Dell Wyse Project Ophelia was still in the news this week, and it even came up in a Business Insider article that positions Ophelia as part of a philosophical change at Dell leading up to an expected leveraged buy-out by private investors (led by Michael Dell) who want to re-shape the company. They say that Ophelia is part of the future model of how we compute, and is a part of the future of IT services delivery.
That's a ways out yet, and so is Ophelia, for that matter. Still, I wanted to try the experience to see if my concerns were on point or not. You can read last week's article, but in essence I wrote that I don't think there is an actual use case for an Android device that plugs directly into a display's HDMI port but still requires a keyboard and mouse (or a touchscreen) to interact with. The applications aren't written for that experience, they're written for mobile. The device is built to do everything, which means it likely doesn't do anything all that well. For example, the experience with the apps won't be as good as a phone or tablet, and it won't connect to desktops as well as a thin client.
I wanted to like the device, though, and since it compares so well with the Android Mini PCs on the market, I thought I'd get one and give it a shot. Dell's VP of Cloud Operations (and former Wyse CEO) Tarkan Maner was quoted saying the price of Ophelia would start around $50, so I bought a $48 Android Mini PC from Amazon. What follows is my review of that device, but more importantly, my take on the use case challenges presented by it.
To set the playing field, the device I ordered was the Ug802 Mini PC advertised as running Android 4.0.4 (it arrived with 4.1.1). It's advertised as having quad-core graphics processing, a Cortex A9 dual-core processor running at 1.2 GHz, 1GB DDR3 memory, 4GB internal flash, support for a 32GB Micro SD card, and 802.11b/g/n wireless. The box I received had no N support, or at least it didn't recognize any of my N networks. I chose this device because it was more powerful than the next runner up at $38, and because the description didn't include phrases like "Full extension let you be like a tiger with wings added" in the description. That is not a joke.
Not quite the size of a USB stick (or a can of Diet Coke)
The first thing you notice with one of these is that it's hardly small. I mean, it fits in your pocket, but it's just about what you'd expect when you think about it as essentially a phone or tablet without the screen or battery. The guts are the same…just crammed into a different form factor. This isn't bad, but don't get your hopes up for something super tiny from Wyse. Will it be smaller? Probably, but not by much when looking at this side-by-side:
That's not a show-stopper, though. It's going to be larger, just don't let "USB stick-sized device" fool you.
Next up to look at is connectivity. The Android Mini PC I bought came with a standard male HDMI connector and a short extender for those with unforgiving displays or mounts. It also has a micro-SD card slot, a micro-USB port for power only (this device is not powered by MHL), and a USB 2.0 port for connecting a keyboard or other device. It can be used with a hub, so that's nice.
Once turned on, you can connect it to a WiFi network, but it only supports B or G networks, not N. That's a pity, because one of the use cases of this device is as a media center, and I think you really need N to pull that off. With this device it doesn't really matter, though, because the WiFi is so terrible you can't really do much anyway. How bad? 50kbps--that's a lower-case b.
In addition to missing 802.11N, this device is also missing Bluetooth. The Wyse device will include this, so that's already an improvement, but whether I have to plug in a USB dongle or not to get a wireless keyboard to work, I still have to lug a keyboard and mouse around with me to use it.
This is really the entire point of getting the device, and my expectations were low. Having said that, I'm comfortable saying that every expectation I had was met in that the general experience really, truly…sucked. I don't mean a little, I mean a lot. I was actually somewhat surprised at first…the mouse pointer worked well, the keyboard seemed to be ok, but then I started digging into menus and trying to get applications.
Since this is Android, it's made for touch, and it's made for you to be able to swipe up or tap the screen to get to the home and back buttons. This UI is still there on the device, but you can't click on it and gestures don't work to make it appear. Sometimes it does, sometimes it doesn't. The only surefire way to navigate is to right click, which acts as a back button. If you're 10 menus deep and you want to go home, that's 10 right clicks. On top of that, the interface has been skinned with a 10ft UI, meaning everything is huge and most of the icons look like crap because they're blown up way beyond their nominal size.
That's certainly inconvenient, but workable, so I decided to try some apps. The first one I got was Wyse PocketCloud, because I figure this will be a staple of Project Ophelia and I use it regularly at home and for work. PocketCloud is my favorite remote desktop client for iOS since it works with everything I need it to (if I needed HDX support it would be different). Pocketcloud installed and ran fairly well, although the lack of touching and full mouse control meant that I couldn't use it. This was the same with all remote desktop apps I tried (2X's, plus a couple other random ones)…nothing was coded 100% for mouse input. If any of them accounted for right clicks the system appears to override that and interprets them as a back button click (which ends your session!).
So the remote desktop use case was a bust, but what about other applications? Essentially it was hit and miss. Some apps, like Flipboard, worked great. Others, like Flick Kick Football didn't. I'm certain it has to do with how the app interprets the input, whether it's gestures, or x,y coordinate taps, or something to that effect. The bottom line is that these applications were made to be used on tablets and phones, not in this form factor and not in this use case. Angry birds played, but you couldn't pinch zoom. Try playing it without zooming…stinks, right?
Speaking of tablets and phones, this device (or at least Google during setup) believes it's a tablet:
I tried watching videos on it, but the device wasn't able to keep up with anything over the network. I didn't have any MicroSD cards to try local execution. YouTube videos played in Chrome (the built-in FireFox doesn't have a Flash Player), but the terrible WiFi meant that I could only watch a few seconds at a time.
If this all sounds frustrating, you're right. There's one more thing that really messes up the use case for me for this device, though, and that is keyboard support. It's quite clear that this is per-manufacturer thing, since my Logitech Revue Google TV works just fine (never thought I'd say THAT!), but using the keyboard on this device is almost impossible. There are a few built-in apps like Firefox (or at least a browser with Firefox's logo) that work, but when I tried to use the native Android Chrome app, the Enter key wouldn't work. The number pad key would simply type a Z, while the carriage return would do nothing. That means that I have to bring up the soft keyboard (which comes up every time you go to a text field because it thinks it's a tablet) and click on the Enter key. And so it goes in this manner--something appears to work all right except for one maddening problem--until you finally give up.
At the end of the day, I can't think of a single reason to use this device. If certain things worked, I could see someone having use for it, be it for work apps or consuming content, but I already have devices that do that and don't require me to bring along a keyboard and mouse. As I said, I had low expectations of a $50 Android Mini PC, but I was really hoping to find a silver lining. I know I didn't get the top of the line FXI Cotton Candy, and I'd be happy to test that out if someone knows it will perform better, but I'm inclined to believe that if they can sell it for $200 that it's not the same as Dell Wyse's $50-$100 device. I could be wrong.
What does this all mean to us and to Dell Wyse?
Dell Wyse has a long way to go to take this kind of platform and make it good enough to use. I think they can do it, but I still don't know of the use case. None of my monitors are touch screen, and certainly none of my televisions are. That means I need to use a keyboard and a mouse or buy touchscreen displays. Plus, if I want to travel, that means I have to bring one of those things with me or hope that wherever I'm going has the proper hardware. From my desktop, this might be ok, but the problem still remains that the experience isn't as good as a thin client for remote desktops, and it isn't as good as a tablet or phone for Android applications.
While I'm certain Dell Wyse can make the experience top notch, they have a long, long way to go if they're starting from the same place as the device I used. Plus, as I wrote for SearchVirtualDesktop in an article about how I'm over Android thin clients, even if they do get it right, the applications are still being written for phones and tablets. If Dell Wyse is trying to change the way developers create their applications, making a device that can provide a decent experience is the least of their problems. What do you think--would you use it?