Reuters: Google phone with T-Mobile contract in Jan | Hacker News Reader

Reuters: Google phone with T-Mobile contract in Jan

reuters.com

130 pointsippisl108 comments

I'm not sure I approve of changing what a HN story links to after there is already discussion related to the old link.

The original article is here: http://www.androidguys.com/2009/12/14/reuters-nexus-one-avai...

Agreed. I'd much rather have a separate discussion now that the news is "official".

It's not that the news is more official, it's that the original link was speculating about a $99 unsubsidized price, which is pretty exceptional for a smartphone. That's not in the article at all.

Really? I kind of like a thread full of 'bah, it'll never happen' nay-sayers being immediately contradicted by the link they're commenting on.

...according to a rumor posted in a comment on another blog.

I'd say there's basically no chance of this being accurate. Otherwise, I'd buy a case of phones and resell them for parts. They surely cost at least $300 to produce just in parts.

I'd say there's basically no chance of this being accurate. Otherwise, I'd buy a case of phones and resell them for parts. They surely cost at least $300 to produce just in parts.

Hmm, the iPhone 3GS teardown is only supposed to be ~ $175. I haven't seen any reason to suspect that the Nexus One is more expensive hardware.

Not to mention that whoever posted this clearly doesn't understand Google Voice (hint: it's not a VoIP endpoint)

Sure it is - you just use POTS to access it right now. I wouldn't get too caught up in product details, when there is no technical reason to prevent it.

And Google just bought Gizmo, which is a Voip endpoint. It already had some Google Voice integration, presumably it'll have more when they relaunch.

Sounds pretty believable and reasonable to me (except for launch date).

Do you have further info on why the phone will cost about $300 to build? In comparison, the iPhone 3GS allegedly has BOM of $178. http://www.isuppli.com/News/Pages/iPhone-3G-S-Carries-178-96...

Curious here: who would buy the "parts"? It'll cost you a lot just to disassemble the phone.

I don't put a ton of faith in iSuppli, but regardless my point is that it is unlikely anyone would sell a device at a loss--and $99 would surely be a loss, yes?--without any obvious way to recoup that money.

Making an open source mobile OS isn't a great way to make money either. Google sees mobile as critical to long-term success and is spending money to make big inroads. If they spend a billion dollars and get 20,000,000 new people as Google mobile users that may be a coup in the future. Hell, they spent more than that for YouTube.

As I mentioned in a previous comment, they know they'll make tremendous amounts of money from mobile search, and app sales (if/when they pick up for android).

Over a year, everyone with android will probably have supplied Google with hundreds of dollars of mobile search revenue.

If the Gphone sells for a low price, and in very high quantities, they will probably lose for every one sold, but will make much more per user than the $100-200 lost on production.

It also means more sticky users in the long run, popularizing android and turning more people towards google. More apps would be developed, other android phones will presumably sell more (unless the contrary happens, in which case it is the Gphone that sells, which means even more money for Google).

I wonder, how many thousands of mobile ads would I have to click on to generate $100 for Google?

You don't have to click ads, they pay by impression too.

true, you don't have to click them: every website you visit on android that has Google ads on it, or YouTube (more ads) would generate revenue to Google just by visiting it. Plus, some ads run for over $20+ per click, out of which Google surely makes a nice chunk out of, which means that, by these parameters combined, you don't really have to click many ads to generate $100 for Google.

I reckon that over the course of a year (the average length of time at which users change their phones), you'd generate Google more (if not much more) than $100 in explicit and implicit revenue.

But every website you visit on iPhone that has Google ads on it, or YouTube generates the same amount of revenue to Google. Right?

I mean, it's one thing if their big competitor here was Microsoft or, in other words, Bing. But iPhone already uses Google by default.

True, but how does that change the situation? More phones means more revenue and more users. That's probably what they're looking at when thinking about subsidizing the handset.

Not sure about this, but I don't think Google takes a cut from Android Market app sales (unless you count the cut that Google Checkout gets, since a lot of apps are sold through that service). Can somebody who's got an Android app in the market weigh in on this?

I assume Google makes a cut from those too. It wouldn't be hard to imagine that those who consume many apps over a period of time might also contribute towards covering for the subsidy.

How much are you willing to pay to put the brakes on your competitor and to catch up fast?

$100 per customer now to prevent them from locking into the iPhone isn't quite so ludicrous if you also have a long-term strategy to get that money back (and then some) using the advertising in the shipped google services.

