UK Government: Children to be taught programming

[Pickle]

I think the RasPi could be the perfect device for learning modern opengl programming. Granted the RasPi is going to be good for any programming based on size, price, and availability. But the inclusion of a gpu in the soc makes opengl interesting too.
I think it would be neat to have (or create ) a set of tutorials based on the RasPi for gles/egl.

Ive always been interested in computers since I had my c64/128. I would have loved to have learned about how I could program these machines myself but never had the tools or background (nor anyone around me) to learn program. I didnt know I could have programmed them. I recall doing some basic in middle school but nothing substantial.
I didnt really get into programming until college where i started with C. Then some C++. At the that time it was simple microchip and command line code. I thought it would be awesome if I could make windows guis's (I wish I knew how simple it was with tools like visual studio).
So I think I really improved on programming when i started my first position. I self taught myself Visual Basic/C++ and refined my skills with C/C++.
Then I discovered the open source handheld called the gp2x. This was a handheld that ran linux that I could program. I began learning SDL and also helped improve my C/C++.
The open source handhelds of today now incorporate gpu's capable of subsets of the opengl spec. OpenGL has been on my list of dream goals of programming. Always seemed like it was so far away.
I began with deciding I would port a simple opengl app. My first was Ken's Labyrinth. Soon I was able to tackle commercial titles like Quake/Gish/Lugaru. In the last month(s) I started writing my own 3d game from scratch. Not only did i learn fixed but I was able to learn programmable on top of it. (thanks OpenGL Bible )

Reason for the little backstory....what if I would have access to something like the RasPi when i had my C64. And I had access to information that informed me that I could program and do what I want with the machine. I think that idea is profound and why this project really could be epic. Whats even more awesome is the idea that I could pass on what I learned late onto those younger than I was giving them the jump to get ahead. It would cool to see what things could happen after that.

[jamesh]

Quote from Pickle on September 22, 2011, 22:50
I think the RasPi could be the perfect device for learning modern opengl programming. Granted the RasPi is going to be good for any programming based on size, price, and availability. But the inclusion of a gpu in the soc makes opengl interesting too.
I think it would be neat to have (or create ) a set of tutorials based on the RasPi for gles/egl.

Ive always been interested in computers since I had my c64/128. I would have loved to have learned about how I could program these machines myself but never had the tools or background (nor anyone around me) to learn program. I didnt know I could have programmed them. I recall doing some basic in middle school but nothing substantial.
I didnt really get into programming until college where i started with C. Then some C++. At the that time it was simple microchip and command line code. I thought it would be awesome if I could make windows guis's (I wish I knew how simple it was with tools like visual studio).
So I think I really improved on programming when i started my first position. I self taught myself Visual Basic/C++ and refined my skills with C/C++.
Then I discovered the open source handheld called the gp2x. This was a handheld that ran linux that I could program. I began learning SDL and also helped improve my C/C++.
The open source handhelds of today now incorporate gpu's capable of subsets of the opengl spec. OpenGL has been on my list of dream goals of programming. Always seemed like it was so far away.
I began with deciding I would port a simple opengl app. My first was Ken's Labyrinth. Soon I was able to tackle commercial titles like Quake/Gish/Lugaru. In the last month(s) I started writing my own 3d game from scratch. Not only did i learn fixed but I was able to learn programmable on top of it. (thanks OpenGL Bible )

Reason for the little backstory....what if I would have access to something like the RasPi when i had my C64. And I had access to information that informed me that I could program and do what I want with the machine. I think that idea is profound and why this project really could be epic. Whats even more awesome is the idea that I could pass on what I learned late onto those younger than I was giving them the jump to get ahead. It would cool to see what things could happen after that.

I agree. I think the OpenGL stuff is a very real plus point for teaching. 3D is where it's at, and even though I work on the Videocore on a daily basis, I know nothing about Open GL (I work on cameras at the moment) - so I would be well up for a decent OGL tutorial!

Eben on the other hand, wrote most of the OpenGL code, so he knows a lot about it, so should be writing the tutorial!

[tangobravo]

Back on the topic of programming in schools, it was GCSE DT that allowed me to do real programming - a very open systems and control course that was basically "design and make something".

I used to play percussion in a school orchestra, which often involved minutes of doing nothing before a cymbal crash or something, and it was quite easy to lose track. My solution was a device to help show musicians their place in the music. It featured a little 2-line LCD display that could show custom messages at certain points in a piece of music. The second line always showed the current bar number, and a count-down to the bar when the message on the first line was set for.

