0.5Gb internal storage. The Galaxy phone I point at can put a microSD in too - it's not quite the same as saying that it's got that as primary internal storage. And that Galaxy has 1Gb internal storage. My point is that YOU pay for the microSD. The device INCLUDE the internal storage in the price. And you get less with this device, for more cost.
And again, my point is cost. If you can get off-the-shelf components to do something similar for VASTLY reduced prices, then you have to wonder what you're paying for. If you can get a Samsung phone - with roughly the same makeup and components for 1/5th the price, and if you can build stuff with similar chips for similar purposes (even including LCD touchscreens, etc.) that use open-standards to communicate and no "hidden firmware" for a pittance - again, what are you PAYING for with this device?
This isn't how much things like this cost, clearly. Maybe the BOARD is more expensive, as it's custom. Maybe a PARTICULAR screen is more expensive as it's in limited supply. But the overall device? It's a bog-standard phone. Quite what "unfree" firmware does a commercial device like a Galaxy have once you've rooted the Android install on it? Pretty much the same as this device - the GSM chip will have a proprietary firmware to ensure radio compliance.
And if not, how much would it cost to replace, say, the GPS functionality on the Samsung with an "open" replacement? Again, nowhere NEAR the cost of this device (which is basically doing something similar by relying on the manufacturing for a previous device to provide the baseline).
In runs of one and tens, yes, maybe this price is reasonable. In runs of 100's and 1000's - no.. it's really not. Not in the age of ubiquitous fast low-power common-bus chips for all these functions.
This reminds me of the open-graphics-card initiatives. The time spent on starting from scratch (and inevitably relying on some closed piece of hardware at the end anyway) means that by the time it comes along few are interested, the costs are enormous, the parts are hard to obtain and the device quickly becomes obsolete (the project that springs to mind was still advertising PCI-only functionality just a year ago, not even PCIe). And it's based on FGPA's that you have to buy from a commercial vendor that doesn't publish their designs...
I'm a purchaser of niche products. I have a GP2X and some of its predecessors and successors, I programmed for it, it was built in tiny runs, cost more than equivalents, and was quickly obsoleted by commercial devices but it was "open" - it was ARM chips with a Linux install that you could code down to the bootloader.
5 times the cost of a (pretty expensive) commercial device that does the same and has the support of a major international company is a lot to ask for a niche product that doesn't do anything "special" (any Android machine, by definition, could run a plain Linux install if you so wanted to do that), is reliant on finding parts from old phones, and needs a tiny production run meaning you probably can never get the parts for it again if it goes wrong.
I'm not against the idea, here. I'm against the execution. There are cheaper phones that have bog-standard hardware that you can replace any firmware with open-firmware if you needed to (or even the entire functionality of that particular feature with another free equivalent) and put them in a case. In fact, here, the case is the CHEAP part. That's the part you could easily redesign to shove any board taken from a phone into. The electronics for a hobby project, however, is never going to get near the cost, reliability, even safety of a commercial product re-purposed.
The ideal isn't mine, particularly, but the execution seems incredibly poor is that's the closest you can get to price-point. Double-the-cost, maybe. FIVE TIMES is ludicrous.