Besides, you have a war chest for just such things... but instead of spending it on acquisitions you can use it to help buy customers, they become happy with you too.

If this is Google's strategy, then they are becoming more like Microsoft every day.

As long as I have the ability to completely remove Google's version of Android, and put on a custom version of my own, I'm not that concerned with the rest.

I'm not really familiar with the subject, but if Google uses their profit from online advertising to get people to use their phones (by selling them cheaper than "possible"), wouldn't that be the same as Microsoft putting ie in Windows? So what do the anti-competition laws say about this?

Google doesn't have a monopoly in mobile phones, so the anti-competition laws probably wouldn't affect them much.

Actually, the potential issue would be whether they are using profits from one monopoly (search) to undercut competition in another competitive market (mobile phones). A guest post at TechCrunch actually had a good overview of the ways Google might face antitrust action:

http://www.techcrunch.com/2009/03/01/what-an-antitrust-case-...

No - as far as I understand it, the issue with IE wasn't that Microsoft used profits from Windows to give away IE, it's that they used a monopoly product (Windows) to give an unfair advantage to IE (out-of-the-box distribution not accessible to their competitors).

2007 average monthly revenue per user at google was $53. add this to the fact the the mobile phone is very important in many areas(for example: location , mobile-web , mobile-tv) , $100 subsidization don't seem like a very large investment.

Where did you get that monthly revenue figure?

"Overall revenue was $5.94 billion for the period ending September 30, an increase of 7 percent compared to the third quarter of 2008."

Just a quick math. Assuming there are about 300mil frequent users of google (i.e more than the occasional search) in the western world. If 4.5 billions came from them, that would be about $15 per user per quarter, so about $60 per year.

Assuming that the google phone user is going to be one of these heavy users, if mobile user are worth $60 per year, then a $100 subsidy doesn't sound that bad, but still, very far fetched.

People would be buying them in droves, and using them in china/india/whatever developing country you think, and there is no way google would be making $60 per year from them.

What I don't believe at all is the 29/mo data plan. There is no way att or t-mobile will put up with this. Remember, the device has to connect to something, otherwise it is just an ipod with wifi.

The other alternative is wimax. Their network would be perfect for something like this, but they are not nation wide yet.

My bet is that android 2.1 will have wimax support.

Adding unlimited data to a voice plan on T-Mobile currently costs $30, so not totally out of the question. AT&T currently charges quite a bit more.

Mobile WiMAX is still a long ways off.

The difference is that you would only have the data plan and your voice calls would go over the data network, and not the call network (IIRC, they use different bands to transmit/receive). So these people would immediately be 'high bandwidth users.'

That sounds very high without a solid source.

For that to be an annual number, however, sounds pretty reasonable, especially given Google's asking price for an annual ad-free Google Apps subscription: $50.

If you think about it Google's entire business model operates this way. How much do you pay for GMail, YouTube, Search, etc? $0 yet Google has to spend $X per user to provide and maintain the service, pay the bandwidth bills, pay the engineers, etc. They're recouping all that money and a ridiculous amount of profit on advertising. I don't think it's too far out of the realm of possibility they would extend this model to hardware in selected markets where they are pretty late to the game (as a platform provider) If this is indeed Google's strategy we will see subsidized Chrome OS based net books next year also.

> ... my point is that it is unlikely anyone would sell a device at a loss ...

Playstation 3, Xbox 360 . . . or are we just talking cell phones here?

Google's source of revenue isn't just selling the phone, similarly Microsoft and Sony's source of revenue wasn't from selling the console. It's all about selling the content and getting people to keep coming back for it.

Perhaps a $99 phone would be absurd for a state of the art smartphone, however making a loss wouldn't be absurd when google is making money off of every search you make, every youtube video you watch and submit.

Parts alone are probably closer to $200, but $300 is a more realistic break-even point when you add in manufacturing costs, all the shipping that has to take place, and the fixed engineering costs.

Also worth noting, while Apple prices theirs higher because they make money on the actual handset, Google makes money on every smartphone that gets into the hands of a user from the ad revenue.

Get enough smartphones into the hands of users and Google will see their revenues increase from the increased viewing of ads. So they needn't sell the phone at such a high margin to make a profit.

Really? The iPod touch is $199. Cell radios really aren't that expensive.

Yet the iPhone costs more than twice that.

And the ADP1 (an unlocked G1) has relatively modest specs compared to this new phone and costs $400 from Google. I don't think that's because Google is trying to make a mint by sticking it to developers.

