EU: Once we have some in-house teaching material to distribute, we'll be hosting a resource centre to aggregate this, along with third-party material that we feel is of good quality (and there's a lot of this stuff out there already). We're debating whether we should offer a community platform to allow teachers to share material they've created, but we're conscious that, at least in the UK, teachers are already quite well served in this area by organisations like CAS.
The H: IT Teachers are hard to find, at least ones who don't work off a guide in a binder. Is the Foundation working with anyone on creating a richer teacher training experience?
EU: Not at present. I think this is the big unresolved challenge of CS teaching – how to deliver what can be a pretty deeply technical subject without producing a cadre of teachers who are also fully-trained engineers (with all the pay and retention headaches that would imply). We're keen to be involved in this conversation, but for now we're restricting ourselves to just providing good-quality teaching support material to help give non-specialists the confidence to deliver our material.
The H: Back in my day, we were taught CESIL as a first "machine code"; making the assumption that trying to use ARM as a first "machine code" would be, shall we say, difficult, is anyone creating a modern CESIL so children can really learn the basics of coding?
EU: Okay, I'd not actually come across CESIL until you mentioned it here. That's a very neat idea, and perhaps we should grab at least the Java command-line implementation from http://www.obelisk.demon.co.uk/cesil/ and put it into our standard install. My personal view is that there's a case for steering kids towards high-level languages for the first couple of years, and then throwing them directly at ARM assembler (which in the grand scheme of things is a pretty benign assembler) once they've naturally come up against the limits of what you can do with interpreted and compiled languages. One thing that's going to reduce kids' interest in assembler is the presence of good compiled languages. Back in the BBC Micro days there was such a massive gap between what you could accomplish in BASIC and what you could do with even a little bit of 6502 assembler; I'm not sure the gap between C and ARM assembler is so compelling.
The H: One worry I've heard is that the Pi is a bit too exposed for classroom use, and although there are cases around, most of them wrap the Pi up and expose the ports leaving the need for a USB hub and its power supply, a power supply for the Pi, keyboard, mouse still needed. Recalling the BBC Micro and Commodore 64 era, they packed (almost) everything into a single box. Are there any plans for something like that for the Pi or are you hoping the Picosystem will provide?
EU: First off, I love "Picosystem"; might have to start using that myself. We're definitely going to be producing a cased version for education, but I suspect that an Amiga-style integration with a keyboard may be a bridge too far for us at the moment. What we may do is offer a unit with a hub integrated in the case to reduce the number of boxes on the table by one.
The H: Thanks Eben, it looks like you have a busy 2013 lined up.