Conclusion Up Front
I have to say: I am very impressed with the Jawbone UP3. I’ve had it for two weeks now and it has been absolutely fantastic. I love this device, and I disagree with the common points around which an apparent consensus of disapproval has formed among Internet reviewers; it seems the reviews for the device range from declaring it a high quality, perhaps overpriced device, to abject hatred.
I thought I’d put my voice out there saying, on the contrary, that it is in my experience a very useful, accurate, and comfortable device, and its hardware sophistication and the great software accompanying it gives me hope that Jawbone will really deliver on their promise to continue its development with firmware updates. I think that’s a winning IT model, and I think we would do well to support it.
The device is not without demerits, of course, and I cover the two big ones at the bottom of this article. Unfortunately, I fear the major critique (the minimal approach to nutrition tracking and the near-uselessness of associated resources such as the food database) is common among devices, so ultimately, I don’t think there is a superior choice to be made in this department.
That said, let me put forth a summary of the capabilities I desired in such a device, and then I’ll move into some more detailed review.
I am interested in tracking the most data with the most relevance to the overarching project of this blog. Already underway is the documentation of the technological platform on which I am basing my endeavor to engineer a spiritual oasis amidst the ruckus of modern life, and a biometric device which assists the wearer in governing the body is a welcome addition. There are few causes which demand more immediate and focused attention from the mind than the body, being that it is the platform from which all can be accomplished as a human being.
In envisioning a biometric solution to this problem domain (governing the body), I immediately conceive of the following monitoring capabilities:
- Emotional State
This seems well within the scope of operation of modern fitness trackers, so I set about finding one which met these criteria. I also conceived of the following preferred technical aspects of the device I would ultimately select:
- Access to the raw data
- This facilitates personal archives of the data along with personal interrogation of the data through whatever software I should find appropriate
- Free Open Source Software
- Support for Open Source platforms such as GNU/Linux or FreeBSD
- Support for the integration of other biometric devices (such as a scale)
I’ll say right now that the middle two bullet points went right out the window at the onset of investigation. There just aren’t any advanced biometric tracking devices currently developed with a FOSS mentality, nor are any apparently supporting GNU/Linux or FreeBSD. And that’s a shame, ’cause this is exactly the kind of project that would benefit greatly and suffer virtually no costs in adopting such a mentality, if you ask me.
The raw data access, on the other hand, is provided for free by Jawbone for the UP devices. FitBit will charge you $40 a year for a premium membership which offers, among other services, access to the raw data, and that seems kinda lame to me.
That said, let us move on:
The Decision to Purchase the UP3
I read through a lot of reviews online looking for a good physiological monitoring / fitness tracking solution. I started with the Best fitness trackers 2015 review at wareable.com which covered the best devices of 2015. From there, I progressively branched out to devices they hit upon which piqued my interest, and then devices others related to those, and so forth.
I considered the Jawbone UP2, which was positively reviewed by wareable.com, but its rudimentary sensing capabilities (limited to, it seems, a three-axis accelerometer) leave me wondering how it’s possibly doing some of the things it claims to be doing. For example, both the UP2 and the UP3 provide detailed sleep pattern records which include awake time, REM sleep, stages 1 and 2 (represented jointly as “light sleep”), and stages 3 and 4 (represented jointly as “deep sleep”). How, one might wonder, does the UP2 device infer from an accelerometer which stage of sleep you’re in?
My guess is that it’s just kinda BSing it. The Jawbone site claims it uses “actigraphy,” admitting that it is estimating your stage of sleep from the accelerometer alone. I’m sure that data is not completely useless, but it’s not extraordinarily confidence-inducing.
The Jawbone UP3, however, seems to have an impressive array of sensors which I would expect might allow it to do a reasonable job of tracking one’s sleep to this degree. It features an accelerometer, two temperature sensors, and a bioimpedance sensor which can collect heart rate, respiration rate, and galvanic skin response data. That doesn’t say to me that it can track sleep flawlessly, but it sure does have a reasonable set of data from which to derive inferences.
The sophistication of the UP3 hardware platform is exactly what I’m looking for in a fitness tracker. Jawbone has stated that the hardware is so rich with potential that firmware updates will continue to be delivered with new features, and I believe them. I think that’s a solid plan for a fitness tracking device, and I’m on board.
Ultimately, the choice came down to the FitBit Charge HR or the Jawbone UP3. However, this wasn’t much of a choice (unless the UP3 was really terrible) because the Jawbone UP3 far outpaces the competition when it comes to detailed sleep tracking and the associated features (such as the smart alarm).