Virtually all new HTC smartphones retail in the $400-600 range. I don't think HTC is making a fortune off them.

Because apple is pretty greedy, and the iPhone has a camera.

Apple is setting a price that millions of people are willing to pay. I wouldn't call that greed.

What would you call greed, then?

Setting a price that's more than what your product is worth in the marketplace.

You can always spot real 'greed' because it's always self-defeating in the end.

Just because they're selling at market price, it doesn't mean that they're not greedy. If they were a bit less greedy, they'd rather have more users for a lower price, which in the long run can actually result in the same revenue. But then I'd get into the luxury brands theory/subject, but meh.

The best estimates still only put production costs of an iPhone 3GS @ $178.96 1

The rumor isn't so crazy if you remember that google's most general mission is to get people online and using the web and looking at their ads.

You can guarantee that they know roughly how much ad revenue the average smart-phone using individual brings to them and therefore they could justify bringing a phone to market at cost or even below.

1 http://gigaom.com/2009/12/09/who-is-getting-rich-off-the-iph...

Best estimate, yes. But it's also the only public estimate.

Remember when Google announced that they were starting a free email service with a ridiculously large 1 gigabyte disk quota for everybody? They announced it on April 1, and everybody assumed it was a particularly implausible joke. How we laughed! And then it turned out to be real.

Keep your skepticism shields up, of course, but this sounds like it could happen.

For a decent phone you need to add: radio subsystem, bluetooth, camera.

Along with these you get problems of heat dissipation, short battery life, long certification process with carriers around the world and god knows what.

It doesn't matter: Google knows it will make good money from mobile search, and if the offer those phones at a market-shattering price, in which the will sell it in volumes, they will more than make up for what they lost in the first place.

You know the ps3 cost in excess of 800 to manufacture at launch, and sold for around 400. The us army/navy, forgot which, bought a bunch of ps3s to harvest the cell processor. But that doesn't mean that's what most people do. Xbox Also cost more to produce, but I haven't heard of anyone selling it's parts.

Yeah I was about to write about this. Those consoles were sold at a loss because they could make up the difference and then some on game sales (through licensing or however that works).

The qestion then becomes how would Google monetize these phone users? Do they listen to our phone calls or text messages and send us even more targeted ads? Are they looking to monetize voice communication they way they did email?

In order to sell at a loss they have to be making it up somewhere.

It wasn't just game sales. Bill Gates has seen the need and value of a PC underneath everyones TV since at least 1995. Set top box software. Web TV. Etc. Xbox is the next step in that strategy. Sony saw the same thing.

And this is no different, I don't think. Google sees value in putting accessibility to its services in everyones pockets.

Where are you getting the $300 figure from? Also, elsewhere in this discussion you say "Yet the iPhone costs more than twice that" (ie >$400, [1]). iSuppli estimates the cost to be $180 including manufacturing and "other" parts[2]. This is very much in line with Apple's typical margins. Apple doesn't sell commodity products so it's no surprise that they could be netting 50% margins (see [3] for iPod cost information). I work in the electronic components industry. The component costs quoted by iSuppli are credible. Companies like Samsung and Apple are able to negotiate very low prices for components from their suppliers. The suppliers are willing to offer components at lower prices and hence sacrifice their profit margins for a couple of reasons: (a) Apple and Samsung buy in large volume (b) Apple and Samsung will always be around. Offering the best prices now is a foot in the door for future business with them

[1] http://news.ycombinator.com/item?id=996710

[2] http://gigaom.com/2009/12/09/who-is-getting-rich-off-the-iph...

[3] http://www.businessweek.com/technology/content/sep2005/tc200...

They're reportedly unlocked, but still each require a $30/mo, 24 month T-Mobile data contract.

You'd have to make quite a bit on those parts to cover the subscription costs.

The only thing anyone knows for sure is that Google handed out a demo phone to it's employees last week.

Every other part of this story is pure conjecture, speculation, and anonymous sources.

I think the Internet is setting itself up for a huge disappointment here.

I couldn't agree more. I was stupefied when I saw this headline on HN.

Still, the headline is only six characters from the truth. Prefix "Rumor:"

But isn't that the point? How valuable is it to discuss rumors?

I heard a rumor last week that Microsoft is going to buy Apple. Should we spend dozens of articles talking about it? I heard another rumor that the movie Avatar is a secret communist plot to take over the world. Is that worth 10 articles too?

