is. That would shed some light on the veracity of Apple’s claims and testing showing that other phones experience this issue as well. I’d also be interested in seeing dropped call deltas for non-US carriers to see if this is an AT&T-specific issue (as most network problems with the iPhone in the US seem to be).
Apple’s near-term fixes are largely band-aids to mitigate the problem, but based on Steve’s comments during his intro and during the Q&A, it doesn’t sound like they’re done researching ways to resolve the problem long-term, and hardware fixes within the phone itself may be pending down the line if they can figure something out. I think Apple has done the right thing by offering a free bumper with the phone, and while it won’t impact the hardware itself, the modified signal indicator algorithm seems likely to provide a more accurate indication of the impact of the attenuation on your network connection. In most cases it will probably result in a drop of a bar or two at higher signal levels where there wasn’t any drop before, but reduce the drop in areas where the signal was already low (2-3 bars). The indicator also reaches to a lower dB level now, so the signal indicator will stay at one bar in weaker signal areas than before. Whether this lower db level will be sufficient for making or retaining a call, I’m not sure. I’d be lying if I said I didn’t think AT&T had anything to do with the “wrong” signal algorithm that was used prior to iOS 4.0.1, given their “more bars in more places” marketing campaign, but that’s the conspiracy theorist in me taking.
If I had $199 burning a hole in my pocket, could afford the cost of the monthly plan, and didn’t already have a 3G iPad to do most of my mobile stuff on, I would get an iPhone 4. In my estimation, the improvements to the hardware outweigh the apparently limited impacts of the signal attenuation problems in daily use. And for full disclosure, I hold my phone in my left hand.