After spending a week using the G1, I can say it's a good start, and a clear indication of good Android developments to come. But the phone itself has some serious problems with accessibility and usability, issues that no number of third-party apps are going to be able to solve. Here's what I loved and hated about the T-Mobile G1.

The Hardware
Body: The body was made by HTC, a Taiwanese company that makes Windows Mobile devices for Motorola, Palm and its own line. This phone is built just like those. The back is classic matted and grip-friendly HTC. The swivel-flip feels almost exactly like earlier HTC phones, only it extends out and then back in again, revealing the keyboard underneath. This motion gives a satisfying snap when opened, though it might be too loud in a quiet office.

Keyboard: It's got numerous problems. First, it's set so that the raised section on the right, with scroller ball and home and menu keys, is always in your way when you're trying to type. This is annoying, even after you figure out how to work around it. The individual keys aren't raised high enough over the body for easy touch typing, though at least the keyboard is backlit, in case you're texting in the dark. The space and backspace key are tinier than we'd like. And it's even more awkward than normal to type while charging the phone, because the miniUSB cable is in the way.

Buttons: There are five face buttons on the device—call, home, back, power/end and menu—and they're all fairly straightforward. Hit home to bring you back to the home screen, menu to bring up a popup menu in your current app, and power/end button to lock your phone or hang up your call. That last part takes the most getting used to, since you're naturally going to want to use the red power button to quit apps or end tasks, but all that does is lock your phone.

Trackball: It feels great, better than on the BlackBerry Pearl, and it clicks down solidly. Still, switching between the trackball and the touchscreen can get awkward.

Screen: The touchscreen is bright, renders text clearly and is, on the whole, pretty great. It uses capacitive touch, like the iPhone, so you use your fingertip, not a stylus, to poke around. There are cases when screen presses don't register properly—they're not too often, but often enough to be noticeable.

Battery: A full charge lasts about a day, mainly because push Gmail grabs the internet every time the account receives an email, and mine receives plenty. Couple that with 3G data browsing and app usage—which you're most likely going to be doing a lot of—and you'll need to get used to a mid-day charge at work. Thankfully charging from near empty to near full takes only about two hours.

Wi-Fi: The Wi-Fi range seems slightly to be on par with comparable smartphones (HTC's Windows Mobile phones, iPhone), showing just about as many Wi-Fi hotspots in my house as the other ones did.

3G: I got noticeably decent browsing speeds, with an actual test registering 433kbps. This, of course, is only the case if your city has 3G access at all, since T-Mobile's only just starting to roll out their network.

Camera: It's passable and on par with previous HTC efforts. It does have autofocus, but other than that there's nothing spectacular with the G1's camera.

GPS: GPS is actually off by default, which produces a very inaccurate location when you try and find yourself on Google Maps. You'll have to switch this on manually.

Other Issues: It's hard to fathom why HTC left out a 3.5mm headphone jack in 2008, same for USB mass storage mode for Windows or Mac. Really? You have to pop out that microSD card and use a card reader every time you want to load a ringtone or a song or a photo or a video? Seriously? Apologies, there actually IS a USB mass storage mode, but you have to use the bundled HTC proprietary mini USB cable. Any old mini USB cable won't do! But yes, it's possible. Also, when the screen is flipped open, it's tilted down about three degrees—really annoying to certain people who like clean lines.

Operating System and Usability
Calling: Making phone calls on this thing works well. Call quality is good, but the screen annoyingly times out after about 10 seconds. If you want to power on the screen again, you have to hit the menu key or the "call" key, which takes you to the dialpad. It may just be that we punch in our credit card numbers or find contacts during a call more often than most people, but always having to bring up the screen again is a pain. And pressing the power/end button, which you'd think would power up the screen, actually just hangs up the call. Annoying. But as for the actually making calls part? No complaints from us.

Texting: Texts are arranged per contact in threads, and works well enough since texting is so simple. No cockups here.

Stability: The one word I'd use to describe the Android operating system is "solid". It's been my main device for a week, and I've yet to see the entire OS hang or freeze (haven't had to reboot yet). Individual apps have crashed or frozen, but Android handles this spectacularly well by using the PC paradigm where you can choose to Force Quit a frozen app or wait for it to unstick itself. This way, very little can take down the entire phone under everyday use. (Buggy hardcore apps that snake deep into core functions could probably succeed.)

Background Apps: Multitasking is one thing Android does really well. Apps can run in the background, receiving data and continuing to "exist," even though you don't see them. The OS handles memory management for you invisibly, giving processes a lower CPU priority and taking away their RAM when other programs need it. For now, examples are simple, like opening a browser, then a bunch of other apps, then returning to the browser. You can use four or five apps before before the browser has to re-fetch data on the web page. Presumably, programmers will soon make more impressive use of the background processing power.

