Cory--

   Danger managed to hack (maybe unreasonably) unlimited bandwidth out of a culture that's very, very addicted to maximizing dollar-per-mhz. The only way they got to do this was by showing that there was only so much bandwidth someone could use plinking keys into the included apps.

   The SDK utterly breaks that.

   Look. Danger has, like, no support within the T-Mobile organization, save for a few accountants worried about the stock hit of a bankrupt Danger. I'm utterly tempted to set up a hidden camera and record what happens if you try to buy the thing at the stores -- it's really That Bad. They made a damn fine product, and made it affordable. They sold it to you, and you made a good pick. But T-Mobile owns its infrastructure, and its bandwidth, and they're *terrified* of what you might do with it, outside of their control. What about encrypted VoIP? What about porn streamers? What about cellular jammers? Better to shut it all down then to let the unknown in.

   This is a problem.

   Several decades ago, the TV networks were banned from producing their own shows -- they had to go and negotiate with independent studios to get anything on the air. The alternative had proved too corrupt, too restrictive, and maybe even too uncreative to last, and a competitive market of pilots and series was born. T-Mobile 0wns Danger, and what's the result? The SDK ultimately gives T-Mobile control over what goes on the Hiptop.

   There's a desperate, thirsty market for high quality, easy to use, cheap and effective wireless access. You'd think that'd be obvious, but look how utterly shocked the business world is at the success of WiFi. I think they really, truly believed there was a massive market yearning to spend a hundred dollars a month on wireless internet access, controlled end-to-end by some goliath -- down to what web pages we're allowed to go to, SBC/Yahoo. T-Mobile may be trying to protect their influence...but I think they're reducing the relevancy of their product. *sighs*

Yours Truly,

   Dan Kaminsky
   DoxPara Research
   http://www.doxpara.com

Ah, but did you buy the underlying network? The Danger servers that are doing all the on-the-fly modifications to the HTML and images and storing all your data? Did you know that the devices themselves are sold at cost, unlike most phones? Danger's strategy is very different from Palm/Handspring/Nokia's. Their interests are much more (though not entirely) aligned to those of the service provider.

I think it's premature to judge Danger on these scores yet. Let's see what they actually end up doing, not what they could do. I very much get the sense that they are making it up as they go, so I'm reluctant to judge them as evil when all they are is unprepared.

Besides, the Hiptop is already so nice compared to what i had before that really, I'd be OK if no other apps were ever made for it (even though I'm writing a Hiptop app right now). I bought it for what it was, not for what it might possibly be. Yes, it would be a mistake for them to restrict development too much. I'm not convinced they are going to; they don't strike me as that dumb.

mahlen

Results! Why, man, I have gotten a lot of results. I know several thousand things that won't work.
        --Thomas A. Edison (1847-1931)

(Point of fact: the vast majority of phones are sold at or below cost with a service agreement)

I bought my Danger -- at full retail -- because I had seen a presentation from a Danger exec where he promised easy and open development for the platform.

If the objective of limiting the SDK is to keep the network overheads low, then throttling connectivity to unsigned apps would accomplish this end -- without restricting the ability of developers to build me apps that improve the functionality of the device (i.e., an iSync conduit that would make the device actually useful as a PIM, or a proper notepad that accomodated more than 20 notes, or a mailer with filtering rules, or a browser with find-in-page).

This is a device that is neither an end-to-end Internet box nor a phone, but rather an increasingly tragic half-man/half-beast with the worst characteristics of both.

It's really depressing, because this is the most promising device I've seen in years. My only hope is that criticism from actual Danger customers will sway the company into doing the right thing before it's too late.

Cory--

(Point of counterfact: The service provider almost certainly subsidizes the cost of your phone, to the point of handing cash money to the retail outlet, in return for the usage contract.)

   iMode was built on the model of "neat stuff becomes available, and people pay a bit to get it".

   T-Mobile really wants that. If shutting down the open sourcers means they get a chunk of the $10 for a non-broken notepad, heh.

   Of course, it's a little like Ford demanding a share of your bridge tolls...

