Wirelessnorth.ca

For the last week or so, we at WirelessNorth.ca headquarters have been trying out Bell’s first foray into the Android world with the Bell Motorola Dext. You may remember that Bell and Telus have been jointly spending billions over the last 3 years rolling out a new nationwide HSDPA network. A big driver for such a major strategy shift was to allow these two carriers to catch up to Rogers in being able to offer the latest GSM based devices like the iPhone and now the Android.

The Dext was originally launched by Motorola back in October 2009 but, after some delays, Bell is just launching it now. Bear this in mind, this delay explains a lot about some design decisions that might otherwise be perplexing about the device. Six months may not sound like a long time, but that’s a whole half a generation in the hyper-accelerated dog years of the current smartphone industry.

What we like:

The phone is built well. The Dext is by no means dead sexy, but it’s solid and it feels sturdier than the first generation Google G1 slider phone.

At launch, the Dext is pretty cheap from Bell. $79 with the usual (egregiously long 3yr) contract or $399 without. Nonetheless, this compares favorably with top-end Android phones which tend to hover around $200 subsidized or $600+ new. Bear this in mind as well. We at WirelessNorth.ca are all for (almost) anything that helps make smartphones more accessible to more Canadians.

The DEXT is no DROID. Whereas the Droid (Milestone in Canada) is Moto’s top-end smartphone, complete with huge screen and snappy processor and the latest Android OS. The Dext not so much. While being a budget Android, Dext is still a well featured device the goodness you would expect: an (ahem, vintage) Android OS 1.5, 7.6MBps 3G, GPS, Wifi, touchscreen, a real web browser, an acceptably useful camera etc.

The Dext also comes with MotoBLUR, Moto’s proprietary social-sauce add-on for Android. When you first turn on the device, the phone asks for your log-on credentials to Facebook, Twitter, Flickr, Gmail and more. So after handing over all the keys to your every online identity (what could go wrong?!), the fun starts. Motoblur aggregates all your contacts and status updates from all your social services. The phone asks if you want to sync all your phone contacts with either Facebook or Twitter. So that whenever you look up a contact you see their current avatar and status. Every phone should do this. Motoblur also has some potentially handy cloud features like online backup and device remote locate/wipe.

Other MotoBLUR features are less handy. There’s a universal inbox that supposed to pull in all sms, tweets and email. But for us it didn’t add enormous value when the likes of Twitroid and Gmail offer slightly better clients on their own. And then there are the desktop widgets. The Dext lets all kinds of social updates stream to your Android desktop. Motorola calls these “happenings” (slightly incongruous hippie connotations anyone?). In some future revision MotoBlur we think this might be a great feature. For the moment, limited screen real estate only displays one tweet at a time, and the most frequent update rate is … “once per day”. For someone with a few hundred to a few thousand contacts on various social networks, an always-on display of one arbitrary tweet per day is… less than useful.

Caveats:

A downside going with MotoBlur is that it saddles the phone with an outdated version of the Android OS, Android 1.5. Astute readers will note that this puts the Dext at about 4 versions of Android behind the latest 2.2 just recently released to users of unfettered Google (NexusOne) phones.

One of the advantages of owning an Android is that you should be benefiting from the rapid iteration of updates as Google steadily improves the OS. We have yet to hear from Bell or Motorola when Dext users in Canada can expect some overdue updates, but we’ll update this post whenever we find out. For now we have to give the phone a hammer for OS1.5.

You may recall the WirelessNorth.ca mobile device rating scheme. We merely start with the expectation that every device should be designed perfectly. From there we subtract one “hammer” just for the average number of times in the day you’d feel like whacking the phone (or it’s designer) with a hammer.

There are two more hammers. The first is the processor. Under the hood, the MotoDext runs an Arm11 cpu. The Arm 11 is an old design of the same class as powered the original iPhone and the Android G1. Newer/better smartphones these days are running the much speedier (but confusingly named) ARM CortexA8 based designs like the snapdragon processor in the NexusOne or Milestone. The Dext cpu is just fine for running all the basic Android functions, and the device is great at streaming video (e.g. Youtube) over 3G at full screen resolution. However, more demanding apps like many newly released Android games can be unplayable slow on the device.

