A look MetaWatch, Google I/O and Potential Sport Uses

imageA little over a week ago at the Google IO conference a company called Fossil not so quietly introduced the MetaWatch.  The watch allows one to wirelessly connect to their cell phone and stream glanceable information, such as who’s calling or new e-mail/text messages.  The idea being to take a normal looking watch (one of the two model actually looks formal dining room classy!) and combine it with the functionality that triggers most folks to pull a phone briefly out of their pocket, such as seeing who’s calling during a meeting.  Applying that to the sports world, a company like RunKeeper could integrate their app such that the phone remains in your jersey pocket or armband, but the information you care about (speed/pace, distance, heart rate via BT HRM strap) is displayed wirelessly on the watch in real time.

But before I get to the details of the watch, there’s one thing that’s of particular note: The goal of Fossil and the MetaWatch isn’t actually to sell directly to you, the end users.  Nope, it wants to re-invent how digital watches are designed ground up from a technology standpoint.  So how are you gonna get one then?  Well…hang on and I’ll explain.

See, today when a company comes up with an idea to create a watch for some purpose they have to design everything end to end themselves.  This means everything from chip selection to firmware that runs the device to the software that you use to then have that device interact with a computer (like downloading workouts).  This costs a lot, with one of the major costs being production of the physical watch itself (manufacturing).  Typically this runs about $1M for the first year for a startup to get a complex digital watch off the ground.  Think of companies like Sportsense, developing the Swimsense watch for example.  When you look at that device, there’s a lot of ground that they had to re-invent.  Some was unique, but some might have been re-usable.  And their costs might have been considerably lower initially.

That’s where the MetaWatch folks want to interject themselves.  They’re looking to offer startups the ability to do prototyping and limited production runs of watches at a fixed and cheap cost – $200 per device.  The key selling point to startups is that they’ve created a device that includes a highly visible LCD screen on the front of it that can be customized to do anything you’d like.  And since it’s written on a fully open source code base, it allows a startup company to customize it how they see fit.  The only component of the watch that isn’t open source is the Bluetooth stack.  This means that from a development standpoint there’s nothing stopping a small company with a cool sports idea.  Well, except $200 and some quality dev time with pizza boxes and Mountain Dew.  Comparing this to previously, where they’d have to invest heavily in initial prototypes and manufacturing costs.

So how does this all integrate into the sports world?  Well, one of the guys behind the MetaWatch is Bill Geiser, who happens to have a long history with sports watches, in particular one of the very first watches designed for swimming, the Speedo Stroke Monitor, back in the late 90’s.  This watch worked by a simple transistor switch that recognized when the water was in the water…or the air based on electrical current.  Thus each stroke could be counted simply due to the circuit being completed when the hand hit the water.  But that was 1998.

Now fast forward nearly 15 years to the MetaWatch.  This watch includes five key features that are of interest to folks in the sport device world: A fully customizable 96x96p LCD screen, an accelerometer, waterproofing to 30 meters (3ATM), up to a week of battery life, and a vibrating motor for vibration alerts.  Add to that it includes the TI 2560 Bluetooth chip for syncing data between cell phones and other devices (laptops/tablets/etc..).  Between all that stuff you’ve now got the majority of the parts needed to create your own sports watch.

The device does have a few areas which could be improved upon though should it want to be leveraged for sports.  For example, it doesn’t have any method for audible alerts (beeping).  And for those in the running/cycling world, ANT+ capability is huge, and it doesn’t have that either.  In talking with Bill Geiser about this, I asked what the path would be should a company want to integrate ANT+ into the device.  He noted that the current TI-2560 BT (Bluetooth)chip in these watches has an identical footprint as the TI-2567 chip, which include both BT and ANT+.  It should be noted that the current BT chip does not support Bluetooth Lightweight (BTLE) and only supports basic rate Bluetooth.  That’s alright…as there’s basically no BTLE devices out there yet in the sports world.  Even Polar has backed away from using it in the near term, instead focusing on their W.I.N.D. protocol for upcoming devices like the Polar/Look KEO pedal based power meter.

So what’s the path ahead?  Well, over the next month the device will start shipping to people (developers mostly) who’ve ordered it.  From there those folks will need to start developing apps for it, either to control the watch from a cell phone, or to just use the watch natively.  From a phone standpoint this will for the immediate future be limited to folks using Android phones, and will require that the Openwatch App be installed on your phone – as this is the conduit between the phone and the watch.  And as noted earlier, this would then allow sports applications like MapMyRide/RunKeeper/etc… to stream the data you care about directly from the phone to your watch display – just as if you had a normal Garmin/Polar/Suunto/Timex sports watch.  And from a development standpoint, if a developer can program for the Android platform, they can program for and fully control the MetaWatch.

I asked about iPhone compatibility and they said that the major blocker today is around time limitations imposed by iOS for background tasks.  Additionally, the iOS platform prohibits calls to certain functions like inbound telephony information and text messages (without jailbreaking).  I also asked whether or not adding a GPS chip into the device was on the table.  They noted that while it has been considered, it would require an interested company to work with Fossil to create a new watch that supports it – a bit of a ‘start from scratch’ type of situation.

