Live data from Hacker News

Update On The TechCrunch Tablet: Prototype A

techcrunch.com

21–30 of 46 posts

Re: Update On The TechCrunch Tablet: Prototype A

#21
post #2

Classy. Do these guys really think you can brew up a $200 web tablet at an old kitchen table? I've been critical of this project before, but my advice on the matter would be to refer techcrunch to the first facet of this effect: http://en.wikipedia.org/wiki/Dunning-Kruger_effect I don't understand how the commenters on TC think it's anywhere near an actual prototype. Groupthink? If they're serious about this project,…

They've got a team of people and multiple companies involved, a basic case, hardware plan and working model, and your response is: "you can't do that, your working model isn't good enough, you're all incompetent and overreaching, you're a bunch of nerds and you're just stealing other designer's hardware." Why does your comment have any positive votes at all? Much as I think the idea is silly (keyboards forever!) "Do…

That's a fair synopsis of my response, yes! The very fact that they have built a prototype already means they're doin' it wrong.

The correct approach when you fly in with a set of specs (which seemed to be $200, touch screen, browser) is to find suppliers and negotiate your build cost down to sub $75 so that you can afford to bring it to market at $200

This kind of budget, with the volumes (small) and capital (small, mike arrington isn't dell) that TC would have for this project means that well before you're at the stage of ordering samples building a prototype, you would have needed to have done a whole lot of work at the negotiation table to even get into the ballpark.

I'm all for optimism, but honestly - this isn't a rails web app - it's hardware.

Re: Update On The TechCrunch Tablet: Prototype A

#22
post #17

Earlier quoted context omitted.

Do these guys really think you can brew up a $200 web tablet at an old kitchen table? Actually, that's a very good sign. Every one of the smartest EEs I know works like this: In apartments or labs filled with piles of random wires and oscilloscopes and beat-up hard drives. You can tell which parts are the important ones because they're the ones that are only half -covered with random junk. The first danger, when this…

You can already pick up eee's and copies for $200 or so... I'm not sure where the market is for a tablet like this personally. It's less useful than an iphone (not portable), and less useful than a laptop (No keyboard).

Less useful for what? If it had the right form factor, it might be more useful for browsing Hacker News over my lox & bagel, which is what I'm doing right now. (Commenting would require voice recognition, which is not up to snuff, yet, or a software keyboard, which would be worse. But a little Apple Wireless keyboard could be a good adjunct for that.)

Re: Update On The TechCrunch Tablet: Prototype A

#24
post #3

Earlier quoted context omitted.

>Do these guys really think you can brew up a $200 web tablet at an old kitchen table? I'll believe it once they say what they are using for the screen...

Yeah. We don't know yet what their screen is or what it cost them. In fact, that information wouldn't be useful anyway: I'm guessing that Michael Arrington might well be able to get one model of any screen he wants for well under $200. But once he's got that screen designed into his product and is trying to contract for 2M of them... well, I'm guessing that Foxconn's sales team is pretty tough, given that they battle…

The screen?

It's Arrington's old macbook air screen.

http://www.techcrunch.com/2008/08/19/apple-is-flailing-badly... Quote: "That $1,800 piece of hardware has now been dismantled for parts for a project we’re working on here."

Re: Update On The TechCrunch Tablet: Prototype A

#25
post #18

Earlier quoted context omitted.

Do these guys really think you can brew up a $200 web tablet at an old kitchen table? Actually, that's a very good sign. Every one of the smartest EEs I know works like this: In apartments or labs filled with piles of random wires and oscilloscopes and beat-up hard drives. You can tell which parts are the important ones because they're the ones that are only half -covered with random junk. The first danger, when this…

And that is exactly my point. It's one thing to build a prototype, and another thing entirely to be able to mass produce a web tablet for $200. And no matter how good the EE, you can't do that stuff at a messy kitchen table - you need serious resources.

Well, sure, but those serious manufacturing resources belong to OEMs in China. As a US-based designer, building a factory is not your problem -- there's no hope of doing that profitably at this rock-bottom $200 price point. Your problem is design -- figuring out which combinations of commodity parts will do what you want. After which you'll take various iterations of your prototype to meetings with OEMs like Foxconn, asking them to bid for the honor of manufacturing the product in bulk and sticking your company's label on it.

Of course, there are lots of potential pitfalls in that process -- there are several middlemen who can mark the price up, there are lots of suppliers to coordinate and play against each other, and there's the constant risk that your design is lousy, or unmanufacturable (when the factory is half a world away, it's hard to establish a good design feedback loop), or that you'll cut so many corners trying to meet your budget that your brilliant design turns out lousy. This is presumably why there are so few decent $200 web tablets.

Incidentally, though I used to be a product engineer in the cellular components industry, I'm no expert on the overall design/manufacturing process -- I was a tiny cog in the giant cellphone machine. For actual knowledge I recommend something like Bunnie Huang's blog posts about hiring folks to build the Chumby:

http://www.bunniestudios.com/blog/?cat=6

UPDATE: I just surfed through my own link, and I have to say that Bunnie is a great blogger. His hints for getting quality designs out of Chinese factories (my capsule summary: "live in China, watching the assembly line, for as much time as you and your budget can stand") are in line with my own experience.