It’s pretty awesome.
Aside from cost, the only real advantage of the FitBit Charge HR (which I could detect) is its continuous heart rate monitoring, and honestly, I don’t think that data is that valuable. The resting heart rate monitored by the Jawbone UP3 is done very well, and I think it provides a solid and important indicator of my cardiovascular health. If I had some reason to really need to focus my exercise so specifically that I reach and sustain a particular heart rate, then the Charge HR would be indispensable, but since I have no such need, UP3 it is.
After 14 days of having the UP3, I’ve been exercising very consistently in line with the somewhat-recently-formulated (though long in development) theory about optimal exercise patterns for human beings (after the trial run, I’ll post the theory and results, documented in part by the UP3, as part of the blog’s project). My brain is very much conditioned to succeed in video-game-like environments, and this tracker sort of game-ifies exercise. Just today, in fact, I received notice that I am in the top 5% of all UP users in my sex and age group when it comes to average steps taken per day.
w00t. I’m on the global leaderboards. I am reminded of my efforts in chasing those PSN trophies in Killzone 3 which required that one rank in the top X% of all players globally.
Counterarguments to Prevalent, Somewhat Silly Critiques of the UP3
I don’t know if anyone will even find this useful (or have the patience to read it), but it seems to me there are quite a few harshly critical reviews of the UP3, and I’d like to respond to them because, basically, I don’t think they’re fair or even accurate, in most cases, and I would hate for Jawbone to be disincentivized from following what seems to me a winning model for the production of these devices (aim for hardware sophistication that can be built upon with expanding software features provided free of charge) because of some foolish review points.
So, here we go:
The Physical Design of the Device
The following reviews all make the same point:
- “The clasp is fiddly as anything and, although you’ll get used to getting it done up (there’s a certain knack that no words could possibly explain properly), it’s never easy. And it’s not all that secure either. We’ve knocked ours off reaching in a bag and it’s also fallen off over night while sleeping.” – wareable.com
- “It’s a little bit awkward to put on, to be honest. It does get easier with time, but it’s still not as natural as some of the Apple Watch strap designs, for example.” – techradar.com
- “band clasp difficult to adjust” – CNET.com
- “It took me nearly 10 minutes to figure out the best way to wear the Up3, and even after using it for a few weeks it still sometimes takes longer than I’d like to put it on. To make things even worse, the Up3 can easily come off if something hits your wrist in just the right way.” – engadget.com
- “unfortunately I still found the clasp to be wanting. It fell off my wrist no less than ten times, including twice in my sleep. Derp.” – gizmodo.com
So that sounds pretty bad, eh? Five pretty big-name reviews all lambasting the device for a fundamental inability to be suitably worn. In fact, I can’t even find a review which doesn’t include complaint(s) about this aspect of the device. I took this critique seriously, and it was perhaps the greatest worry I had when I purchased the device; if it doesn’t stay on comfortably, it’s not going to be a viable fitness tracker. I chose to purchase it nonetheless because I simultaneously held the position that this critique is the most likely to be based on user error or unwarranted disgruntlement.
I’m happy to report that, in my experience, those critiques appear to be little more than unwarranted disgruntlement. The device has not fallen off of my wrist once in the 14 days I have worn it. Not once! I walk, run, lift weights, play racquetball (!), toss small children into the air, and serve as a jungle gym for them, and none of it has caused the band to become dislodged in any way. The clasp hasn’t even come loose at all. The worst that has happened is that the clasp can get my hair stuck in it if I run my hand through my hair without thinking about it (I have long hair), but, y’know…it’s not an unresolvable crisis. The hair does come out of the clasp if I move the clasp in the appropriate direction. Rocket science, this is not.
I’ll grant the clasp is not an intuitive thing, but anyone with the greater reasoning powers of a human being is capable of understanding it quickly, adjusting its size properly, and wearing it successfully. This is a non-issue that, for some reason, every review writer on the Internet seems to find problematic. Allow mine to be one voice which says “not so” in this case.
Because the wareable reviewer above declared that the proper way of putting on the device cannot be put into words, I thought I might offer my very simple method of attaching the device to my wrist:
- Place the device on your wrist, as it will sit when worn, without attaching the clasp.
- Rotate your wrist so that the top of your wrist (where the device’s touch panel is situated) faces your thigh (or knee or whatever) and hold the device steady by the pressure between your wrist and thigh.
- Adjust the clasp as you see fit. With the device firmly held in position save for the bands which will be joined by the clasp, manipulating the clasp is quite easy.