Window Shade: Google's most unique multitasking helper is the notification window shade, which serves as an infodump of all incoming emails, messages, IMs and missed calls. Tapping a notification will take you to its corresponding app. No matter what app you're in, the shade drops smoothly into place when you pull it down, dragging your finger from the top. (Just opposite the window shade is the pull-up app menu. If you run out of room on your three desktop screens, you'll be visiting here for lesser used programs.)

Long Clicks: One convention that's used often—but not consistently—is the long press. Long presses are a mix between right clicking and playing the lottery. Hold down an area of the screen—you may see a menu pop up or you may get absolutely nothing. Long click on the main screen and it asks you which app shortcut you want to move to your desktop. Long click on the text message screen and you'll be prompted to delete or view a thread. Long click on Google Maps or a page in the browser, however, and nothing happens.

Interface: As we have observed, the UI suffers from general usability issues such as inconsistent actions or surprisingly unclickable regions like the browser's URL bar or the home screen's clock. But when you use it, you realize it is kinda pretty. Like the window shade, many of the transparencies, transitions, fade-ins, fade-outs, popups and other UI elements are slick, and definitely win out in aesthetics over smartphones like Windows Mobile. Compared to the iPhone, it still loses, but this comes down to a lack of multitouch capability—on the G1, for instance, you zoom by clicking + and - magnifier buttons. Like I said, it's definitely a solid OS, but it also needs some real work by some UI experts to make it easier to pick up and play with.

Apps
Contacts: Phone contacts sync nicely with Google's Gmail contacts—great if you use Gmail, and an extra place to backup your contacts if you don't. You can even scroll through them fast by dragging a bar on the right. The problem though is that the quick-scroll dragger is hyper-sensitive, and holding your finger still in one place can make the phone jitter between letters. Each contact has a default phone number displayed under his name—when you tap a contact it feels like you're dialing his number, even though you're just pulling up details.

Mail: There are actually two mail programs on the G1: Mail and Gmail. Mail lets you manage five accounts, while Gmail makes you tie your phone to just one account. But Gmail is one of the best apps on the phone, giving you 90% of the desktop features you use on a day-to-day basis. Archiving, labeling, reporting spam, deleting and starring are super easy and sync to webmail almost instantly. The best part of this Gmail implementation is that it's push the only push Gmail on any mobile device (Helio's phones also have it). T-Mobile failed to mention its cool keyboard shortcuts—I had to fiddle to figure out that you can hit "r" for reply or "a" to reply all. (Surely there are more.) A dumb flaw is that it won't auto-complete names when you start with someone's last name. I have to sort through 10 Brians to find Lam's address, when I should be able to just type Lam and have this be smart enough to figure out who I mean.

Marketplace: The Marketplace is divided into Games and Applications, with sub-categories such as Lifestyle, Productivity, Shopping and Tools. Downloading and installing apps are pretty much 1-click, like the iPhone App Store, and most apps launch just fine. However, since most developers don't have an actual Android phone to test their apps on, a lot of programs will be sluggish or even crash-prone in the first few weeks. Expect this to be fixed soon.

IM: The IM app is a very good client that supports AIM, Google Talk, Windows Live (MSN), and Yahoo. It's intuitive, works well with the keyboard and even offers background notification—unlike iPhone—so you can switch to other apps but still get incoming messages delivered to you via the top status bar.

Browser: The G1 browser, like Chrome on the desktop, is based on WebKit, the open source browser engine that also powers Safari and Mobile Safari. This means it's pretty damn good. That said, the lack of multitouch gestures in Android's version makes zooming a pain. It doesn't have Flash support (YouTube gets forwarded to the YouTube app) and it doesn't auto-zoom to maximize the column you want to read in your display. It can, however, remember your password for logins, like a desktop browser does.

Google Maps: Gmaps has most of what you'll find in the desktop version, including Satellite, Traffic and Street View. Once you turn on GPS, the phone's fairly decent at locating where you are even indoors, and Compass View is a gimmick that works sometimes and doesn't work other times—but then again, spinning around like an idiot makes you look like an idiot all the time.

Music Player: It's no iPod, but the G1's built-in music player gets the job done decently. It fits in fairly well with the rest of the Android experience, but we're definitely looking at third-party apps like TuneWiki to pick up the slack here. That's not to say the Music app is bad—it's perfectly fine. It's just not great.

Third-Party Apps: Some of the more promising apps like Tunes Remote, TuneWiki and Video Player aren't as fleshed out and stable as we like. Tunes Remote lags and crashes a lot, TuneWiki can't find our music and Video Player only supports a handful of codecs. We expect these all to be fixed soon. Other apps like AccuWeather, Barcode Scanner and Pac-Man work just fine despite being developed on the Android emulator. We're looking forward to good things here.