The main logic was on a Basic Stamp 2 (programmed in a BASIC-like language). There was a serial port connection to link the device to a PC with a visual basic interface in order to input pieces. The PC side had a drag-and-drop interface with the ability to position pauses, repeats, messages, and time signature changes in a simplified music notation. The device could be reset via the serial port, and for the first second or so waited for the PC to send new pieces to it.

The other part of the puzzle was how to input the speed at which the orchestra were playing. I had a switch input that went to a PIC chip, the PIC programmed with another high level language called JAL (just another language) which enabled interrupt routines to be written. The PIC timed the switch presses, and if the user stopped pressing the switch, the PIC would continue feeding the main Basic Stamp with an average of the speed between the previous few presses. The interface between the 2 was just a toggled output on each beat.

The enjoyment I got from that project was what set me on the road I've followed since. It exposed me to data sheets, PCB design, multiple languages, interfacing between different microcontrollers and microcontrollers and a PC. It was the openness of the course and the teacher that enabled that to happen - but it didn't work for everyone. The projects had to be self-motivated, I remember a "Polo Dispensing Machine" which at the end of two years was a wooden ramp with a polo-tube-sized hole in the bottom.

[crundy]

Quote from barnaby on September 22, 2011, 17:03
I was surprised recently because my Nephew (doing his GCSEs) said he has a project for school which involves them learning some Python

Nice! That would come as a surprise to me too, having just finished. Did he do a full GCSE in IT?

I stand corrected, he's just started his A-Levels (or AS levels or whatever the hell it is those crazy kids do nowadays). Not sure what the actual subject is, but apparently they're learning exactly how eveything in a computer works (including the difference between stack / heap memory, useful for OO languages) and a few programming languages, which includes python, .NET and I think he said java and C++. If that course was available when I was at school I would have taken it defo!

[crundy]

Quote from jamesh on September 23, 2011, 09:27
I agree. I think the OpenGL stuff is a very real plus point for teaching. 3D is where it's at, and even though I work on the Videocore on a daily basis, I know nothing about Open GL (I work on cameras at the moment) - so I would be well up for a decent OGL tutorial!

Eben on the other hand, wrote most of the OpenGL code, so he knows a lot about it, so should be writing the tutorial!



I know it's for Android, but this book is absolutely brilliant for learning game development in general and teaches OpenGL. It's worth running through the book to learn how game development and OGL work and then moving to another platform, because other books just don't teach it properly.

[max1zzz]

I so hope this actually goes through (and that im given the option to drop this pointless BTEC) i can't stand to do another IT gcse that i never finish (because i get board and start writing batch scripts or messing with HEK) but knowing my school we will get told "No, you can't do that blah. blah .blah BTEC looks good on your cv blah. blah. blah. that looks bad to ofstead, which makes the school bad (even though we have a bad reputation with the community, and no one takes any notice of our good ofstead report) blah. blah. blah.

[barnaby]

a very open systems and control course that was basically "design and make something".

Oh, that sounds fantastic. In my resistant materials course we were constricted to following a brief because "that's what you do in real life". Fortunately I still managed to make something useful.

Not sure what the actual subject is, but apparently they're learning exactly how eveything in a computer works (including the difference between stack / heap memory, useful for OO languages) and a few programming languages, which includes python, .NET and I think he said java and C++.

*drool* (apart from .NET, which I would resist being taught

That sounds great. Good for him! If his school is sensible to run courses like that, perhaps they'd be the type who would put raspis to good use.

Thanks,
Barnaby

[stuporhero]

Don't resist the urge to learn any language. You can't judge without learning it!
And learning one language can show light on a technique you could use on other languages (thank you "Dive into Python" for object orientation!)

And remember - a language is a language. Programming is like talking: You learn to do it once, it's the languages you add later on in life! etc. etc.

[rjw57]

Quote from tangobravo on September 16, 2011, 08:28
The "C++" course (mainly C really, OO isn't touched on) is only 4 weeks with 2 hours a week, so there really isn't time for people to get comfortable with the concepts if they've never come across them before.

I've supervised that course. It isn't really C++. It is a Pascal course that was converted to C. Then the printf's were replaced with std::cout.