There are some things I have written which have given me some confidence that I am a superior wordsmith, but that three step process is not one of them. These reviewers are serious whiners.
Lack of a Screen
- “There’s also no smartwatch skills on offer at all, sadly – it seems a shame that you can’t be alerted to incoming texts, emails and the like from a paired smartphone, not even just by haptics.” – wareable.com
- “The Bad: No on-band display means you need to check phone for stats” – CNET.com
- “Notably absent is a display of any kind—whose absence absolutely kills this thing” – gizmodo.com
- “Why is Jawbone so averse to putting screens on its wristbands? Basically, it’s all about battery life….But while I can understand Jawbone’s devotion to long battery life, I’d much rather give up a few days of charge for a usable screen.” – engadget.com
My God, why is this an issue? NOT EVERY DEVICE NEEDS A SCREEN. Not every device needs to be alerting you about text messages and emails! I am reminded of a Dilbert cartoon in which Dilbert stares at a ring with a single-character display to browse the Internet.
I actually think it’s a smart design choice to leverage the fact that no one without a smartphone is likely to buy this device. Jawbone decided to provide an elegant band with absolutely minimal user interaction, and they succeeded. I appreciate it. I don’t need more devices demanding my attention.
The interaction with the device is performed almost entirely (one does put the device into sleep mode and activity mode through a simple tap and hold gesture, but that’s it) through the application installed on any Android or iOS device. I have an iPhone 4S, and installing the UP app was a simple endeavor. The application seems universally acclaimed, and for good reason; it’s attractive, functional, and convenient to use (for the most part; see the critique at the bottom of the article).
- “Sometimes it syncs with the band, and sometimes it won’t. Sometimes it would refresh and sometimes it won’t.” – gizmodo.com
I have yet to experience any Bluetooth issues. My wife’s UP3 primarily connects to her Samsung Android device, and mine to my iPhone 4S. We also have an iPad Air to which mine connects on occasion. We’ve never had a problem. For what it’s worth, there that is.
Lack of Real-Time Heart Rate Monitoring
- “Which is pretty much how I feel about the heart rate tracker, too—because it doesn’t measure your heart rate in real time. You paid $180 for this sweat heart rate monitor you wear on your wrist, so it stands to reason that you might want to know what your heart rate is, y’know, whenever you want? You can’t. It’ll clock your resting heart rate every day when you wake up, and that’s pretty much it. This is so, so dumb.” – gizmodo.com
First: if you were under the impression that you were purchasing a “sweat heart rate monitor” (I presume you meant “sweet”), you somehow gained a false impression of the device. Furthermore, as I said previously, I think the real-time heart rate feature could be nice; I wouldn’t refuse it as an additional feature, but it’s largely unimportant. The resting heart rate data collected each day is objectively valuable and useful information to have, and I’m glad to have it. If you want real time heart rate tracking for some reason (and I’d love to know what that is), then this device isn’t for you. This should be understood from the onset. Jawbone doesn’t try to hide it.
The following line serves as a nice epitome of the reviewers of this device, if you ask me:
- “I see my resting heart rate is 57 one day, 60 another, 58 the next. Jawbone’s app hasn’t told me anything about what this means, or what I need to do next with that data. It feels useless.” – CNET.com
Seriously? Are you faulting the device for your ignorance? The purpose of the resting heart rate is that it serves as a baseline for judging your fitness. This is why it is useful information. Don’t become confused over day-to-day fluctuations; look for trends over weeks and months.
The Sleep Feature Set
- “You’re supposed to double-tap the device to wake it up, then long-press it until the little orange running-man shaped light goes off and the moon shaped light comes on. The problem is I usually have to double-tap it half a dozen times before it wakes up, which is pretty annoying. Even more annoying is that I have to activate it at all. Fitbit, Basis, and several others automatically log your sleep. You don’t have to remember to switch on a night mode when you go to sleep or remember to turn it on right when you wake up—which, to be honest, is something I forget to do about 70% of the time. It feels really… behind.” – gizmodo.com
Well, you are too pathetic to simply tap a device when you go to sleep. What can be said? This can’t possibly be seen as ruinous to the device, but I guess if you’re just that incompetent and easily annoyed by simple things, I understand.. The device does allow you to recover your sleep data if you forget to notify it that you are going to sleep. I’m sure Jawbone could’ve automated the whole process, but there’s probably a significant accuracy benefit in giving the device this one, little piece of information quickly and easily before you go to sleep or get out of bed.
