On HP’s Touchpad
I got the chance to play with an HP Touchpad at Best Buy today for about a half-hour and wanted to share some of my thoughts about the experience. I’ve been eager for many months to finally get my hands onto a WebOS tablet. After my experience, I will repeat what I have posited before and even go a step further: WebOS is certainly the most legitimate competing platform to iOS in the tablet space and, in some ways, has the potential to outdo even iPad.
Unfortunately, “potential” remains a key qualifier, but we’ll get to that in a bit. First, I’d like to discuss what, to me, is perhaps the key usability battle in the tablet space, and the one in which WebOS is strongest: app switching.
Allowing only one app to run on screen at a time makes a great deal of sense for the tablet market for several reasons. First, screen size, resolution and the need for fairly wide tap spaces for fingers would make it difficult to run multiple apps (and certainly more than two) on a tablet screen at a time. If you need any proof of this, look toward the VNC apps out there and try to control your Mac or Windows PC purely by touch; it’s doable but not efficient.
Secondly, running only one app on the screen at a time helps to cap the necessary resources to power a tablet experience. Running at most, say, one app actively, one music stream in the background and a few lightweight notification threads in the background is the constraint that allows tablet battery life to be so remarkably good on average, especially compared to laptops that are much larger. Software is efficient and operating systems (or at least iOS) are cutthroat, prizing every kilobyte of RAM as if it were 1990.
Despite the logic, full-screen apps aren’t how most people are used to working with a computer. The idea of having a browser window, chat window, Word document and Twitterstream going on the same screen on a Mac or PC is commonplace; on a tablet it’s almost heretical. Because of this, app switching is an incredibly important part of the tablet experience; people need to be able to get in and out of the apps they’re using quickly and accurately.
In iOS, users select apps by tapping on one of the icons on the home screen, known as the Springboard. Up to six of these icons may be placed on the dock at the bottom of the screen. All others are stored above; the user can swipe back and forth to browse their collection of apps 16 at a time. Folders can also be used to help categorize things, and expand to reveal their enclosed apps when tapped on, each capable of holding a dozen apps of their own.
iOS 4 introduced multitasking, whereby a new, underlying set of app icons appears when double-tapping the home button; these represent active applications. With iOS 5 comes 4-finger gestures. A pinching motion will close out to the Springboard while swiping left or right will jump immediately into other apps. However, this method is flawed in that there’s no way for the user to reorder the apps that are brought up by swipes. Rather, they appear in reverse-order of how they were opened. This can often be problematic.
For example, say you’re browsing the web and realize your Wi-Fi connection is acting up, so you jump to Settings before going to Mail as you initially planned. You wanted to reference an email and a website, but now you have to wade through the Settings screen to get between Mail and Safari unless you once more leave Mail manually and reopen the browser. This may be the epitome of a first world problem, but it’s also a serious issue for iOS because it breaks the magic of the system, if only for a fleeting moment.
Compare this to the WebOS, card-based system. Once again, a dock exists at the bottom where popular icons can be stored. To the right side of this dock exists a “more” button, in which all other apps are hidden. They are categorized by tabs into Settings, Favorites and All Applications. This leaves the default screen largely empty, leaving a hole where iOS would simply put more icons. It’s in this newly-liberated area that the WebOS card system takes hold. WebOS shows active previews of each active application and allows users to swipe between them, like they might with a deck of cards. To quit an application, one merely swipes up on the card, spinning it out into the universe. To enter an application, a tap will bring it to full screen. To return to the deck of cards, one swipes up from the bottom of the screen while in an application.
This is old news to anyone that’s used a WebOS device like the Pre or Pixi but is a revelation for the tablet market. WebOS relies– correctly– on a visual model of navigation, rather than the mental model required for the iPad. With WebOS, you’re looking at slightly-smaller versions of the screens you wish to choose from. With iOS, you’re either get no visual cue (four-finger swipe) or can only navigate based on recalling which icon represents the content you’re trying to view. This can be surprisingly problematic. For example, let’s say that you were reading a PDF document this morning but played a few games since then. You remember only that it was a PDF and what it looked like, without being able to recall how you received the document. Were you viewing it in Mail or in Safari? Perhaps it had been imported into iBooks already, or you were previewing it in Dropbox? With WebOS, you see the document, you tap it, you’re back in business. With iOS, it might take a while to figure out where you were at last.
The card system is absolutely, as I’ve hoped for quite some time, a better solution than the icon-based navigation offered by Apple, especially on a 9.7” screen. WebOS loses some of its advantage on smaller screens, where an icon might be an easier visual cue than a 2 inch version of a full page document. But in the tablet space, I have no doubt that the WebOS implementation is superior, especially given its speed of execution.
However– and this is the heartbreaking part– the Touchpad itself is simply not a good device at the moment. It feels sturdy in the hand and not too heavy, although I did notice an almost anti-gravity type experience when picking up an iPad 2 right after holding a Touchpad for a half-hour. The hardware is generally not an issue. But the software remains critically broken in many places once the user leaves the excellent navigation area.
Many are small issues in the greater scheme of things: I had some difficulty connecting to Wi-Fi at first, the Bing Map software was fussy about my location and provided uninformative error messages and the spellcheck is woefully lacking, missing simple things like contractions (cant to can’t, wont to won’t, etc). But the greatest issue is the screeching halt that the Touchpad will often come to when performing the simplest of tasks. For example, switching from landscape to portrait mode took over ten seconds while loading DaringFireball.net, an exceptionally lightweight website. The mail client took a good 3-5 seconds to boot up for the first time. Resizing window panes in mail (another neat UI technique similar to Twitter on iPad) would trigger at least a split-second wait time and again.
These sort of issues are simply unacceptable and were never present on the iPad, even on day one. What’s both good and bad about these issues is that they’re so blindingly obvious– and the performance of the hardware so incredibly good in spurts– that it simply must be the software to blame, and not a lack of horsepower from its 1.2GHz, dual-core Snapdragon processor or 1GB of RAM. This means that the issues can– and probably will– be fixed in short order. But it raises the question of why they didn’t wait to launch it until the 1.0.1 patch was already in place. In a way, I’m afraid that HP has squandered the opportunity to immediately leap into second-place for overall quality in the tablet space because they didn’t want to delay the Touchpad anymore.
Aside from performance, the lack of a solid music or movie store on the device certainly doesn’t help matters, and for the platform to launch with only 300 or so apps on the day the iPad surpassed 100,000 is challenging, to say the least. But these issues can be resolved. The app catalog will take care of itself if HP can solve improve performance and garner hardware sales, and Amazon’s MP3 Store would feel right at home here. Indeed, the only glaring issue is the maturity of the platform.
Speed must improve, and not merely on the next Touchpad but on this Touchpad. Truly competitive platforms can’t survive on the promise of the next iteration, because Apple will simply leapfrog them again. The Touchpad must be brought up to, or close to, parity with the iPad 2 in terms of performance before a Touchpad 2 or iPad 3 ship, period. As it stands, the Touchpad shows us two things: how very excellent WebOS can be as a tablet OS, and how very far ahead of everyone else Apple truly remains.
It’s a point of sales problem. It’s an expertise problem. It’s a marketing problem to consumers. It’s a price point problem, and it’s a software richness of content problem.