The last hammer is for battery life. Brand new, with everything turned on, the device will just barely last you the day. Play with some apps or stream a few videos on your morning commute and you may find yourself running low by lunchtime, or have the phone run flat on you just before you’re trying to meet someone for dinner. Unlike, say, BlackBerries, Android devices are not known for good battery life. In this case, a less efficient processor combined with a lot of background processes don’t help either. MotoBLUR does provide some battery saving modes, but we sill found ourselves wanting to carry around the charger a lot.

These three hammers are the worst we can say about the phone. If you can live with these drawbacks, the phone has a lot to offer. It beats the pants off any feature phone and it compares well against what else passes for budget smartphones out there. For web/internet connectivity the Dext beats any EDGE-only BlackBerry, and (assuming we eventually see some updates) you can say at least it’s not based on a dead-end platform like WinMobile6 (which against all rational sense all carriers are still selling).

Conclusion: The Bell Dext is probably one of the best deals out there for getting your hands on a relatively inexpensive Android

So, with caveats, we can recommend the Dext. The most important caveat is that it won’t last you well a full 3year contract. It’s a good 2-year phone… that Bell is selling 6 months late. But then again the stats say most people do replace a phone every 18months anyway. Otherwise, you can expect it’s CPU to struggle with future apps as faster Androids become the standard. You can also expect the already tight battery life to certainly not improve after a few year’s use. Of course, more future-ready Androids will drop to it’s price point at some point. But, for a window of time the Bell Dext is probably one of the best deals out there for getting your hands on a relatively inexpensive Android. With “only” 3 hammers against it, we’re prepared to give the Bell Motorola Dext a firm rating of “WirelessNorth.ca: Mostly Satisfactory”.

Thanks again to Bell for kindly providing the hardware for our testing.

The Good The Price Android OS Solid build quality A couple handy moto-blur features The Meh Modest screen size The rest of moto-blur (but future versions could be promising) The Hammers Pretty terrible battery life Android 1.5 1997 called and want’s their mobile processor back

UPDATE June 14, 2010: That didn’t take long. Virgin Mobile Canada (which you know is also Bell) just launched the HTC Legend at $79 ($349 no plan) with a slightly faster processor and Android 2.1. The iPhone4 release also pushes the price of the iPhone 3GS down to $99. The poor Dext is now a tough sell unless it’s price comes down significantly (and/or until Moto comes out with a damn Android 2.1 or better update).

If you have played with an android for any length of time, you may have noticed the neat battery tracking feature in the Android OS. When the battery reaches 15% the device will warn you that it’s low on power, and there’s a little button next to that that says simply “Why?”. Click on “why?” and the device will give you a nice bar graph breakdown on where the battery went. Here’s a screen cap from the google developer blog, apparently of a device that doesn’t get used much:

This is also a neat feature, because it gives us some insight into the design constraints for modern smartphones. Of course everyone’s millage will vary, but here are the averages across 5 different recharge cycles over a week’s time of daily usage for own android:

Note we’ve lumped together a few things like android core apps, android system, and android OS into one category of “android stuff” cause we’re not really clear on the difference between those things anyway. The foreground apps here were mostly things like gmail, maps, some foursquare etc.

A few other observations: holy cow do screens suck power! Still want that big beautiful screen on your next smart phone? Also power usage of background apps is highly sensitive to their update frequency. Apps like background news readers or twitter apps can actually suck up to 30% of your battery (we had this problem at first) until you tune down their update frequency. This is definitely something app devolopers should be sensitive to too.

How does your usage vary? let us know in the comments.

Still waiting for your iPhone 3GS? You are not alone. Since the launch of apple’s swankiest new unit we’ve seen the waiting list stretching out months for anyone foolheardy enough to have ordered one online, or from a Rogers call center rep. Rogers retail stores do have them, but only in handfuls at a time. One store rep we spoke with said he was only getting a third of the number of devices they were requesting on a weekly basis.