In all, I think the MetaWatch will present an interesting landscape for sports technology startups to quickly prototype and get beta and first run products in front of test groups.  I don’t see it changing the game or development cycle for major and established players like Garmin/Polar/Timex/Suunto, but I do see it as an opportunity for smaller or no-name companies to quickly break on the scene.  I also see it as a very intriguing opportunity for classic phone app developers to dabble in the device world without any more investment than a $200 test kit.  And, of course down the road perhaps we’d get to a full color model not unlike a simple Apple Nano strapped to a watch band that allowed you to run whatever apps you wanted.  Think of a Garmin FR310XT…but with the RunKeeper app running on it.  Of course, today this is just Fossil’s first generation device…but I think the future is bright. For now though, baby steps.

(P.S. – Yes, this may be a bit geeky for some…but I know y’all like the mix of sports technology and race reports, so fear not – tomorrow will be back to deep triathlon thoughts.)

Retweet 10Like 6Google +1 0

10 Comments

  1. nice overview! you probably hear about the new USB accessories that google released for Android link to goo.gl and the possibility it can bring for Android gadget devices.

    Reply
  2. When you mention that they “should” use a chip with ANT+ and/or GPS, my first thought is why?

    Would it not be “better” if the watch is stupid, and the ANT+ and GPS functionality is built into th phone it’s connected to.

    That would make it a lot cheaper I hope.

    And if you build in GPS and ANT+ you have a new Garmin watch, and not a easy accessibel “wrist display” for the mobile phone which you could then have in your saddlebag.

    Or am I missing some important point here?

    Reply
  3. ANT+ build in the watch would be nice, since you would not have to carry your phone on the run/ride.

    For HR/Power/Cadence the watch would “talk” directly to your HR-strap, SRM-crank or foodpot.

    I would love to see such a watch from Apple/nike+. Even if the watch would just display phone calls and messages (at first) an would have a stunning “unibody” apple design. I would buy it right now.

    Thanks Ray for the great post!

    Reply
  4. But if you want a watch with ANT+ so that you can avoid carrying your phone on your ride, then you can buy a Garmin NNNN or Timex or Adidas or Nike or (insert random ANT+ GPS enable watch name)

    Nothing new there. I think that the point for me with a watch like this would be a “small and easy accessible” display for my phone which does the data recording and powers all the GPS ANT+ chips. Hopefully with a baterylife to match, the indicated 1 week I feel is ok, but no longer than what you need.

    Hoping for more interesting discussions around this device. :)

    Reply
  5. Alex

    It seems a lot like the Sony Ericsson LiveView: link to sonyericsson.com

    Just, a bit better with the waterproofing.
    I do not think it is neccesary to include a GPS-chip in the watch because you most likly already have GPS in you phone. The biggest problem I see with this is that you can’t really swim with your phone.

    Reply
  6. great post, ray.

    alex, the watch can operate on a stand-alone basis (w/o connection to phone); so it is possible to swim with it. The watch has both an accelerometer & an ambient light sensor. Coupled with the right embedded software either of these sensors could be utilized to detect a stroke cycle.

    Reply
  7. Seems like a pretty cool idea, i’d get it just for ability to glance at my watch and know who is calling, or who sent a text, etc. Any devise that is going to require me to carry my phone while running or riding is not really something I’m going to bother with; what’s the point when a Garmin 310xt can do all those activities without requiring that i have an iPhone/android strapped to my arm?

    Reply
  8. wow, this is an awesome post Ray !!!

    I was looking for this on the TI store, but they only mention the BT support (TI-2560 chip). Is a metawatch coming that supports both BT and ANT+?
    That’d be perfect since you could interface with both your ANT+ sensors (powertap, HR, etc..) and your phone. Best of both worlds !!
    Do you know if that model (with TI-2567 chip) is coming?
    With that we’d be able to develop an amazing watch with all information on one watch. I’m ready to start coding for that !!!!

    Reply
  9. Ignoring the sports aspects it’s an interesting concept. Technically it’s very interesting. It takes the idea of having real time information that’s available from your phone and puts it on your wrist. But the down side is that watch displays are so small compared to a phone’s display. (And how many people no longer even wear a watch these days, instead using their phone for telling time?)

    The main thing that I think could be a drawback is that the start ups who sell the watch to the end user will want to differentiate themselves and not all have the same physical design. At the very minimum they’d want the watch in different colors, and I’m sure most, if not all, would want to have a different shape, form factor, etc. Is that part of the plan?

    Looking at their web site it almost seems specious for them to say that it’s open source since nothing is available for download. I’m guessing that you’ll get the sdk when you buy a watch. I think it would be better if they offered an emulator that ran on your computer, like google does for android, so that you could try before you buy.

    Reply
  10. lumpynose,

    Using the same internal module while changing the outer design is exactly how this was designed. Someone with a business opportunity that requires a different look, color, material, shape, ATM rating, or brand is something that we have contemplated business models to handle. Before then, the standard watches can be used to develop and proof of concept.

    Preliminary API documentation will be posted within the next couple of weeks. The remote protocol will likely be the first to go up. The source projects will be available for download when MetaWatch ships in July (expecting to use Github). If we finish earlier, then we’ll put that up early too.

    I like the emulation idea, but there are not any effective tools to do emulation of a TI MSP4305438A. The most we could hope to accomplish would be a simulator for the remote protocol, but if you are interested in the embedded firmware, this will not help you much.

    Between the API documentation, examples, and discussion I hope you are able to get enough information to decide if you want to play with it.

    Announcements, demos, and pre-order are a little ahead of the documents right now, but if you follow @Meta_Watch on twitter or go to http://www.metawatch.org and join the mailing list, you’ll be alerted when new documents are up.

    Reply

Post a Comment

Your email address will not be published. Required fields are marked.
If you would like a profile picture, simply register at Gravatar, which works here on DCR and across the web.

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>