Verdict
The G1 phone and the Android operating system are not finished products. There are only three working Google Apps here—Gmail, Maps and Calendar—while Google Docs, Google News, Google Reader, Google Shopping, Google Images, Google Video, Blogger and Picasa are nowhere to be found. What's the deal?

We have high hopes for third-party coders to fill in gaps Google intentionally or unintentionally left in this OS. There's already a video player, and we're sure VLC will try and port some kind of version over. But your question is not whether the phone will be great down the line, it's whether or not it's good enough for you to buy it now.

The answer depends most on who you are. Despite all the UI quirks and bad design decisions, it's still better than other smartphone OSes out there. It's not perfect, but for people who like tinkering, its cons are outweighed by its pros such as Gmail and the Marketplace. Hopefully Android updates and more ports of Google apps will augment not just future phones but this one too. This isn't something you're going to give your mom for Christmas, but if you're an adventuresome gadget guy with some money to spend ($179) on a totally new, pretty exciting venture, then why not?

1. It's open! The single best thing about Android is that's a modern mobile phone OS that's also almost completely open, unlike some other locked down mobile OSes. (There are a few restrictions in accessing the hardware for security reasons.) It's based on Linux, and once Google has released Android, most of it will be totally open source, so it'll be incredibly easy to dive into its guts and mess around, which will help build a robust developer community, along with all of the other benefits of using open software. Most of its other awesome traits grow out of its openness, actually.

2. We'll keep the Steve references to a minimum here, but Android will accelerate the process that the iPhone helped kickstart last year—the gradual devolution of carriers to open, dumb pipes. Before, carriers controlled every single facet of what a phone could and could not do. They still do to an extent, and it's not completely "anything goes" on the iPhone and Android, but together they have and will make the mobile landscape change far more rapidly than before. As Opera CEO Jon von Tetzchner told me a few days ago, just a few years ago, carriers thought that the idea of full internet access was ridiculous, not to mention dangerous, threatening the obscenely lucrative business they have set up around minuscule bits of data like text messages, crappy "web portals" and ringtones. Android phones will be constantly connected and totally revolve around the internet, incessantly sucking down ever cheaper data—a carrier's worst nightmare not so long ago.

3. You'll have tons of hardware options. Android's designed to be versatile, so lots of manufacturers will be putting it on lots of phones—ones with QWERTY keyboards, touchscreens, T9, outrageously spec'd out phones, as well as ones that are kinda crappy in the spec department, actually. But this also provides a common platform for developers, making it easy to put their apps on millions of phones. It's the benefit of any OS that runs on a lot of hardware—like Windows or Linux, etc. Of course, this is also the Windows Mobile argument against all of the other proprietary OSes like Palm and BlackBerry.

4. There's even more potential for amazing apps than the iPhone, because developers are almost completely unencumbered by arbitrary rules and restrictions. So awesome apps like Podcaster or Instictiv Shuffle won't be mercilessly killed for not fitting into a tightly controlled framework or navigating a byzantine approval process.

5. It'll have the best Google apps experience of any mobile device, and play super nicely with Gmail, Gcal, Maps and everything else Google puts out. Or at least it damn well better, since you know, it's Google's baby. Simple, direct syncing with Gcal is tops in our list, since doing it on the iPhone requires sacrificing a goat while chanting from a book covered in the skin of baby unicorns.

Bonus reason: Not an iPhone. And our software geek sister Lifehacker has some more too.

1. Google can see into your soul. If you've ever been wary about how much Google knows about you, how are you going to feel when they're all over your cellphone? While a lot of the reason Android came to be was just to get people really using the internet on their phone (because when people use the internet, they use Google), we won't be surprised to see contextualized local ads, kind of like the sidebar ones you see offering you a date from hot local girls in Brooklyn or whatever hovel you're holed up in. But this will be hot girls just around the corner, since the phone will know where you're at.

2. It's not on the US's two biggest carriers, AT&T or Verizon. Statistically speaking, you've got one of them. But so far only the two runts of the majors, Sprint and T-Mobile are going to have Android phones. T-Mobile's 3G network is pathetically tiny compared to the other three, and well, Sprint's the only carrier actively losing subscribers, if that tells you anything. It's possible we'll see some Android action on Verizon's mythical open network though.

3. Buuut, carriers still have the right to gimp Android to their liking, precisely because of its Apache licensing. So a Sprint Android phone could have its built-in "store" stocked only with, say, Yahoo! apps—or no store at all. In Verizon's hands, the UI could still look like it fell in a bucket of gaudy red paint.