   What's sad is they could have it both ways. Charge for online delivery, on the basis that it's so much more convenient to -- as a programmer -- receive a check from T-Mobile every month than to manage sales oneself. *sighs*

   Thing is, T-Mobile doesn't care. They don't! I called up their sales line like seven months ago, cash in hand ready to sign up. Why couldn't I? They were giving like 200 minutes voice for $40, and I needed more. I wanted to pay $60, or $70, and get 500 or a thousand minutes or some junk.

   I ended up spending my money on something else.

   I'm in the same quandry again, and as much as I absolutely want the product, I don't know if I can trust it to actually work on my behalf. I can hope. But it'd be nice to be able to do more than that.

--Dan

According to EE Times, the Hiptop costs $150 to manufacture.

If T-Mobile wants to be a consolidated billing agent for app developers, that's fine, but if I want to install freeware on my phone the carrier should have no say in it.

The Danger server farm costs money to run; fine, let Danger send those users a bill.

Mahlen: Regarding how "it's premature to judge Danger," their time is running out... When the hiptop first came out, I could not have been more excited and enthusiastic about it (and it's heretofore spurned potential).

I'm halfway through my one-year contract and they've just about burned off all that good will. If things don't improve, drastically, come October, I won't be renewing.

Dan: I switched up to the $60/mo 500/Unlimited plan that they finally offered a few months ago. So, that'd work if you're going to pick up one of the upcoming hiptop colors (they've just about given away all the current Sidekicks), but if this program loading fiasco doesn't get sorted out, forget it, why not go w/ a Symbian or Palm device that actually *has* tools and a developer community (err, not 'developer dollars').

That being said, I don't think there's ever going to be the hope of a true 'open' smart-phone platform anytime soon as long as the carriers have final say. But you know, those new tri-mode prototypes ( + mesh networks) sure look interesting...

[one can dream, right?]

I don't think the carriers have final say unless the device manufacturers want them to. For example, people seem to be using the P800 on various US cellular networks even though it isn't approved or sold by any of them. The only problem is that you'll never see a P800 bundled with service; you have to buy it separately. This is a model that worked fine for years in Europe. But Danger has chosen the proverbial "deal with the Devil" instead.

Note that any Bluetooth GPRS handset can be used to connect any (Bluetooth) PC running any code to the same cellular network (which has notably failed to collapse as a consequence).

Cory--

   Ah, but the real potential is in Always On, and Bluetooth (through this encapsulated-PPP junk i've been studying) ain't always on. I need to read more about how GPRS works, but I think Danger's managed some kind of semi-permanence, maybe at the cost of wedding themselves to T-Mobile's infrastructure...

   Tragic. Annoying. Again.

--Dan

General rule: It's only a platform for whoever gets to decide what runs on it. Right now, it sounds like it is either Danger's platform or T-Mobile's platform, but it's not the end user's platform.

If data 'overuse' is a problem, then they should just charge for the data. Pick a $/megabit and stick with it. If this mucks up your voice model, then pick $/megabit with low latency, and another $/megabit for fewer guarantees. FedEx or second-class mail - but no more. You'll just confuse the end-user.

They might think they want to charge by "what the bits are worth" - but this is dangerous territory. Bits for stock trades may be quite valuable. But bits for my horoscope are worth almost zero. Bits for video are probably worth even less on a per-bit basis. You can only climb up the 'bit value' scale if you are prepared to climb down it as well. It's market pricing, and the market NEVER guarantees to cover your costs.

Welcome to life working with telcos. Why are European operators stumbling badly in data offerings beyond SMS?

They've got developer forums and SDKs but do they have reasonable integration, reasonable commercial agreements or staff that get it? No. There is not much innovation coming out of the world of voice and GSM operators and it's poor. The marketplace for developers and application providers is limited and controlled by operators who **just don't get it**. And, this is only with 2G.

3G depends on a massive uptake of data useage as mobile voice becomes commoditised. Nonetheless, the pricing for data in 2G is so unrealistic that you would have thought that it was as valuable as immediate voice.

Danger, an innovator and doing great stuff, is, like many, many innovators in the mobile/wireless space, being brought to its knees and will probably go under like the others. The operators and their shareholders will continue to moan about how they can't get their ARPUs and data useage up while developers will turn their backs on the operators and see the welcoming arms of WiFi.

Meanwhile, the operators will continue to pretend that they are portal based broadcasters and provide useless data services to their customers. Fools.