[barnaby]

Don't resist the urge to learn any language. You can't judge without learning it!
And learning one language can show light on a technique you could use on other languages (thank you "Dive into Python" for object orientation!)


True, and I agree — learning objective c has changed how I treat JavaScript and php. Learning assembly has changed the way I understand computers! But I have no use at all for .NET, and I wouldn't be able to do anything with it anyway, on a mac.

In terms of a school course, I think cross platform languages should be used exclusively (at least to teach the fundamentals of programming), with options to use other, platform-specific languages of the students choice at a higher education level. When the principles are more abstract and the student is more independent they should have the choice of language, platform and IDE.

That's how I'd do it, anyway.

Cheers,
Barnaby

[Soul_Est]

I truly enjoyed reading through this thread and though I wish I could add something useful or enlightening but I can't.

I first learned to program in Grade 11 using batch files and then Turing in DOS 6.22. All this was housed on a single floppy disc. The Grade 12 course got cancelled since there were only four people (myself included) in the course. College was worse for me by comparison.

The school that I went to was RCC Institute of Technology for their Computer Networks Engineering Technology program. To say it was a disaster was putting it lightly. The network was the worst of any post-secondary institution I've been in as was the IT department. The labs were lacking in equipment to say the least. The program was mostly Microsoft based since 'that's what you will find in most SMBs'. Database classes were done on SQL Server 2008. Windows classes spanned three semesters. The programming courses (C++, Java, Assembly) went by too quickly which it hard for me to keep up. The MCUs used in the Assembly classes were locked down to a Windows only IDE (No Arduinos there!). Having a Mac was also a no-no which sucked for me (got picked on a lot). The school also ended up buying the International Academy of Design Toronto school and giving them a better network and better equipment and had them pay less tuition. My class ended up paying more for our final year. In the end, I lost interest and hope. I didn't get the 'advanced degree' and so far have yet to find anything in the field which I have studied for. Moral of the story: Don't go for a program just because they say you can finish school and get into the field sooner. Do your homework.

@obarthelemy
I really like how the French have set up the technical courses in their system. Open and free. Has me wondering how Finland, Sweden or Norway have their systems implemented especially with such great companies having been born in those places.

[andywe]

Question - how can a very complex processor with a lot of proprietary media processing silicon be used to teach UK Children programming ?
I'm assuming that when the boards eventually appear there will be various complex APIs for driving the (admittedly quite impressive sounding) array of media processing hardware on the BCM2835
To quote
"Dual Core VideoCore IV® Multimedia Co-Processor
1080p30 Full HD HP H.264 Video Encode/Decode
Advanced Image Sensor Pipeline (ISP) for up to 20-megapixel cameras operating at up to 220 megapixels per second
Low power, high performance OpenGL-ES® 1.1/2.0 VideoCore GPU. 1 Gigapixel per second fill rate.
High performance display outputs. Simultaneous high resolution LCD and HDMI with HDCP at 1080p60"
Most of this technology is closed source ... with some "open-ish" driver layers ...
What are UK schoolchildren going to be shown to do with all this "wonder technology" ?
For ideas maybe we should be looking at was has been already achieved technologically with related projects such as e.g. Gumstix, Armadeus, Beagleboard, Pandaboard, Tin Can Tools' Hammer, ...
From the point of view of taking advantage of media processing wizardry then I would guess examples should be learned from iPhone, Android and, possibly, Microsoft's nth attempt at getting into the mobile telephony market.
iPhone, Android and Microsoft Mobile Telephony offering application development involves mastering quite complex frameworks ... and their associated technologies and languages ... MVVM, XML, XAML, CocoaTouch, Objective C, Java, C#, RESTful clients and servers as well as a whole raft of web technologies (Ajax, JQuery, CSS3, HTML5, Silverlight .....).
How will these be integrated into a suitable curriculum, or skills development program ?
Should the RaspberryPi be a "case study" of what is possible ? albeit one that, apparently, is generously supported by Broadcom.
From the point of view of developing future programmers ... what about incorporating an FPGA into the RaspberryPi ... ?
I'm already "tinkering" with many of the above embedded Linux boards ... and others as well ....
They are more expensive than the RaspberryPi but have reasonable support for connecting peripherals ... and various addon boards are available for them also ...

I would hope that the RaspberryPi, too, will stimulate a little ecosystem of addon developers ... and, hopefully, some of these will actually be designed and made in the UK ...

[liz]

In exactly the same way a very complex ink, made from proprietary ingredients, can help teach children how to write.