Also I'm not sure if someone edited "Rumor:" out of the title but it is currently neither in the HN title or the Reuters article.

I guess the interest in and plausibility of the rumor will govern how much it is discussed. If the rumor being true will affect people (say iPhone developers) then it seems natural that people would want to discuss the likelihood of the rumor and what the consequences of it will be.

My reaction was "This sounds way too good to be true."

My reaction aloud was "Yeah, right."

I will bet 2:1 odds that Google will not sell the phone for $99 to people with old Google accounts. I'll put up USD$100 (against your $33), payment to be made by donation to the winner's favorite charity, proof by forwarding the email receipt. Reply to this comment or see my contact details on my user page to take up the offer and work out the details.

Edit: changed from 2:1 odds to 3:1 odds (edit 2: the original edit changed the other side's amount from $50 to $33, though as jfoutz pointed out I'm not thinking of ratios correctly, so I've changed the ratio back to 2:1 and left the $33 as $33).

Edit 3: thanks kw_ for taking the bet! For the record, the main reason I think I'll win is because the iPhone reportedly costs $179 to make[1], I'm assuming the cost of the Google Phone will be similar, and I don't see it making sense for Google to sell it at a loss. I choose the charity Village Outreach, because it is the top-rated charity at the charity-effectiveness-judging organization GiveWell[2].

[1]http://9to5mac.com/gigom-who-rich-iPhone-34040

[2]http://www.givewell.net/charities/top-charities

odds are a tough way to present a wager. I often mistake the syntax. you had it right the first time ;)

the : implies the inclusive form 3:1, 3 out of 4 trials in your favor vs 1 out of 4 trials in my favor.

http://en.wikipedia.org/wiki/Odds

Also, no bet, but good luck to you and your charity.

I'll take up your bet. Emailing details.

I'm not the betting type, but I could see why Google might do this. A big strength of the iphone as a app buying platform is that everyone already has a payment method set up, so you just have to click a button to buy. Google could require people to buy the phones with Google Payments, then they'd automatically have a "spend money" button hooked up, ready to buy apps and whatever else.

The only way this even sort of makes sense is if Google's primary goal is to force a dramatic drop in cellular data plans. A cheap, desirable phone that's not locked in to a carrier via a contract could, potentially, force intense competition on rates. The phone is removed from the equation when selecting a carrier, and consumers can change carriers on a whim with no penalty. Service has to improve and prices have to drop.

I don't know though. That looks a little too big picture for an actual business strategy.

Problem is you can already by phones without contracts and carriers are in no rush to negotiate to lower prices. If intense competition you mean 20 carriers, I agree. But most countries have 1-3 carriers. And most haven't moved on price at all. They might add features and bandwidth, but rarely do you get a price break.

First hint this is total bunk: There is no prepaid $29 data plan on AT&T. The Pick Your Plan menu includes a $19.99 100MB data option, thats the maximum prepaid data option. On T-Mobile, the internet-only postpaid plan starts at $29.99 for 200MB and $39.99 for unlimited data. I doubt the prepaid plan is CHEAPER, if it even exists (and I couldn't find it) and I don't think the carriers will be creating new plans just for this phone that isn't even being sold through them (in the scenario described).

This is an unconfirmed rumor. It looks like it was sourced from a semi-anonymous comment on another blog post.

More important question: If this product is for real, and it takes off, are Google's partner carriers prepared to deal with the surge in traffic?

If this is really true, I will consider ebaying my droid, coughing up the $150 cancellation fee and saying goodbye to Verizon forever.

The U.S. badly needs an open (i.e. European-style) cell phone market. Google has (hopefully) realized this and is doing something about it.

I wonder if these will work in Europe. If so, I'll get someone to ship me one soonest... that's a decent price for an unlocked Android phone, which I've been wanting for a while.

The article on [1] rumours pricing, costing €20 per month for roaming on HSDPA. One just has to wait until it is released to find out I guess, but I too look forward to seeing the device being launched in Europe/UK.

[1] http://www.androidguys.com/2009/12/14/reuters-nexus-one-avai...

I know one Google employee who has done exactly that already.

So it's an HTC phone built with Google's input, will be carried on T-Mobile, and Google sells an unlocked version. The G1 could be described in exactly these terms!

Seeing this kind of rumor posted to HN always makes me wonder if it has jumped the shark.

The Dream is the G1, it was never known as the Dream. It was known as the Dragon/Passion and now Nexus One.

