Earlier this week, Michael Arrington created quite a flurry of interest by announcing the TechCrunch Web Tablet. Of course, any resemblance with the Redux Model 1 is purely coincidental. Nevertheless, Michael and his team might learn a couple of things from our own experiment, started ten months ago.

First, the announced pricing of $200 is pure wishful thinking. If an established company like Asus cannot sell a subnotebook for less than $300 while manufacturing millions of them, it’s unlikely that a custom-built tablet sporting a thinner case and expensive multi-touch screen could retail for anything less than $400.

Second, sourcing critical components such as the multi-touch screen might be quite challenging. Last time we checked, no large multi-touch screen was available, and developing a custom one would cost quite a bundle. Also, should such a screen become available from one of the established suppliers, it is very likely that Apple would buy all available units for an extended period of time, as is the case today for the iPhone’s multi-touch screen. No one should ever underestimate Apple’s mighty buying power.

Third, open source hardware design looks nice on paper, but it’s virtually impossible for any cutting edge product, mainly because of patent issues. When trying to bring such a device to market, one has to deal with the collection of patents that are owned by large manufacturers such as Apple or Samsung, and cross-licensing of patent portfolios is pretty much the only way to make it work, which in turn makes open source design a non-starter. Last time I checked, Apple owned quite a few patents in the area of multi-touch user interfaces…

Fourth, a web tablet without 3G wireless and GPS capabilities won’t be very useful. Problem is, building that into the device is more complex than it looks, for a host of reasons. First, very few chipsets integrating 3G, Bluetooth, Wi-Fi, and GPS are available today, especially if you’re trying to drive a high-resolution screen (say WXGA). In fact, there is only one (Qualcomm Snapdragon), and you cannot source it unless you license Qualcomm’s IP portfolio. This will cost you several million dollars in lawyers fees alone, assuming that Qualcomm even agrees to talk to you. And because Qualcomm is still battling Nokia on a nasty patent litigation case, the chipset isn’t really available to many manufacturers. Another option would be to contract Samsung to design a custom chipset for you, which is what Apple did for the iPhone, but you should be ready to spend anywhere from $50M to $100M for the design alone, and to guarantee that you’ll buy millions of them in the first 24 months or so. If you have that kind of money, and if your addressable market (readership) is large enough, this might be a viable option. Otherwise, forget about 3G and GPS. Second, putting all these wireless technologies into a single enclosure is actually more challenging than it looks, especially with respect to the design of the antennas. Antenna design is extremely complex, takes an awful lot of time and resources to debug, and very few engineering teams know how to do it for 3G, Bluetooth, Wi-Fi, and GPS altogether. If you’re planning to do that, make sure that you get access to this talent first.

Fifth, if you decide to forgo 3G and GPS, you might be able to pull it off, and the only complex part will be the battery. If you want an ultra-thin form factor, you’ll have to work with a battery manufacturer that can design a custom polymer battery for you, and this can take a very long time. Finding the right battery manufacturer is no easy task, and the thinner the device, the more complex the battery design, which in turns creates some interesting project management issues — even the smallest changes to the internals of the device will have ripple effects on the battery’s geometry. This is where crowd sourcing might turn out to be a liability more than an asset, for large crowds tend to have a harder time making their mind up. So here is my advice: freeze the feature set early, and design your device from the inside out.

Sixth, assuming that you could make the hardware work, you’ll quickly find out that hardware was actually the easy part. Software is what makes a device work (or not), and the simpler the form factor (no buttons, multi-touch screen), the harder it is to get the software right. If you don’t believe me, just ask yourself why the iPhone does not support copy-and-paste yet. User interface design is very difficult, and is best handled by a benevolent dictator, which is why the Mac OS X user interface is so much better than any Linux Desktop I’ve seen. Again, crowd-sourcing looks nice on paper, but…

Seventh, once you get the user interface done, start working on power management. This is something that you have to pay very close attention to at the hardware level (every mW counts), but it’s also critical at the software level, and this is the kind of thing that can easily create a 3 to 6 months delay in the project. And remember: the thinner the form factor, the smaller the battery, and the shorter the battery life…

Eighth and last, once you’re done with all that, you’ll likely end up with a device that is twice as expensive as you originally thought, and twice as thick. It’s also very likely that you’ll end up with a smaller screen (7" vs. 8.9"), making your device not much different than the host of Mobile Internet Devices powered by the Intel Atom CPU. So here is the million dollar question: why bother?

This takes us back to the Redux Model 1, and the critical lessons we learned in developing its concept: bringing a new form factor to market is immensely difficult. It takes a lot of effort, costs a lot of money, and usually fails, for largely unforeseen reasons. At the end of the day, usability is the result of complex interactions between hardware, software, user interfaces, and you won’t know whether your design works or not until it’s done and millions (or tens, or even hundreds of them) have been spent.

Also, most of the complexity and cost involved in designing such a device lies in three areas: chipset, antennas, and software, and most new form factors require custom developments for all three of them, which makes no sense at all. This is the reason why we decided not to build the Redux Model 1 as a standalone device, but as a shell (or jacket) wrapped around a modular core called Central Networking Unit, or CNU.

The concept for a CNU is very similar to the modu, with the main difference that our original idea was not to have the CNU act as a standalone mobile phone. Other than that, the two concepts are very similar, and based on the idea that all complex and expensive parts (chipset, antennas, and software) should be externalized, built once, and re-used across a wide range of devices sporting different form factors. By doing so, you save a lot of money on hardware, but also on subscriptions to your wireless service provider (sharing one account across multiple devices).

So here is my advice to the TechCrunch team: do not try to re-invent the wheel, no matter how fun that sounds, or how cool it makes you look like. Instead, lead your crowd to develop some custom jackets for the upcoming modu. It will be cheaper, easier, and a lot faster… That being said, I wish you luck in your project, and I will certainly buy one of your gizmos if you manage to build them.