The reason? The iPhone is costing Rogers too much. Apple’s elegant one $199 price doesn’t jibe with Rogers subscriber revenue. Or at least the math doesn’t jibe as well with the kind of margins that Rogers has gotten used to earning off of smartphone subscribers. Rogers is suffering margin compression with the iPhone 3GS. While they love the halo effect of the iPhone, they’d love even more if they could lure you in the store with thoughts of iPhone dancing in your head… and then sell you the exact same voice and data plan but fob you off on, say, a much less subsidized HTC Android for example.

We don’t know if Rogers is revenue sharing on the iPhone with Apple but, apparently, the difference between the actual price of the device and mandated $199 subsidized price is such that Rogers is going so far as to aggressively market the Android down to $79 (and $39 for students).

This is not such a bad story for consumers. The android is a half-decent device, and $39/$79 is the kind of entry point for smartphones that could really start driving penetration of smart phones in Canada (if not, you know, for those whopper monthly bills you’ll still enjoy but we digress). And a few more androids or blackberries in the global marketshare pie chart does help to keep apple honest.

The news is spilling over the interwebs that the first Google Android handset is close to (finally) seeing the light of day. Long maligned by the likes of Apple, Microsoft, Nokia and everyone else as mere vapourware, it’s now decidedly looking less and less vapourous. The first model is the HTC Dream (nice name). Supposedly to materialize between November and January, depending on your level of optimism. In any case, the dream really exists, it just passed FCC trials. You may remember it took similar time for Apple’s first iPhone to move from FCC to production and launch.

The significance for north of the border? The first handset directly supports the same 1700MHz 3G band (AWS) just auctioned in Canada (thanks T-mobile). An entirely hypothetical bonus, at least until any/all players get around to rolling out service on the shiny new ether. But one hopes a sign of good things to come. Mmmm open handsets, open mobile platforms. The idea almost makes us misty.

ISupply has released the results of their tear down analysis of the new 3G iphone. Apparently the component and assembly costs of the device is $179. Forgetting for a second the hundreds of millions Apple likely has and will spend on the design, development and marketing of the device… that’s pretty cheap. So want to know what the things are made of? (they’re not just full of stars apparently)

So there you have the ingredients. Though I don’t think that just the list and a trusty ikea alen key is going to help you put one together.

Anyhow, here is the real kicker:

“At a hardware BOM and manufacturing cost of $173, the new iPhone is significantly less expensive to produce than the first-generation product, despite major improvements in the product’s functionality and unique usability, due to the addition of 3G communications,” said Dr. Jagdish Rebello, director and principal analyst for iSuppli. “The original 8Gbyte iPhone carried a cost of $226 after component price reductions, giving the new product a 23 percent hardware cost reduction due to component price declines.”

Imagine, for a second, a world where great warehouses in shenzhen churn out pre-designed Android devices (which would basically the same components) or devices with the newly-free S60 OS or cheap-ish windows mobile. Then imagine the churning out of such devices a year from now 23% cheaper and with a few more features, then 23% cheaper than that another year later… all hitting the shelves of Walmart, Canadacomputers or 7eleven at nominal markup.

This (thank you Moore’s law) is another reason why the next several years are going to be a very interesting time to be in mobile.

Monday 14th saw the latest Mobile Monday gathering at the Fort York Armoury. Nick Patsiopoulos and Shyam Sheth, from Yahoo! And Google respectively, shared their company’s vision on the future of Mobile Applications. The choice of subject is no coincidence, following one of the most hotly debated topics in the industry – Is the future of mobile applications in native applications or within the browser? Several in depth online discussions have been made on the subject, for instance Dean Bubley and Michael Mace [see also Is the future in browsers or apps? previously on WirelessNorth.ca].