Additionally, the wareable folks reported that they rose from sleep, left the bed, came back, and found in the morning that the device had failed to notice. I can’t say I’ve had the same experience; in fact, I’ve been consistently impressed with the apparent accuracy of the sleep tracking. While I can’t comment much about the stages of sleep and their accuracy, I can say that the device always seems to accurately report when I am awake and when I am asleep. Other reviewers seemed to agree with this.
This sleep tracking feature is important to me. For my entire life, waking up has been a horrid experience (save for every, say, one of twenty occasions on which I feel remarkably good). So horrid, in fact, that significant others in my life have not particularly enjoyed my company in the morning any more than I enjoyed..well..anything in the morning. As I grew older, I started to read and understand more about sleep and it seemed perfectly plausible that, given my notoriously deep sleep (it is not too easy to wake me, and when I am awakened, my consciousness and later recollection is questionable for a few minutes), I could reap great benefits from being awakened at an optimal time during my sleep pattern. The mere idea that I could now own a piece of technology capable of doing this for me drove me to purchase the UP3 rather quickly upon its discovery, and with my wife’s eager approval.
I am happy to report that I have been awakened very pleasantly by the smart alarm on many days. I haven’t actually gotten up in response on every day, but the days in which I do get up seem a hell of a lot better than being woken up in the ways I used to be woken up (children jumping on me or alarm clocks, primarily). It seems to really work, and I love it.
- “The UP3 is charged by a short, proprietary USB cable, because of course it is. But it’s worse than that: It has a magnet that supposedly holds the device onto the charging pins, but often doesn’t work in practice. If you don’t line it up just right, it won’t charge. ” – gizmodo.com
Dude. You have to charge the device using that cable once every seven days. I agree that the magnetic connection is not established properly each time an attempt is made, but for the love of God, it lights up to show you that it’s working properly. Just attach it again if you miss the mark. Is this that big of a deal?
Actual, Non-Silly Critiques of the UP3
I think the primary area of critique is in the food logging, which so many applications have attempted to do well, and so few have ended up doing very well. Jawbone has partnered with myfitnesspal.com (whose services I do not use and with which I am not well acquainted) to gain access to their repository of food information, but sadly, the vast majority of the food options in the repository are simply incorrect. Take a simple entry for broccoli, for example. It’s correct if you use gram measurements, but if you choose the “one head” option, you will now be staring at a food item with 40g of carbs, 9g of protein, 1g of fat, and, somehow, 16 calories.
And that’s one of the better entries.
Basically, in my experience, you have to just bite the bullet and enter all the nutritional information yourself 90% of the time, which is a shame. You simply can’t trust the data in the repository, and as a result, you have to interrogate each entry for accuracy, which ends up taking more time than just building an entry yourself given the minimal set of qualities available for entry; the food data repository, as a result, is borderline useless.
Very, very occasionally, I’ll run across entries I can use, but the application doesn’t provide a simple way to view the nutritional content of food items. You basically have to add the entry to your food selection, then switch to the screen showing your selections, and tap into the item to see some details. Even within the item, on the iOS app, you can’t see certain details like cholesterol or sodium, making it a total pain to try to infer the item’s nutritional value from the consequences of adding it to your food intake for the day. It’s not unheard-of to mistake a bad entry for a good one given that you can’t see all of its details, only to discover its badness when your end-of-day totals are way off somehow. Neither is it possible (so it seems) to modify entries to correct them, so even if an entry is correct save for its declaration that a slice of toast has 4,000mg of sodium, you can’t simply wipe the sodium entry out and use the new custom food item you’ve created as a result.
If they could sharpen up that food database (unlikely) or at least give the user the power to more easily view and modify it, that would be swell.
Additionally, the food items track only the bare essentials when it comes to nutrients. No vitamins or minerals are tracked (save for sodium), and that’s a big bummer. I used to use caloriecount.com, and I find their interface superior in both terms of features and usability, but their food database suffers from the same issue. I’m afraid any user-generated food database is going to be largely useless without some serious, perhaps professional, data curation.
Aside from that, the Smart Coach feature is not as useful as it could be. I’d say about 30% of its notifications are worth reading, and that’s a fairly low bar I’m setting. I very much appreciate the occasional statistical reference to the total population of UP users or some other data source, and I wish they would be more frequent. But, this platform is under development, and the UP app gives users the ability to like or dislike notifications to guide the platform’s development, so I’m making use of it and hoping others have similar interests and opinions.
Those are my two big critiques of the device, but given that the critiques are common to basically any similar platform I’ve seen and/or used before, they aren’t extremely weighty. It’s not a perfect device, but it is an excellent one.