Any AT&T users just ready for something different? I love my iPhone, but I got an original one so I've had the same experience for three years now. I might pick this up if it's compatible with AT&T's network, just to see the other side of the coin.

Would you sell your privacy for a bargain on a phone? I think I might be tempted to. Sheesh.

is there something specifically wrong with Android phones vs other platforms or is that another tired reference to the Schmidt quote?

I don't know about wrong, per se, but Android phones are absolutely tied to your Google account. Doesn't have to be your MAIN Google account, of course, but still.

Sure, to the extent you want to access google services. But isn't that a given?

My desktop is tied to my google account to the same extent that I choose to use google search and email.

No, to the extent where the first step in booting up your Android phone for the first time is filling in your Google account information (at least on the HTC Dream I tried). It's closer to Windows asking you for your MSN or Live or whatever login the first time you boot it up.

And here it is. I have a specific account for only my GVoice. It would, rest assured, be tied to that one. Not much else happens on that account.

I love Google. I use Google. And I would have said the exact same thing two years ago. At least Schmidt came out and eventually said what pretty much every company that has shareholders stands for. Ergo, this is not a tired reference to anything, really.

It just so happens that Google by their very nature already knows a lot more about me than my current provider (vzw) will ever know. And using a phone on their service will increase Google's knowledge that much more.

I'm honestly not much a smart-phone person. I was an early adopter (Windows CE phones) and ever since then, it's given me a bit of a bad taste in my mouth. Smart-phones have not changed much, really. At their core, most are not as good at being a phone as some of their less-snazzy counterparts. That was my original gripe.

What has changed: ubiquity of mobile-friendly content and increased bandwidth to handsets. Now it doesn't matter much. Smart-phone users are more likely to use their devices for everything BUT talking on the phone.

At any rate, I think the fact that android ties to a google account is a bit of a privacy fail, but IIRC Sidekick does the same thing, roughly (though not to Google). Then the fact that by default the calls will go through GVoice means that Google absolutely knows everyone you call, everyone that calls you, every SMS/MMS. The privacy implications of that are similar to Google setting up its own high-performance DNS (that's been hashed over in a few threads here on YC as well, poke around).

Still, with all that... I'm kinda lusting for this gadget.

That just sounds too good to be true, especially the data plan situation.

Android Central addresses the rumor status of this story: http://www.androidcentral.com/getting-little-something-our-c...

* There is no "Google Phone." * HTC, not Google, is the manufacturer. * Currently, no carriers have been announced. * We don't know if it will be available for purchase by the general public. * We don't know when the Nexus One may be made available for sale. * We don't know what it will cost. * The Nexus One, for all intents and purposes, has the basic features of today's high-end smartphones.

More hype-hate here: http://forum.androidcentral.com/htc-nexus-one-opus-one/4182-...

And Droid already has android 2.1 ported to it: http://www.androidcentral.com/android-21-ported-motorola-dro...

AHAHAHAHAHAHAHAHAHAHAHA!

wipes a tear

Seriously, VOIP, over US carriers? Yeah, they'll let that happen...

T-Mobile is the little guy right now but it has the most forward-thinking policies, however with limited 3g coverage.

I have been a T-mobile customer since 1994 (they had 2 handsets and 2 tariffs back then, business and personal) and this is the first time I have heard anyone describe them as forward-thinking.

I hated them when they were VoiceStream.

They were OK when they were Mercury/one2one.

To be fair, their dataplan in the UK is pretty reasonable. Customer support for anyone who isn't on BlackBerry is dreadful.

I've only been a customer for about 5 years, but in that time I've heard their CEO saying they won't be banning or hunting down people who use their phones for tethering, I've seen them roll out the Android phone before anybody else, and I've had really good experiences with their network (in msp and in the rocky mountains). I don't know if any of those are "forward-thinking", but I've actually been a happy customer, rather than a grudging one.

I pay $60 per month for unlimited voice and data, and I can tether as much as I want with no fees, etc.

tmobile's main problem is network coverage (around San Francisco it's quite bad) ...

But its policies and customer service are quite reasonable. From what I gather, at&t has just as bad network coverage but doesn't like you to tether and charges a lot more.

To get measurably better service one needs sprint or verizon, which charge even more than at&t

Supposedly they're starting to roll out HSPA+ now, which ain't too shabby

Google Voice is actually not VoIP. At least not on your end. It's basically just a slick interface to a long distance calling card.

