Finance director for the company, Alexandre de Rochefort, says that even a company that’s done extremely well on the iPhone can have trouble breaking even on Google phones:

We have significantly cut our investment in Android platform, just like … many others … [The Android Market] is not as neatly done as on the iPhone. Google has not been very good to entice customers to actually buy products. On Android nobody is making significant revenue.

That’s the essence of the App Effect: High volume, customer pressure and nudging from Apple drive iPhone app prices down break-even levels, which gives app developers two options. They can either charge higher prices for equivalent apps on Android, for which they will be crucified by customers, or they can match their prices and hope that enough of Android’s comparatively small, fragmented user base just happens to stumble across said app in the barely navigable App Market. An attractive business proposition, I say!

So what needs to happen? Either Android adoption grows (which it’s doing), the App Market gets much easier to navigate (a desktop app, maybe?), or you know, both. [Reuters]

No wonder they “desire to have a carrier relationship” to subsidise the price. Sony’s long had a tough time cracking the US market on its own and the X10 deserves better.

Bonus news: Remember how Sony Ericsson’s page said it would arrive this February? Well, it’s now gone back to “to be announced”. Don’t be too concerned if you’ve got your eye on one… it’s still slated to arrive by mid-next year. [New York Times]

Both phones only work on the 900/2100MHz HSPA networks, which means you’re on Voda or Optus (and not Telstra like we heard), unless you’re happy sitting on 2G speeds. The Tattoo will also let you buy custom body cases for $30 from Garskin.

The lack of a carrier partner means that there won’t be any over-the-air software updates, which probably won’t phase too many Optus Dream owners. And while that price is a bit more expensive than getting an unlocked version from overseas, HTC were quick to point out that you won’t get a proper Australian warranty if you grey import.

For everyone who has been waiting for the Hero to launch in Australia, what do you think about this launch strategy? Is selling it unlocked a good or bad move on HTC’s part?

Well, rather, I know that it’s definitely something about the phone itself. It’s missing that spark, the emotional drippings of OMG, that say, the Droid has. The X10 is perfectly fine. It’s just that everything outwardly belies the specialness of what’s going on inside. The all plastic-build feels just slightly better than cheap, the shape is kind of awkward unless you have gorilla hands, the design — it all just feels, well, incredibly ordinary.

What stood out in terms of hardware was the screen (at four inches, with a resolution of 854×480, it’s big), and the camera, which borders on amazing for a phone. The extra betaness of the software meant we couldn’t really get a grasp on how deadly the 1GHz Snapdragon processor is, since while things were quite speedy, the phones also froze a lot, apps crashed or wouldn’t start, etc.

Sony Ericsson’s approach to customising Android with its own Nexus interface is mercifully respectful of your boundaries. That is, if you don’t want to use their TimeScape or MediaScape UI for managing your contacts and media (though you probably want to in the latter case), you don’t have to — the default android contacts app is still there. The only thing you can’t escape is all the blue. If you take Windows Media Center and imagine it ported to a phone, that’s pretty much the Xperia X10. But with more “infinite buttons”.

MediaScape actually feels like Media Center, the way it arranges and presents your photos, music and videos. The concept behind TimeScape should be pretty familiar at this point. Everything from a contact, like Facebook, emails, Twitter, photos, text messages, calls, whatever are integrated into a single interface, so you can check their status updates or get a hold of them however you want. That infinite button is what coagulates all of the services into a cohesive contact.

For the most part, I think the Nexus interface works (even as crashy as it was today on these pre-production phones). It’s easy enough to navigate, it stands out against the other custom Android interfaces with lots of bright colours and transparencies and it’s not bad too look at. More than that, you only use it as much as you want. (Some people might want a more complete Android UI overhaul, and that Nexus doesn’t quite deliver.) I don’t see how Sony Ericsson is going to get developers to put out apps just for Nexus, though I’m not sure how much that matters. It’s an example of Android’s potential in the hands of phonemakers who actually know how to design interfaces.

Sony Ericsson’s always had trouble getting their smartphones to crack the US, since they haven’t had carriers footing the bill for phones to make them actually affordable. It seems like they’re trying to make that change with the X10, saying it’s “our desire to have a carrier relationship for this product.” So you might actually get to use one. It still doesn’t set my pants on fire like the Droid did, at least not yet, so I’m not sure how much you actually want to. But maybe it just needs more time.