4. Android is designed to run on a ton of different of hardware—phones with and without touchscreens, with and without QWERTY keyboards, phones with amazing specs, crappy phones, and everything in between. While this is a strong point as mentioned above, it could also be a point of suckiness. That means there won't be a consistent Android experience, and it'll depend heavily on the device you're using. Devs told us that you'll likely see different versions of their apps, so that on weakass phones, you'll have more diluted apps, which might be an issue for people picking up a cheap Android phone expecting to do everything a more expensive one will.

5. Relying too much on developers to fill in features could result in a phone that's not quite totally seamless and consumer-grade across the board. For instance, from what we've seen in the SDK, there's not a built-in, Google-made media player. It's rumored that the excellent TuneWiki will be Android's default player, which is great, but doing this for too many key features could make things a bit bumpy, since you're talking several developers instead of just one.

Bonus reason: It's not an iPhone.

Worse in the long-term is that the analyst's contacts tell him that developers are not exactly snapping up Google's SDK—they're too busy developing for the ten million other OSes out there, like Windows Mobile, Symbian, OS X and BlackBerry. It wouldn't be too surprising, given stuff like infrequent updates to the SDK, unless you're one of the super special developers with access to the privately updated one. The fact that bad news about Android keeps rearing its head isn't a good sign itself, since noise this persistent usually has a bit of signal embedded in it. [Barrons via Electronista]

The reporter noted that “The trio of Google execs also used the opportunity to talk about the inroads the company is making with its own branded mobile phone as a replacement for the iPhone.” [That appears to be a mix-up by the reporter, with Sergey and Larry actually talking about not producing their own phone, according to Reuters's Ken Li's notes in SAI.]

But, TechCrunch does have its own source, who "swears" that the Ammunition Design Group "is designing the Gphone and that it is a seriously beautiful device." They've worked with companies like Palm, Hewlett-Packard, Dell Computer and Logitech in the past (a phone they designed for Sprint is pictured above). Not a confirmation by any means, but TechCrunch usually has decent info—so take that for what it's worth.

The more probable explanation for any sort of design work on a handset for Google is that they could be prototypes for marketing or other promotion—no one can make the call whether it's for production yet. [Tech Crunch and Mediaweek via BGR, Silicon Alley Insider]

That's just one reason it needed Android. Unlike Windows Mobile, which is all tangled up with Windows, Android's totally centered the web—where people naturally go to Google. While Android vs. iPhone is shaping up to be the new Windows vs. Mac (or open vs. closed), the iPhone actually proved the thesis that easy mobile net access is really easy access to Google: Christmas Day, the iPhone, "fewer than 5 percent of all smartphones worldwide, drove more traffic to Google than any other mobile device." By making Android all about net connectivity and giving developers a common platform to develop for hundreds of phone, the bet is that even with tons of third-party apps, it all comes back to Google. The web is the platform as much as the actual code-y bits.

Naturally, handset makers fear losing their brand in the hype, even as Google argues it means they don't have to waste time on the OS, but can concentrate on hardware and their own proprietary apps. (Course, if you're of the mind it's all about software now, then Google's argument is funky bunk. Hardware will matter maybe as much as Dell vs. HP—maybe that's a lot to you, maybe not so much. Besides, has HTC really had an identity in the first place?) Motorola is actually betting big, putting the original team behind the Razr on its Android phones, hoping it'll be a path to newfound glory.

Even if it (or anyone else) succeeds, ultimately they'll still just be a cog in the Android machine. By the same token, even if Android itself bombs out, as long as it forces open access to the internet, Google still stands to rake in the rewards. [Wired]

Assuming that the phone really is in production and will be delayed until 2009, that leaves the question of whether the Android platform as a whole will be delayed, since many other manufacturers besides Google are planning their own phones. Krakow's source doesn't say, which probably means it's still on track. Either that, or horribly, horribly delayed and the guy doesn't want to induce a panic. [The Street]

Update: Google spokesdudes say that the Android platform is a-okay.

We're still on track to announce Android-powered phones this year. Some of our partners are publicly stating that they plan to ship Android phones in the fourth quarter.

It sounds sort of fun but also sort of awkward, as I can't really imagine running around aiming my phone at some stranger on a crowded street. It would also be really annoying to get attacked while you were out on a date or something, having to explain that it was just some guy who was engaging you in your phone's virtual first person shooter. In any case, there's a beta open for this, although with no Android phones out yet I'm not sure how you can participate in it. [Product Page via Ubergizmo]

For better, for worse or for Google, it sent the formerly locked-tight mobile-phone industry into an open-access hysteria. True, Android and the Open Handset Alliance are still (mostly) pretty logos and promises on paper. But for a bunch of talk, it's lead to some very real action, like helping nudge Verizon out of its ugly, red-walled garden. All this gesturing might actually change the industry.

