Glad to know there are real Cheap OSS PDA project which target the low end market. But wait...
I've played around Linux + DragonBall + 4~8MB Flash + LCD Panel + Wireless design for 3 years. With 2 designs onhand, 1 aborted prototype, 2 boss(es) and at least 6 OEM vendor meetings I can tell YOU BIG BOY OF AMERICAN cannot easily find this product off the shelf:
#1 : Buyer from American
PDA is NOT a toy. Most people buy for personal use. They are mostly middle class, geek, engineer and student. Poor men don't need a Personal DA. For US and EU market, when a electronic product sell To sell any IT-grade electronics stuff in US, you need a support network that span the whole America continent. Any manufacturer who make this stuff need to talk to big agent. A less-then-$50-PDA-directly-compete-with-Palm-and-O SS-inside is absolutely not CONVINCING!
#2 : Buyer from !(US || EU)
You need localized version of software. This is petty hard for CJK character: Smart IME, big font image size,... This effectively cut out cost-benefit of OSS kernel. Any one ever try to make a handwritten recognition algorithm for CJK character set know what I mean.
A PDA must talk to PC to get new software and upgrade. Many Asia, Afica and some Middle East contries just don't have this condition. Any non-PC based content and software distribution is relatively expensive.
A low end PDA is not much than Game Boy+Memo(Calendar)+Talking Dictionary. For the price of 8M Flash you can buy a electronic memo, a DragonBall can buy a talking dictionary. Why? because there are cheap single-or-two chips design from Taiwan and Korea. Any DragonBall based design, though interesting for any embedded system designer, just cannot compete in real low end market.
#3 : OEM factory
OEMs which can handle PDA manufacturing are either contracted to [any big brand you name it], or only talk to big agent (see #1).
SME (Small/Medium Ent.) which really want to consider immediately face these:
The production cost of making a Visor and a U-NAME-IT-OSS-PDA are not much different. There are cost in Inventory, strict QA + QC, testing, import/export/FCC certification, etc. Added up, most SME are out of question.
Let material cost / unit = $50 x 80% = $40. In 100K quantity (If you don't know why no manufacturer want to take 10K order, stop reading) it add up to $40 x 100K = 4 Million US Dollars! Unless you get good deposit from big buyer NO SME CAN take this amount of Payable.
Motorola is a REAL-HARD-TO-TALK-TO big ass. If you don't have a contract/PO onhand from recognizable Brand DON'T EXPECT ANY technical support / pre-sales meeting, at least in Hong Kong and China. That said, any SME OEM must tackle ALL technical problems in post-design and pilot-run stage without vendor support.
#4 : You geek
You can't expect much people work on OSS project other than Eng-German-Japanese speaking guys:-( OSS Linux based PDA still need many works in UI and usability design. To make it into international market/community, you need many people to work on different area.
As I know quite many OSS based consumer devices are killed in design/prototype stage for some or all of this reasons. That's why you need to pay that much to buy a car Mp3 player or TiVo. To sell it you need to put many STAE-OF-THE-ART technology inside - from custom ASIC to special LCD panel to hard-to-license-band - no matter the consumer need it or not.
That's what I think: If this project is REALLY for 3rd countries with not-for-profit fund support, that's GOOD. If you want to make it a commercial product, think twice.
Slashdot Mirror
I've played around Linux + DragonBall + 4~8MB Flash + LCD Panel + Wireless design for 3 years. With 2 designs onhand, 1 aborted prototype, 2 boss(es) and at least 6 OEM vendor meetings I can tell YOU BIG BOY OF AMERICAN cannot easily find this product off the shelf:
#1 : Buyer from American
#2 : Buyer from !(US || EU)
#3 : OEM factory
#4 : You geek
You can't expect much people work on OSS project other than Eng-German-Japanese speaking guys
As I know quite many OSS based consumer devices are killed in design/prototype stage for some or all of this reasons. That's why you need to pay that much to buy a car Mp3 player or TiVo. To sell it you need to put many STAE-OF-THE-ART technology inside - from custom ASIC to special LCD panel to hard-to-license-band - no matter the consumer need it or not.
That's what I think: If this project is REALLY for 3rd countries with not-for-profit fund support, that's GOOD. If you want to make it a commercial product, think twice.