Example: Sony Ericsson Timescape, the “signature” Nexus app, sounds a lot like HTC’s Sense in that it pulls in a bunch of services like Facebook and Twitter for your contacts, but it’s more straightforwardly organised around timelines, which you can break down by service, events or whatever. There’s also an infinite button. For infinite discoveries.

MediaScape looks uncannily like Windows Media Center, if you stuck it on a phone. It’s actually nice. The camera software is swanky, with facial recognition and tagging like a real Sony camera.

Apparently, it’s not going to be restricted to just smartphones, but could reach some of their more “mainstream” phones. How they’re going to get developers to develop apps for Nexus though, is another question entirely.

We’ve heard rumours like this before, but this time there are a few distinct elements that seem credible, maybe even enough to make us rethink our previous position. The rumblings are a bit vague, but point to an outsider-made but Google-dictated device, sort of like how Microsoft’s first Zune was actually made by Toshiba — and in the case of the Google Phone, there are a couple options for the possible manufacturer. The obvious choice is HTC, who’s been the major hardware manufacturer of Android devices, but TechCrunch hears that the source of the hardware will be Korean, not Taiwanese, which likely points to either Samsung or LG.

Samsung has a long-standing relationship with Apple, supplying tons of parts for the iPhone, so maybe LG would step up to the plate and develop this phantom device. LG’s no stranger to Android, but has been a minor player up to this point — maybe they’ve been working on this mysterious Google Phone in the meantime, which is supposedly aiming for an early 2010 release.

[TechCrunch]

All the apps that count will be ported to every one of them. It’s a completely different situation from the PC market, where software’s built to run on a Windows or a Mac. Mobile apps require very little development, so it’s much easier to bring them onto every platform.

Ozzie, Ozzie, Ozzie. You’re just jealous that the Windows Mobile Market doesn’t have enough fart apps, aren’t you?[Venture Beat via Brian X. Chen]

TouchWiz is the first custom Android interface that’s worse than the standard one, and shows what kind of horrible things emerge when Samsung’s interface designers are left unchecked. Here’s how I think the design process went, roughly: The designers dropped a bunch of acid, stared at old Atari games while binging on Taco Bell, then proceeded to shit all over the phone for hours and hours.

If it’s not inherently ugly, like text input screens with awful ’80s neon orange and blue, it’s gratuitous and redundant, like the 3D app cube. Or an entirely separate menu of Samsung icons for apps. And some things, like moving the slide-out menu to the left instead of its traditional place on the bottom, actually work against the way you use the phone — the menu gets in the way now, since I’d often bring it out by accident while changing between desktops. It’s just… terrible. Worse, Home Switcher, an app that reverts phones back to the stock Android home screen, can’t erase Samsung’s disgusting mojo. The Behold II would be 10x better with a vanilla build of Android 1.6.

Even the phone hardware is a mess. The front of the phone is an orgy of buttons: seven, to be precise, not including a d-pad, with a dedicated button for the app cube. The lock key isn’t just on the side but it’s kind of hidden, flush against the bezel. The USB port is weirdly shoved on top. And, uh, what the hell is up with the back plate?

Two things are good about the Behold II — Samsung’s custom camera setup comes straight out of their point-and-shoot cameras, and is packed with features, like extensive manual controls and burst shooting, and it’s very fast unlike the rest of the phone. The other is the AMOLED display which is nice, though marred by the same kind of bluish tint as Samsung’s other AMOLED Android phone, the Moment.

Take a good long look at the Behold II though: It’s a warning to other developers what not to do, and a scary look at one dark possible future for Android, in its infinite permutations. Not just deep fragmentation of the platform, but customised crimes against humanity, perpetrated in the name of Android. It makes me want to cry, except that my brain’s too mushy to make my eyes work.

The $US100 Pixi made sense once upon a time, when the Pre was $US200 and Palm needed a phone for the masses — like the Centro, oncer upon a time — to establish the webOS as a real platform. A leaner, cheaper version of the Pre was a good idea. Now, you can get the real thing — faster, stronger, screenier — for under a hundred US bucks. The Pixi’s existential crisis is not insignificant.