Android aside, we can really give Google props for helping to break the mobile industry via its hard fight for open-access provisions in the 700MHz spectrum auction. The provisions themselves could change the game for all telecoms, and Google doesn't even need to win!

If anyone thought Google was approaching the limits of its growth earlier in the year, Wall Street proved 'em wrong when gPhone hype pushed its stock past the $700-per-share mark, making it the fifth most valuable company in America. It's comfortably resting now at the hype-free level of about $670, over $550 more than the initial price of around $120. (At the time, many investors say that was too high.)

Google kept up the "don't be evil" motto by pumping lots of money into green energy and green research. Maybe this is because they believe in a better tomorrow, but maybe it's just to power their Matrix-like server farms without resorting to turning runaway Microsoft employees into living batteries.

On the Google application front, its office suite is still no Office-killer, and we still have some kinks we wish could be worked out of the Reader, but Google did make our lives easier in a couple ways, like by dropping the invite requirement to Gmail, adding IMAP support and getting even more iPhone-friendly. Yay. And, just the other day, it threw down with Wikipedia by launching its latest collect-all-knowledge component, Knol, which'll pay contributors—with ads. (Surprise.)

Owning YouTube kinda caught up to Google, and we're kinda sore on that front—it got sued by Viacom, bitched at by other media companies and then had to implement a content-filtering system. But possibly even more egregious was its addition of overlay ads.

Now for even less palatable stuff: Google knows pretty much everything about you and everyone you've ever loved and that probably won't ever change, even if it is trying to be a little less creepy. For instance, Google will still have all the info you think you're erasing from Ask.com. And, it still plays nice with ethically dubious—to be generous—Chinese censorship regulations to ensure its grip on a chunk of the massive and growing market over there, which doesn't exactly line up with the "don't be evil" screed. Even Sergey thinks so.

And the ads. It's much pretty much a given the catch on any Google goodness is that there'll be some sort of advertising wedged in there—true, nothing worth paying for is totally free, and Google doesn't appear game for charging monthly fees for every little service, like Microsoft and Yahoo! now do. But Google going mobile means ads becoming an increasingly unavoidable part of your mobile life— AdSense for Mobile is all fired up. It seems the lovely mobile industry freedom Google's fighting for isn't free. It costs a buck-oh-five—or a contextual spot.

Final grade: B But we think this grade is slippery: next year, Google will either be A+ or D. What Google is doing is either a massive greed-motivated build-up or a series of initiatives intended to change the tech business (and our world) for the better. In two semesters, we think we'll know if either is true—or if both are.

back at its headquarters, Google is already operating an advanced high-speed wireless network under a test license from the FCC, according to people familiar with the matter. The company has erected transmission towers on its campus for the network. Prototype mobile handsets powered by the Android software are currently running on it.

More over, according to the usual "people familiar with the matter," they're grinding out a plan to toss in all $4.6 billion or more all on their own because going in with partners promises complications in bidding setup and outcomes, as well as pissing off company A by going in with company B.

Why even jump into the messy, messy world of network operation when they've already got a hand in the jar with Android? The feel-good reason is that Google wants networks to be more open and competitive to drive more innovation and better, unrestricted services without artificial carrier walls and lockdowns.

The selfish side is that open networks mean carriers can't make it hard to use Google's wares or make Google pay carriers to get at customers. There's also the potential revenue of subscriber fees—hey, open doesn't mean free—and ads across the network, according to at least one analyst group.

Whatever Google decides, we'll know by Dec. 3, the deadline for declaring an intent to bid. And should they win, don't cream your jeans about the wild, woolly West of an open wireless network just yet, since Google—and any other winners—won't get their hands on the network until 2009, when analog broadcasts go dark. [WSJ]

•Programs, built by using Android's SDK, can be submitted January 2nd through March 3rd 2008.
•The top 50 programs earn $25K to continue development.
•Those in the top 50 are entered into a final round where ten $275k prizes and ten $100k prizes will be awarded.
•A second round of the contest starts in the second half of the year, when handsets running Android launch.
•"We believe that developers will be drawn to, and flourish in, the open and collaborative environment offered by the Android platform. The Android Developer Challenge is designed simply to catalyze that reaction."

Pay people decent money to do decent work. Who would have thought of that? [Android]

The only companies that join consortia are the ones who are too stupid or shitty to make a great product on their own. It's like, Hey, we've got forty spazzo companies that can't fuck their way out of a paper bag; let's put them all together and maybe they'll magically become some kind of big bad powerhouse.