Kids can use this platform to run software like Kids Ruby, Game Maker, Scratch, PyGame and all that good stuff. Best of all, they can own it, keep it in a pocket, and parents don't have to worry too much about the cost of a broken one. The excellent multimedia is a bonus; and it's a good way to get kids who don't know yet that they have a facility or interest in programming to want to own a Raspberry Pi anyway.

We've said before that we're not writing a curriculum; we're providing a tool. Things are already looking up with the new Computing GCSE (and we were talking to the head of that exam board just over a week ago); there is also a lot of educational software being written for and ported to the device (not by us, which just underlines what I said about providing a tool). After school clubs, self-directed learning and some fun stuff with parents are also important.

[scep]

Quote by a person on the Clapham omnibus, today
Question - how can a very complex processor with a lot of proprietary media processing silicon be used to teach UK Children music?

I'm assuming that when the keyboards eventually appear there will be various complex APIs for driving the (admittedly quite impressive sounding) array of media processing hardware on the Yamaha EZ-200.

To quote:
61 regular-sized touch sensitive keys
Number of Polyphony (Max.): 32
Tone Generating Technology: AWM Stereo Sampling
Style Control: INTRO, MAIN x 2, FILL-IN, ENDING
Tempo Range: 11-280
Transpose: -12 to 0, 0 to +12
GM: Yes
XGlite: Yes

Most of this technology is closed source ... with some "open-ish" driver layers ...

What are UK schoolchildren going to be shown to do with all this "wonder technology" ?

Ummm... learn to read music and how to play simple tunes, with a minority going on to become session musicians, concert pianists, music teachers, band members, keyboard designers, jingle writers etc? The rest will have to make do with having been exposed to one of humankind's greatest accomplishments.

[Bakul Shah]

A relevant article in the Graniuad: http://www.guardian.co.uk/educ.....-computing. I am afraid the situation is not better in the US schools.

I like "anarcho-capitalism" : )

[andywe]

Quote: In exactly the same way a very complex ink, made from proprietary ingredients, can help teach children how to write.
I like the analogy ... but am not so sure ... you can write using chalk on a blackboard, and complex coloured inks are great for artists, or those who write in purple or sepia ...
Writing is a means to an end ... its quite a focused skill ...
Writing computer programs is much, much harder ...
There are graphical programming tools such as LabView (and its Lego Mindstorms incarnation)
and Risbee from POB Robotics in France
There are also promising tools for Mind Mapping and Concept Mapping ...
These have all been around for quite some time and run on the usual operating systems ...
They have not caused a "revolution" in either teaching or inspiring youngsters
Over the many years I have been involved in running programming courses it has been very rare for the number of women in a class to exceed 10% ...
Courses attracting larger numbers of women have tended to be web design courses - mostly of the HTML and CSS variety
With PHP courses the percentage reverts to 10% ...
I hope that developing of "Interactive" software might be more appealing ... and certainly have been gratified by female artists interested in attending Arduino workshops ... though, here, the difficulty arises when trying to create original programs, rather than simply midly tinkering with existing sketches ... I'm afraid that there is not getting away from having to master C and microcontroller principles to a reasonable level in order to be independently creative ...

I have taught maths and science (Physics, Chemistry and Biology) at GCSE and A level and have thought quite hard, on and off, about use of computer technology in science teaching ...
I can see potential uses of relatively cheap and relatively powerful embedded systems but nothing truly innovative comes to mind ...

I've also worked on developing teaching modules involving PyGame and Scratch ... even, at one stage working (at his mum's behest) with a very hyperactive teenager with a passion for developing computer gaming software ... Scratch was not a great success, even though I am a fan, Stickman animation was much better received ...

Quote 2:
Ummm... learn to read music and how to play simple tunes, with a minority going on to become session musicians, concert pianists, music teachers, band members, keyboard designers, jingle writers etc? The rest will have to make do with having been exposed to one of humankind's greatest accomplishments.
.. Touche ... but not quite what the "blessed Cameron" and his "acolytes" have in mind, I think ...
My impression is there are after gifted and inexpensive "techies" to be used by "business types" ...

Quote 3. A relevant article in the Grauniad: http://www.guardian.co.uk/educ.....-computing. I am afraid the situation is not better in the US
schools.