Yahoo’s Blueprint, the core platform for Yahoo!Go v 3.0, seems to be part browser and part native application. The philosophy behind Blueprint’s beta version is breadth over depth, with the intention of accelerating development of applications for the mobile web. Yahoo! Wants to give newbie developers to the mobile medium a standard set of tools, to overcome the hurdles of native application development. In summary, Yahoo!’s pitch to content developers is – you focus on building your data and idea, let us focus on presentment and look’n’feel on the multiple devices and OS’s.

But do I really want to have Yahoo controlling how my application/widget gets displayed? How would I differentiate my content’s user experience from that of the next app, if they all look the same? We all know looks matter – but here everyone is a good looking mannequin (Nick – if you are reading this – please feel free to comment).

Google in contrast, would like to take a more web centric approach, leveraging web tools and conforming them to the mobile space. Who hasn’t heard of Android. Google’s open platform philosophy and web-centric approach is here to stay – add Google Gears Mobile to the growing list of web driven mobile initiatives. Nothing new here, watch a few Android videos on Youtube and you’ll get the point. In summary – the IPhone is here to stay, we will have more iphone like devices, and the browser/web based approach is the way to go.

The most interesting comment in the post presentation discussion, centered around consumers, segmentation and devices. If Google’s and Yahoo’s focus is on the iPhone consumers, where does that leave average, non-sexy device users?

And Shyam made a valid point – the real users of mobile internet and services will be the iPhone (and other high-end device) subscribers. If that is the case, then as a content developer, one should really target the browser and high end device market. Don’t worry that the smaller devices don’t render well or do not have sophisticated browsers, that user is not really looking for a mobile web experience in the first place. And, with the churn in the device market, the average device capability will tend to what is considered higher-end today anyways!

Focussing on the smartphone type user, you are targeting an early adopter – they are the ones who really care enough about content and technology, and will drive the initial traffic to you. The mid-late adopters follow the popularity trends set by the early adopters. If you succeed with them, you have greater chances at mass market adoption (read Seth Godin’s Purple Cow).

But really, all these discussions are academic – if the road (and by road I mean the mobile data ) to the content have high tolls (and by tolls I mean high usage charges) and are blocked by the network operator. To have mobile content take off, mobile data needs to take off – and prices need to get cheaper to make that happen.

The next installement of MobileMonday promises to feature some guest speakers from Japan – the frontrunners in the mobile data and content space. Should be interesting and I hope they talk about how their cheap data rates helped accelerate data usage and drive traffic towards applications.

Heri asks: “if you had to choose between developing for the iPhone or choosing android, which one would you pick?”

The choice is easy for now, there’s lots of iPhones out there, not so many androids. Build iPhone apps. Get your app out there, learn and iterate. Build an Android version when there actually androids out there to run it.

Android could be big news, I breathlessly await a flood of cheap highly capable off-brand Chinese or Taiwan made Androids to come rolling in across the pacific. (who would need to buy a subsidized phone and 3year contract anymore?) But even in the optimistic scenario, I see Android as a 2009 story at the earliest. Not till then will we a significant number in the field, and not till then will we really know what the killer features and form-factors will end up being or if Android is a hit at all.

I don’t think the form factor you see in the emulator will be it. It’s a bit fugly.

If you are a developer, I’d say learn the iPhone first. You’ll find more work, and more immediately rewarding work, in the near term building for the platform with the larger installed base.

Anyone care to disagree?

This article by John Lombardo A developer’s perspective on Google’s Android is a good introduction to Googles open Android platform for future mobile devices. Android sounds like an  interesting platform:

“Unlike many embedded operating environments, Android applications are all equal — that is, the applications that come with the phone are no different than those that any developer writes. In fact, using the IntentFilter API, any application can handle any event that the user or system can generate. This sounds a bit scary at first, but Android has a well thought-out security model based on Unix file system permissions that assure applications have only those abilities that cell phone owner gave them at install time.”

While you can’t get an Android phone yet, there is still lots to play with Google’s SDK kit and the phone emulator which runs on your pc.

Whether you can make a business out of building Android apps or as a consumer whether you’ll get a chance to use them, depends on how good and how well distributed the devices are once they finally reach the market. With the clout of Google behind the platform, Android could be good.