There are, of course, numerous cons to the singular "hand of God" approach, but the problems with consortia that FSJ lays out are real. It's hard for everyone to agree, and rarely do all involved push for the "greater good" at their individual expense. It might be different this time around, but with so much at stake, martyrs for the Alliance probably won't exactly be lining up. And oh yeah, there's no phone to wrap our hands around yet. [FSJ via Daring Fireball]

Another niggling issue for the juggernaut pair is Google's probable bid for wireless spectrum in the FCC's upcoming 700Mhz auction. If it winds up turning up its chunk into a mobile network, it'd be pulling a Microsoft by competing and partnering with them simultaneously, which apparently they don't take too kindly to.

Of course, this is on top of all the other reasons they weren't exactly clamoring to jump aboard in the first place. [WSJ]

For starters, Sprint and T-Mobile are big companies but they don't have the momentum or subscribers that Verizon Wireless and AT&T do. T-Mo's the baby with 26 million, though that is part of a global subscriber base that's vastly larger (over 109 million). Sprint has an impressive 54 million subscribers, but is currently in a public panic about waning performance. Both, contending with bigger gorillas, have reason to latch on to the biggest of them all, Google.

There's a second reason for T-Mo and Sprint to be onboard: HTC. HTC has had a good run with both of those carriers, introducing two of the coolest recent products exclusively on them: T-Mobile's Shadow and Sprint's Touch. Verizon and AT&T have relationships with HTC as well, but it's easy to see how HTC would feel comfortable developing an Android product for the two smaller carriers.

Qualcomm's high-level involvement in OHA is actually very surprising, both because it is such a close partner of Verizon, and because it is one of the biggest architects of the closed application-and-service environment that you see in US carriers. Brew is Qualcomm's baby, a closed Java equivalent that may be easy for developers to work with but comes with strings attached.

Clearly, though, Qualcomm's participation signals something else, that it is time for even the master walled gardener to embrace the open field. At least, that's what I hope it means, because its presence means that a Verizon Android handset can't be too far off. (Verizon today told us that it "shares the goal of more open mobile application development," and that this competitive move on Google's part shows that innovation comes without the need for "legislation nor regulation," so here's hoping.)

The point is, even if you're neither a Sprint or a T-Mobile customer, you should be happy with today's announcement. Google said explicitly that the OHA is open to everyone, and that the absence of partners (for example, AT&T and Verizon) should not be taken as exclusion. But having Sprint and T-Mo ensures equal development time for both networks, GSM/HSDPA side and CDMA/EV-DO. T-Mobile promises to have their HSDPA (3G) network up by early next year, in time for any Android smartphone to take advantage.

When it matters, there will be HTC (and perhaps Motorola, LG and Samsung) phones that work on the networks of all four carriers. It's a sped-up version of the Treo or BlackBerry scenario: once a handset gains a certain level of popularity, even the mega carriers must yield to their demanding customers. When the Treo or BlackBerry first came out, there was no CDMA version so Sprint and Verizon customers got the shaft, and it took years of development, not just a quick port, to give them what they wanted. Thankfully that will not be the case with Android.

The iPhone is different: it was developed for GSM/EDGE only, and with an exclusive deal with the largest US GSM carrier. Although Apple hasn't confirmed the 5-year exclusivity rumor, it could easily remain locked out of everyone else's hands, in spite of unprecedented demand.

Speaking of locked, while we can assume that the early Sprint Android handsets will be partial to their carrier because of the lack of SIM cards with CDMA phones, HTC told us that it will sell unlocked Android phones directly to consumers, so that anyone with an AT&T contract who wants in can get in early.

There's also a difference between Google's and Apple's approach. By creating an open platform, Google is trying to make money not on software or hardware sales, but by creating vast hordes of ad-susceptible phone users. Google can be less selfish about design, and less worried about stumbles on the road to perfection. Google boss Eric Schmidt told us today that they would not be in the business of clamping down on independent development, and from the sound of it, would be encouraging carriers to adopt a hands-off policy toward third-party development.

What Google is saying to carriers is that customers are grown-ups. They can own a PC for years before it's chock-full of viruses, malware and memory hogging crap they don't need. Why not give them access to a world's worth of software—and expose them to the same acknowledged risks—with their phones? iPhone owners should be happy that Google's move simply adds pressure on Apple to open the SDK faster. But the larger question we should ask is this: Why will Android succeed where Symbian, Palm OS and Windows Mobile have failed?

It is possible that Google's muscle, outspoken pledge of openness, ability to bring much of the familiar PC experience to the Android phone, and vast think tank of collaborators who can avoid the mistakes of the smartphones that failed, the carriers that squeezed too hard and lost their grip, and of that famous experiment in gadget lust and contractual obligation that is the iPhone. But then again, it could be no more than the fragmented world of Linux development, without much mainstream splash.