That said, VoIP is surely coming to cell carriers. The FCC has already indicated that it isn't happy with user agreements from e.g. Verizon that restrict what applications you can run on their network.

From the article:

"Google will push full VOIP usage on these, meaning no voice/sms plans needed at any carrier. Voice calls will go through Google Voice on Data SIM cards and will provide unlimited free voice calling. And SMS is replaced by unlimited free Gtalk."

No one said Google Voice is only VOIP. I have it, I use it for non-VOIP calls. The article just says they're planning on doing all of the voice calls over IP. Considering there aren't data carriers that have affordable unlimited free calling, that's the only way you could get "unlimited free voice calling".

The article is wrong.

Do you have sources?

Specifically, they're saying you don't need a voice plan at all, just data, so they'd be using only VOIP, which Google Voice supports. No where are they saying that GV is only VOIP.

but GV is not the google phone, right?

So fine you can use google voice on your old phone, but this is a separate fact from the question of whether the google phone uses data/voip for voice calls.

Correct on both counts. But the Nexus One will be able to use data/voip for calls as well as a regular wireless calling plan, just as my Android Dev Phone 1 does now.

Interesting. Out of curiosity, do you (the user) notice a difference in the UI or call quality when it makes the call over data/voip vs as a regular wireless call?

(on the blackberry google voice app it only supports regular calls, which are all done with forwarding, callbacks, etc.)

Google Voice can send calls to Gizmo5 (a SIP provider they just bought). Phone number forwarding is the feature they let most people use, but there are a number of people doing straight VOIP with it.

T-Mobile already does this somewhat--you can switch between WiFi/VOIP and the mobile network. UMA is great.

http://support.t-mobile.com/doc/tm23449.xml

Can you actually use VOIP over their network, or just over WiFi?

I think it's either voip over wifi or voice of cellphone network. What it sounds like to me is that this phone has an 802.11 chip in it and can be used over a data network.

The bit about not needing a wireless carrier seems to suggest that.

Not that I know of. But thanks to a smaller 3G network footprint T-Mobile is being pretty scrappy with the WiFi. Not a bad play as that can cover you at home and work. As AT&T has shown, you probably don't want to be making calls over the 3G network anytime soon.

So, T-Mobile is pretty bad at explaining all this, and majorly screwed up branding and markering.

T-Mobile has two related but different services, which have been called T-Mobile @Home, and T-Mobile HotSpot @Home (now called wifi calling). @home is a home-phone replacement service, which comes UMA-equipped with a POTS phone plug for you to plug a phone into. HotSpot @Home is actually the service the parent is referring to, which can be considered 'femtocell using wifi'.

The way it works is -- your phone needs to connect to T-Mobile's GSM core network one way or another. Normally it connects over the mobile network, but UMA-equipped phones can connect via a VPN (IPsec) connection to T-Mobile. Once connected it speaks the same GSM language as it would speak over the air, so everything works, voicemail, phone calls, text messages, etc.

UMA is actually pretty awesome -- it can hand-off midcall back and forth between wifi and the cell network, unlike AT&T's femtocell solution.

Another HUGE hidden feature: As long as you can connect over wifi, your phone can connect to T-Mobile's network and send/receive calls, text messages, everything, using home area rates. Anywhere in the world. Billing over UMA works just like connecting to the cell network when you're at home.. so if you have unlimited at home you get unlimited UMA, or if you have 500 minutes you get the same 500 minutes.

This works on ANY wifi network, not just T-Mobile's routers. T-Mobile's Hotspot network of Wifi routers is specially configured to let UMA-equipped phones connect to T-Mobile's GSM core without having to pay.

In summary, T-Mobile's UMA service is awesome, but T-Mobile doesn't seem to be able to tell the world how awesome it is.

Sidenote: I used to work for HTC, and I kept trying to convince everyone of how important UMA is. It's a technologically superior solution to femtocells, but for some reason femtocells seem to be 'winning'. They are slightly easier to explain (a miniature cell phone tower!), but they actually really suck in comparison to UMA. It's tragic to see it happen.

Isn't the wifi radio a much bigger power hog than the cell radio?

Actually, no.

UMA-equipped phones support what is called Unscheduled Automatic Power Save Delivery, an extension to wifi which reduces wifi battery consumption both when idling and when on a call to about the same level as a cell connection.

UAPSD (branded as WMM Power Save) is part of the WMM WiFi extensions, which are a part of most consumer wifi routers.

Reply on news.ycombinator.com