Re: Update On The TechCrunch Tablet: Prototype A

#26
post #13
post #2

Classy. Do these guys really think you can brew up a $200 web tablet at an old kitchen table? I've been critical of this project before, but my advice on the matter would be to refer techcrunch to the first facet of this effect: http://en.wikipedia.org/wiki/Dunning-Kruger_effect I don't understand how the commenters on TC think it's anywhere near an actual prototype. Groupthink? If they're serious about this project,…

Well, a beagle board, [beagleboard.org] is only $150 bucks in volumes of one so if they can find a touch screen for 50 bucks they basically have their 200 dollar device.

Last time this came up I posted this:

Lets do it with stuff we can get and see what happens:

Main Board: (truly open, schematics and all) : http://beagleboard.org/ ($149)

Screen : Lilliput 8" 1024x768 touch screen (http://www.case-mod.com/lilliput-809gl80npct-lcd-touch-scree....) ($259)

Network : 802.11g wifi (http://www.usbgear.com/USB-Wi-Fi.html) ($30)

Battery: 7.4v LiIon 5200mAh (4 hours run) http://www.all-battery.com/index.asp?PageAction=VIEWPROD&#38.... ($50)

Case: Injection Mold plastic (http://www.emachineshop.com/molding_machines/injection_moldi....) ($10)

Cost to build a single one, ignoring volume discounts and cost to assemble: $498.

I still think the price is going to end up being around this. You could actually go and make this today. It would be cool if someone were to just throw up an instructable and a compact flash image of the OS for the beagle, then we could all go make our own. Someone could design a nice plastic case and have a bunch made (or just have emachineshop hang on to the molds) so they'd be cheap for all.

Re: Update On The TechCrunch Tablet: Prototype A

#27
Browsing the comments on TC, and I found this quite telling/interesting.

http://www.techcrunch.com/2008/08/30/update-on-the-techcrunc...

In case it goes: "A couple of weeks ago I was asked to make a simple cad model for the prototype hardware. http://tctablet.altronix.se/#2.11

I would like to finish the work but since last week I have not been able to to get in touch with them on either irc or skype… :/" Refer

Re: Update On The TechCrunch Tablet: Prototype A

#28
post #17

Earlier quoted context omitted.

Do these guys really think you can brew up a $200 web tablet at an old kitchen table? Actually, that's a very good sign. Every one of the smartest EEs I know works like this: In apartments or labs filled with piles of random wires and oscilloscopes and beat-up hard drives. You can tell which parts are the important ones because they're the ones that are only half -covered with random junk. The first danger, when this…

You can already pick up eee's and copies for $200 or so... I'm not sure where the market is for a tablet like this personally. It's less useful than an iphone (not portable), and less useful than a laptop (No keyboard).

And yet, this is exactly what I've been looking for for the past year or so. Not every computer needs to be a content creation device; I want a computer for my kitchen that's used for retrieving recipes. I can make the software to store and present the recipies, but I can't find a keyboard-free tablet (basically a 10" nokia 770) that would be useful for displaying the recipies in my kitchen, mounted on the wall.

One man's useless junk is another man's dream...

Re: Update On The TechCrunch Tablet: Prototype A

#29
post #13

Earlier quoted context omitted.

Well, a beagle board, [beagleboard.org] is only $150 bucks in volumes of one so if they can find a touch screen for 50 bucks they basically have their 200 dollar device.

Last time this came up I posted this: Lets do it with stuff we can get and see what happens: Main Board: (truly open, schematics and all) : http://beagleboard.org/ ($149) Screen : Lilliput 8" 1024x768 touch screen ( http://www.case-mod.com/lilliput-809gl80npct-lcd-touch-scree... .) ($259) Network : 802.11g wifi ( http://www.usbgear.com/USB-Wi-Fi.html ) ($30) Battery: 7.4v LiIon 5200mAh (4 hours run) http://www.all-ba…

Your link for the lcd touchscreen is truncated with elipses. That is, the actual href="" part is truncated. Is this a feature of the hn software, or did the link get pasted in wrong?

Re: Update On The TechCrunch Tablet: Prototype A

#30
post #2

Classy. Do these guys really think you can brew up a $200 web tablet at an old kitchen table? I've been critical of this project before, but my advice on the matter would be to refer techcrunch to the first facet of this effect: http://en.wikipedia.org/wiki/Dunning-Kruger_effect I don't understand how the commenters on TC think it's anywhere near an actual prototype. Groupthink? If they're serious about this project,…

No, it's doable - BUT not the way they're marketing it. As a pure gadget and not a PC, it's very possible. Take out the touch screen, add a keyboard, use ARM instead of x86, and you're all set. I blogged about it in detail here: http://neosmart.net/blog/2008/what-the-techcrunch-tablet-sho...

A keyboard for this kind of device would be stupid. If that's what you're after, just get an Eee. And browsing the web without a touchscreen or some kind of pointing device would be a horrible experience.

They've got the concept nailed for the most part...now the challenging is building it for that price.

Post reply on HN