The Dream looks like the iPhone, and is five inches long, three inches wide, touch-sensitive, swivels, and hides a full-sized QWERTY keyboard below. It also switches to landscape mode when it's tilted. On the software side, it's got Google's own YouTube, plus email, text editing, and a virtual machine to ensure all applications work on all the various types of hardware. HTC's even considering developing a real version of this reference design in the same second half of 2008 that they're going to be releasing their Open Handset Platform phones. [Forbes]

9:50: USA Today asks about gPhone name again. Coming? Another way to think about the gPhone, says Andy, is that there will be thousands of Google phones, some you like, some you don't.

9:46: NYTimes asks what kinds of agreements will be in place between carriers and Google on ads, and will they have to share ad revenue with carriers? And will this phone change business models in the phone industry? Google answers: Sharing is likely. Our strategy is to share. Android allows it, but doesn't require it. If anything, it provides additional opportunity.

9:44: Sasha asks if it's possible to create completely locked-down phones. It is, apparently, but Google says that you can do that, according to the license, such a device wouldn't be able to run many apps and would be much less powerful and competitive.

9:43: So if you don't like the phone you get from your carrier, if it's Android, you're free to switch it up? Google: "that's an answer for the carriers."

9:39: How will this phone compete with development communities of other OSes, and will yet another platform fragment the dev community. The difference, Andy says, that this one is open. And it's open source. Eric says that because of this, even competitors might both pick it up, if the platform is good enough.

9:39: Eric says he could picture using iPhone in one hand and Android in the other. Strange.

9:33: Chicago Tribune asks what the diff is between the Gmail apps on phones now, and the Android ones. Eric answers that the gPhone, with a real browser, won't need customized programs and websites, so it'll be easy for devs to support the phone by supporting any desktop browser. Android will be the first to do this broadscale.

9:29: Someone asks, were Microsoft, RIM, and Apple to join the alliance, and what would they say? The question is being dodged but he says the alliance is open and if people want to join they can.

9:29 Financial Times asks if a real gPhone is coming. Eric says that we're not announcing, but if we were to, this would be the platform for it.

9:27: WSJ asks about Google's web applications and ads will work with the new platform. Andy responds that Google's flexible ad platform should be fine, esp since one part of Android is a very robust HTML browser. But you won't see an ad-driven browser on this phone for some time.

9:24: BBC is on. They're asking if HTC is going to use Android exclusively. Peter Chou says that they're going to be using all OSes equally. Moto says that this is important to accelerate their development, but they do have some commitments from other OS devs.

9:23: The Q&A is starting.

9:21 Sergei Brin: Sergei is talking about how Linux helped him start Google, and now he hopes that they can use open source to develop great handsets. Today's mobile devices were more powerful than the servers he was using 10 years ago, so he's excited.

9:19: VP of China Mobile, biggest carrier in the world. An open mobile platform will accelerate China's smartphone adoption, he believes.

9:16: Ed Zander, CEO of Motorola is on the phone. Says nothing about handset development.

9:13 Paul Jacobs, CEO of Qualcomm, is speaking. He's chatting about 3G. Qualcomm has been working on Android for the last year, committing significant resources. They've been working on the 7000 chipsets for ideal multimedia functionality. They're upcoming snapdragon chipset will do better data throughput, 3d, GPS, and mobile TV at better power consumption.

9:11: Peter Chou, CEO of HTC Corp is speaking. He believes that the point of a mobile phone OS is to provide quick access to the content and functionality of a mobile phone. Android can do this, he believes. We plan to release the first mobile phone in the second half of 2008 with others to follow. "Windows Mobile Sucks!" Just kidding, he didn't say that.

9:08: René Obermann, CEO of Deutsche Telekom, parent company of T-Mobile, is speaking. Squishy quotes but no new data...our current plans include the launch of an android handset in 2008 in the EU and US with heavy wireless and web2.0 capabilities.

9:07: Eric is saying that users want better features. And that you couldn't do this with one company. That's why they've started the alliance of 34 companies.
Access to less expensive phones with more services and better UIs. Elliot Speaking.

9:06: We want to start a mobile phone platform that includes new services and features we couldn't imagine today. First mobile phone platform stack, including OS, apps, etc. Most liberal open source license, too. It's important to say that this isn't an announcement of the Googlephone. It's the announcement of a platform for lots of phones.

9:05 Started. Eric Schmidt is here and speaking.

