HOME · Twitter · Flickr · LinkedIn · publications · @ Ars Technica · Running IPv6 (Apress, 2005) · BGP (O'Reilly, 2002) · BGPexpert.com · presentations · iljitsch@muada.com

Apple Watch speculation

Posted 2014-10-07

Almost two years ago, I speculated about what an iWatch might be like. So let's see how well I did. And then speculate some more about the things we still don't know.

So that's 22 correct and 12 incorrect or missed predictions/speculations for a 69% score. Not too bad.

When watching the unveiling last month, it was obvious that the design has been in the works for a long time and was almost certainly finalized some time ago, while the software is still a work in progress, and it's hard to say what the status of the electronics is. Also, Apple isn't telling the whole story quite yet. Sure, sending someone your heartbeat is cute. But that's not a reason to include two LEDs and two photosensors on the back, which is obviously expensive in cost of hardware/manufacturing and power use.

Speaking of power, I'm really interested to learn more on what Apple has managed to accomplish here. It's amazing how much faster mobile CPUs have gotten the past five years—although some people are already speculating the growth is leveling off. That same technology can be used to make a much more modest CPU use very little power. People say that on smartphones the screen is the part that uses the most power, but that's not really true: you can watch 10 hours of video on an iPhone (screen on but doesn't use much CPU) while you can play games (that also use a lot of CPU/GPU power) for much less than that. But on the watch, the screen could easily be the biggest battery hog.

I'm thinking both screen and mobile CPU/GPU technology are a lot more mature than they were even a few years ago, so would it make sense to release newer, better versions of the watch every year? I'm thinking the watch would be more like the tv in this regard: the internals are updated at some point but marketing-wise, this is a non-event. Also, I expect Apple to sacrifice margins initially so they can use the best screens and batteries for the 1.0 models. Margins will improve as screens and batteries of that quality get cheaper.

In the past, 1.0 versions of new products often had obvious flaws, especially in hindsight. I'm hoping the technologies are mature enough that this won't be the case for the watch. Probably the biggest limitations of the watch 1.0 will be lack of cellular connectivity and GPS. It could be a very long time until those radios are efficient enough to be powered by the battery inside a watch. Or maybe only a few years.

For now, the watch will use an iPhone to connect to the internet and determine your location. Most of the time, that won't be a big deal as you're going to have your phone with you anyway, because you can actually read somewhat-longform text on it as well as type on it. (It looks like the Apple Watch doesn't support any form of text entry on the device itself.) However, the iPhone is relatively fragile, so there are times when you don't want to bring it with you. For instance, I took up inline skating a couple of months ago. I've had a few falls. In one case, I landed squarely on my phone, but fortunately it was in a heavy-duty case. I also ruined my earbuds on another fall. So leaving the iPhone at home and listening to podcasts on an watch with bluetooth headphones would be great when skating.

Ben Thompson thinks that a standalone watch will arrive within a few years, and Apple is making sure the ecosystem to support such a device will be ready when that happens by leveraging the iPhone to provide connectivity for the watch. In their podcast, he and James Allworth argue about whether that's an appropriate strategy or Apple should have chosen to either release an unconnected watch now, or wait until the watch could handle its own cellular connectivity.

In my opinion, that's a decision best done on a per-app basis, just like on the phone. Your iPhone doesn't turn into a brick when you put it into airplane mode, although obviously some applications don't do much without connectivity. Also, there's an intermediate step: Wi-Fi. I don't look forward to typing WPA2 passwords on a screen the size of two postage stamps, but once you've set up your home/work/school networks, you can be connected most of the day without needing an iPhone or a cellular contract. (Which makes me wonder how schools will react to kids/students wearing this type of smart watches.)

Of course Apple brilliantly managed our pricing expectations with "starts at 349 dollars". Before this, I thought at that $200 the Pebble watch was at the upper end of what I'd be willing to pay. But by the time we get to stand in line, $350 will be old news and we'll pay that without blinking. And then $80 for an extra band, or $700 - $1000 for the stainless steel model. Obviously the gold models will be much more expensive, but the question is: how much more expensive? Apparently Rolexes and the like can cost many tens of thousands of dollars. I'm thinking just like Apple is dragging up the low end, they'll probably drag down the high end. $3500 sounds like a nice, round number that still quite a few people can afford. Maybe they'll go a bit higher, but I doubt they'll break the $10k barrier.

Search for:
RSS feed (no photos) - RSS feed (photos only)
Home, archives: 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014