What’s Neutered vs. the Pre

• Slower processor (using an older ARM11 architecture vs. faster ARM Cortex A8)
• Smaller and squintier 400×320, 2.63-inch, 18-bit colour screen (vs. 480×320, 3.1-inch, 24-bit colour)
• 2-megapixel camera (vs. 3-megapixel)
• No Wi-Fi

Pixi Perfect Design (Just About)

There has not been a candybar phone more perfectly designed and executed than the Pixi. It’s exactly the size and shape a phone that aspires to be small should be. It’s a Hot Pocket sliced in half, but flat and glossy on top and round and rubbery on the bottom. It almost feels fake, like a concept that you hope is a real phone but isn’t, except that in this case, it really is. It fits inside of an iPhone, if you wanna get more literal.

Plastic, slightly sticky Rice Krispie keys, arranged in four rows form a keyboard so electrifyingly good it’s thrilling, like finding an actually sweet wind-up toy in your cereal box (Rice
Bubbles, of course) every time you type. The keys are tiny, but have a deceptive amount of rise, so your fat thumbs can feel out individual nubs, which pop in this remarkably satisfying way when you click down. The size-to-goodness ratio might just be the best on any keyboard I’ve used. If there’s any reason to pick the Pixi over the Pre, it’s if you type a shocking amount on your phone, because the Pixi’s is better by like an order of magnitude. Or eleventy.

Between the screen, with its shaved corners, and the keyboard is a stretch of empty space. Until you run your finger along it, and a spark — a line of light, really — emerges. The LED strip, invisible when it’s not indicating something, replaces the ball on the Pre, which I always thought was a weird little speed bump when you stroked the gesture area, anyway. It’s kind of beautiful, the stark aesthetic of it emotionally tinged with sci-fi imagery, from Gort to Cyclop’s visor.

Two things are wrong. The screen lock button on the top left is a little too in touch with the overall robustness of the phone, so it’s hard to push and doesn’t provide enough feedback. On the opposite end of that spectrum, the trap door covering the micro USB port feels flimsy and aggravatingly snaps shut, making plugging in a USB cable a struggle worthy of a Homerian epic every single time.

Tinkerbell Would Be Pissed

The Pixi is slow.

Achingly.

Maddeningly.

Ripyourhairoutandsmashitagainstthewallingly.

It lags, it hangs, it stutters, it freezes. A lot. A simple fact: Multitasking isn’t better than unitasking when it takes longer to get shit done. An example: I wanted to take a picture while I had the browser and and App Catalog open. Simple. The camera froze spectacularly, rendering the entire phone completely unusable for well over 30 seconds — whenever I tried to flick the camera card away (cards are apps), it would shoot halfway off the screen, then appear back in its place. Somewhere between 30-45 seconds later, it regained composure. That’s with just three core apps open, and no active syncing happening in the background.

True, I could sometimes have up to four apps running without problems, at least for a minute or so, before things starting getting cludgy. But it hangs even with just a single app running sometimes. (Just try opening a website.) And every time you open an app, there’s a solid expanse of time that elapses that you can feel, and it gets old real quick. Maybe webOS is just more transparent about load times than the iPhone, which masks them with title screens, but the whole experience of using this phone is like swimming through very pretty jelly, with one arm, wearing a cast-iron suit, or something like that.

Screen, Camera and Other Hardware

You get used to the smaller screen and it’s 80 fewer pixels, mostly. It’s cramped, but you’ll only be directly, painfully cognisant of it from time to time, like when you’re reading some text outside of Palm’s own apps or navigating web pages. It’s not a crappy screen, but it’s not exceptionally bright or vibrant, either. The touch accuracy seemed less spot-on than the Pre too, though that could’ve been the effect of smaller targets because of the tinier screen, like the drop-down menu for apps in the top left corner, which is just a sliver on the Pixi.

The Pixi’s 2-megapixel camera is unimpressive. The comparison shot above was taken with a 2-megapixel iPhone 3G in the exact same lighting and place, snapped within 10 seconds of each other. The camera app, when it’s not freezing up, is quick to shoot once you press the button though, which is definitely something.

And I’ll just say it: No Wi-Fi sucks, since there are lots of place in NY where even Sprint’s 3G can’t penetrate.