•They're hoping to make a better phone, ultimately. (And sell a ton of ads and services, of course, along the way.)
•Android, an open system for handset dev, is the first joint project and core product of the alliance.
•There are 34 members of the group, including NVIDIA, Intel, Texas Instruments, Synaptics (haptics!), Marvell, Qualcomm (chips), Motorola, Samsung, TMO, Sprint, LG, HTC, KDDI and DOCOMO from Japan and China Mobile Comm. Corp.. Basically, a lot of companies sick of Windows Mobile Slop and other closed up phone systems like the iPhone.
•Who's missing is interesting: Nokia, Sony Ericsson, Blackberry/RIM, Apple, Verizon, and AT&T. Oh, did I forget to mention Microsoft?
•Handsets coming in 2008, second half.
•Nov 12th, the Android early look SDK drops.
•Android built on Linux, made available as open source via the Apache v2 License.
•Companies can dev custom functionality to Android without contributing the source code back to the community.
• HTC's first prototype is the dream. It's the first set of hardware details we've heard of.
•Chen's rounded up Gphone details from a couple of mainstream publications. There's nothing you haven't read above, but a few more quotes.
•Couple of Videos on the Android. One by devs, one by kids.

I'll update this post as more comes. Just got off a conference call, actually. No new details. This is the basic outline of what we have for now.

[All gPhone stories on Giz]

HTC's CEO, who makes lots of Windows Mobile phones for business users, says the resulting phone by Google and its 34 friends will be targeted toward consumers. The SDK for developers will be available by November 12, which will then enable people to hit that 2H 2008 launch date. The software's actually been in development for three years, starting from the Google acquisition of a company called Android Inc. The most important part of this is that the platform will be free for handset manufacturers to load onto their phones, beating out other operating systems like Windows Mobile, Palm, and Symbian in terms of price. [NYTimes - USA Today - CNN]

Final worthwhile bits: Spring is the worldwide ship date, and a weird mention that the "Google Phone...could blow open this model by providing easy access to the Internet at PC-type speeds." Juicy, pertinent info or random hypothetical? We'll know soon. [USA Today]

Helpful chart the WSJ whipped up, no? Other possibilities include easier sharing of music, photos and video with friends across carrier networks, as well as better internet access and utilization—you need it for the ads, after all.

The wild, wonderful world of millions of innovative, useful apps on millions of open devices that we're tantalized with for three-quarters of the piece gets doused with a nice, healthy dose of cold water at the end, however, by the hard reality that developers will still have to conform to carriers' rules "if they want them to work seamlessly on the most popular networks." So we're probably looking at baby steps in the industry rather than full-on ¡revolución! But hey, it's a start. [WSJ]

The NYT is confident referring to the gPhone as a "software" package, rumors of which we have heard countless times. According to the extensive report, the gPhone will be made available by the middle of 2008 and is being produced by Google's partners (HTC, among others?) Google's contribution will be an open OS that shall be distributed freely and will earn revenue via advertising links. Interestingly, the model of bundling free software has been pretty successful in the past, as exemplified by Internet Explorer.

It seems Mr. Rubin is the right man for the project, as one of the founders of Danger Inc., the company behind the excellent Sidekick smartphone series, he seems well qualified to deal in the competitive market. That is not the only experience he is bringing to the game either; back in 1990 Mr. Rubin worked with Apple on a project called Magic Cap, which focused on a groundbreaking platform for portable devices. The venture fell through, as it was apparently too far ahead of its time. (Is this how the gPhone and the iPhone inadvertently cross development paths?) Mr. Rubin went it alone, he blew his entire savings on producing his vision of a mobile platform that would be open to all software developers. The name of Mr. Rubin's project? Yup, you guessed it—Android.

Whatever the gPhone is officially confirmed to be—phone/software suite or both—the potential impact Google will make to the way we communicate will be definitively game changing, if successful. With all this information seeping through, it is only a matter of time before the covers are dropped on the big G's big plans. For now, the wait goes on. [Valleywag]

As we mentioned before, CNet affirms that the OS should be open-source and Linux based, but we shouldn't expect to see it in stores soon. The number of companies involved (over 30) will make it difficult to move quickly on this. I'm just glad that this thing is moving out from the shadowy depths of rumorville. [Cnet via Gadget Lab]

What has been amazing to watch is that investors keep bidding up the share price and the company responds by earning so much money as to bring each new dream back into the not-preposterous range.

According to the story:

Richard Doherty, research director at The Envisioneering Group, says Google is driving toward getting device makers to produce basic handsets equipped with a "Gbutton" that takes users right to a Google texting screen. The goal: Get GPhone texters in the habit of querying its vast databases for directions, phone numbers, movie times or restaurant tips and woo advertisers to sponsor GPhone-delivered answers, Doherty says.

While we're excited to see what the Gbutton is all about, we're a little afraid of how involved the carriers might be in GPhone rollout. Of course, when Google does sign a deal with a carrier, it will be a little hard to tell who is selling their soul to whom. [USA Today/Yahoo]