"..A less positive experience has been the impact of state-monopoly postal services and punitive tariffs - often as high as 100% - on availability in markets such as Brazil.
There, a $35 (£23) Pi will currently cost you the best part of $100 (£66).
I believe that these measures, aimed at fostering local manufacturing, risk holding back the emergence of a modern knowledge economy in these countries."
I've just been reading 23 things they don't tell you about Capitalism, by Ha-Joon Chang (a Korean economist based in Cambridge, like Eben... perhaps they should meet up in a pub sometime?). One of his key points is how rich countries got rich by creating tariffs, which they then expect poorer countries to forgo under the pretense that it'll improve poor economies.
The problem is that it's one rule for us, and another rule for them. Broadcom, the BCM2835 and to an extent, the Raspberry PI is a product of this; it creates low-barriers to entry for wealthier entities like Broadcom and high-barriers to entry for less wealthy entities, like Brazil or ... me.
For example, the Raspberry PI (which I believe is a great idea) is roughly the same price as a FIGnition, in effect, about £25 compared with £20 for my product. Yet Raspberry PI is about 10,000 times more powerful (and complex) than FIGnition. That's because firstly, the Raspberry PI (Model B) is built in China (which reduces costs).
Secondly, it's massively mass-produced (which reduces costs).
Thirdly, they have the sophisticated tools to both design the chipset and the PCBs - a hobbyist can't build one at all even if given the components, a perfectly steady hand and a microscope (an infinite barrier for a hobbyist).
Fourthly, they have access to all the design documents - Broadcom don't even publish them for the general public (another infinite barrier for hobbyists).
Fifthly, they have volume access to suppliers, at a much reduced (secret) cost, which the general public (like me) don't have access to.
So you can see, there are barriers I face that others don't. But this is OK for me - I wouldn't produce a Raspberry PI instead of a FIGnition anyway, because I want people to build computers that can be understood - and that means building rudimentary computers that can be reproduced by the general public. That's because I believe that in the same way we teach kids to read using phonics rather than War And Peace and we don't believe that books have superseded the alphabet; if we want kids to master technology, they really need to start at the basics.
Similar barriers though apply to developing countries. There are inbuilt barriers to them, starting with for example, the fact that the brightest people in developing countries end up in places like Cambridge. Developing countries therefore often want to protect their own markets, for example, Ghana likes to grow and eat their own chickens, but instead Europe dumps our excess chickens on them, so they can't develop their own market.
It's unfair. But it's doubly unfair for a protectionist company Broadcom that makes use of cheap Chinese labour to force open markets in developing countries for their own benefit. Surely, surely, if Broadcom really did want Brazil to have lots of cheap Raspberry PIs, they'd simply give the information to Brazil to procure their own production runs - ship the raw chips to factories in Brazil and let them make their own PIs. There would still be the chip trade barrier, but not the for the added value of board production and assembly.
Problem solved - and the PI would be used for its intended purpose, not for sticking in thumbs and pulling out plums. Please Eben, you're a nice guy (as I recall from the BBC@30 event); don't use the PI as a free-market political tool :-)