Software

The Pixi comes with webOS 1.3.1. Palm’s definitely tuned things up since webOS originally shipped in June with stuff like more support for Yahoo services, the ability to buy songs over 3G, performance improvements and other interface sprucing up, but it’s not a radically difference experience than the one Chen documented exhaustively here. (In other words, read that for the software review, since it’s basically the same, just much slooooower on the Pixi.)

What’s different now is that there are over 300 apps in the App Catalog, and Palm’s dumping fresh ones in every week. So the app situation is greatly improved. The problem is that it’s still behind the rest of the pack — iPhone, Android and BlackBerry — and being fourth-place development priority for cross-platform developers with limited resources it not a great place to be, so Palm’s got a rough road here. Oh, one interesting point, since this is supposed to be the smartphone OS for multitasking, is that while an app is downloading from the App Catalog, you can’t browse for other apps — if you leave the download page, it cancels. So I hope you’ve got good Sprint reception in your house.

The other thing that’s changed is iTunes. Palm’s former official way to sync your media to your phone is broken. Irrevocably. Even if Palm does restore iTunes syncing with its dirty hack (no really, it is a dirty hack, impersonating an iPod with a false USB ID) the entire model is screwed. Putting people buying your phone in the middle of a stupid pissing match that you’re destined to lose isn’t cool. In the meantime, Palm’s official solution is for people to sideload or use third-party apps like doubleTwist.

Don’t Buy It

The Pixi is screwed. It’s totally cannibalised by the technologically superior Pre, which you can find for under $US100. Even if you can get the Pixi for $US30, it’s worth trading up to the Pre for $US40 or $US50 more if you’re absolutely wedded to the idea of a webOS phone, simply for the speed and screen. Mostly the speed, since the Pixi is brain damaged, three-legged dog slow, as nice as the hardware is on the outside.

Not to mention, for a hundred dollars, there are phones that just offer better experiences and aren’t in the same awkward position Palm is in the smartphone fight. I’m talking of course, about the Droid Eris, Android’s $US99 darling on Verizon. The entire reason to buy the Pixi — a value proposition — has completely evaporated. And I almost feel bad about that. Almost.

Incredible keyboard (for the size)

Awesome design and build

webOS is nice

Camera sucks

Palm’s dumb iTunes fight

Slow

I mean, slooooooooooooooooooooow

Word is that Apple has added a new layer to the approval process, called a static analysis tool. This particular static analysis tool is intended to scan for the use of private APIs in submitted apps, and flag them if it finds any. For quite a few people, evidently, this means that apps that were kosher a while ago are, with their next updates, very suddenly not. Time for a FORENSIC RECAP! From Craig Hockenberry, developer of Twitterrific, and Guy Who Noticed This Early:

To which John Gruber (Daring Fireball) knowingly responded:

So yeah, what the hell does any of this mean?

Apple gives developers a bunch of public application programming interfaces (APIs) which are essentially documented, permitted tools or calls that they can invoke in their apps. For example, Apple has an API for recording microphone input, which developers can use, and which Apple, by publishing them, has basically promised to keep intact and working. Private APIs are calls and features that only Apple uses, and which they don’t really tell developers about. There could be a few reasons for this: either they specifically don’t want developers to use them, for security or consistency reasons, or they’re not finished and subject to change, which means that for devs to use them would be risky — their apps could just break with the next system update, since these private APIs are, in effect, volatile. Remember all those early jailbreak apps, before the App Store was open? Those were built using entirely private APIs, many of which became public later. Anyway!

There’s always been an official ban on the use of private APIs, but Apple hadn’t really been enforcing it to date, partly because it’s just hard to tell sometimes — unless the private API is used in a terribly obvious way, finding them is a matter of taking to apps with a fine-toothed code comb, which the current crew obviously can’t do, and which machines — as in, software —would be good at. With these new static analysis tools, Apple has created a machine filter for apps that breach this rule.

Given that actually App Store policy hasn’t changed, this shouldn’t amount to anything more than better rule enforcement for app devs. Shouldn’t. Introducing something automated like this, even if it’s a “serious tool, not simplistic” as Gruber claims, is bound to end in tears.

Congratulations, developers! Your next appeal against app rejection will be to a piece of software, which has no capacity to feel your pain. Devs: Let us know if you’ve run up against this thing in the comments and if it was fair. [Twitt-ah]