In the case of IT this may very well be true ... in the case of Robotics I do feel that there are some very good initiative in the US ... e.g. Carnegie Mellon's one week courses for teachers showing then how to use robotics technology in STEM teaching ... mostly oriented towards either VEX or Lego Mindstorms systems. At a recent UK Lego Mindstorms finals (I was told) the greater majority of the teams were from Public Schools / Grammar schools with hardly any from Comprehensive Schools ..
I don't think that this is because Comprehensive schools are particularly academically poor ... its just that teachers have to cope with a much wider range of abilities and lots of administrative and reporting demands on their time ..

Quote 4: there is also a lot of educational software being written for and ported to the device (not by us, which just underlines what I said about providing a tool). After school clubs, self-directed learning and some fun stuff with parents are also important.
- Getting standard open source Linux applications to run on the RaspberryPi (providing they are not to resource hungry) is no big deal ...
- I would love to hear about really well attended after school clubs using the RaspberryPi when it is eventually available in large quantities ... but I think it will be a while yet ...

Quote 5: We've said before that we're not writing a curriculum; we're providing a tool.
My problem is "what kind of tool" precisely - other than a very cheap embedded Linux computing platform (which is a good thing, and no mean achievement) ...
However, I ask, how will this "tool" lead to innovations in "teaching kids to program" ...
Undoubtedly part of the answer lies in establishing a very active community and a social network ... which I think RaspberryPi is doing well, and I think the very low price helps also ...
One of the nicest comments I heard about the Arduino community (from Mike McRoberts who runs Earthshine Design) is that it is a friendly community and when the same "basic question" gets asked by yet another novice, instead of a "put down" there is a helpful and considered reply ...

[gladoscc]

Problem: Not everyone enjoy programming

[scep]

Not everyone enjoys history/maths/music/languages/etc. Should we not teach these things then?

The programming thing is a bit of a straw man though. The aim of the Raspi Foundation is to promote the study of computer science and related topics, especially at school level, and to put the fun back into learning computing. Computing is not programming.

The ultimate goal would be to get computing (as in computational thinking) on the curriculum from a very early age - and to give all pupils experience and exposure to computing throughout their education - so that they can make considered decisions when it comes to choosing what they study at GCSE and beyond. And to give them an invaluable life skill on the way. And who knows - they may even enjoy it

[andywe]

Quote: promote the study of computer science and related topics, especially at school level, and to put the fun back into learning computing.

Computing is not programming.

Point taken ... but I can't let you off the hook so lightly ...
Computer science does involve programming, and logic and maths and ...
Most computer science degree courses do very little as far as embedded systems are concerned ...
Thinking of fun as "having a lot to do with play and discovery and exploration" ... then the mobile telephony and social networking and computer game playing areas come to mind ... (and, of course the Arduino ... )
Here most of the fun seems to be in making use of the technology rather than coming up with new products, ideas, paradigms ...
Maybe the RaspberryPi could be morphed into a super cheap communications and presentation device ... ?
There have been quite a few attempts at doing similar things e.g. with gumstix and OpenMoko ...

[scep]

Quote from andywe on December 15, 2011, 09:43Thinking of fun as "having a lot to do with play and discovery and exploration" ... then the mobile telephony and social networking and computer game playing areas come to mind ... ... The average teen uses their phone mainly for texting and media play. This has very little to do with play or discovery or exploration. As for social networking being these things, the FB meme of 2011 was "lms". Enough said.

Here most of the fun seems to be in making use of the technology rather than coming up with new products, ideas, paradigms That's the problem (though I'm not sure that "fun" is the right word). We need active creators - -not passive, slack-jawed consumers. Tools like the RpPi are not a panacea, but they can only help in the push towards a much needed sea-change.

[andywe]

Quote: That's the problem (though I'm not sure that "fun" is the right word). We need active creators - -not passive, slack-jawed consumers.
Maybe "fun" is the right word ...
In 2009 I ran an Arduino all day workshop and UKUUG, and in 2010 a similar Arduino workshop at Europython 2010, but with some Python thrown in ...
Both workshops were full ...
It was such a great pleasure to see so many "hardened software engineers" having literally "so much fun" ...

[scep]

The Arduino is fun, yes - and I like it a lot - but you can hardly build a curriculum around it. We're talking about massive, long term, profound changes such as teaching Computer Science in the primary years and rewriting the National Curriculum - hopefully supported by cheap, general purpose PCs like the RasPi - and you keep banging on about an electronic prototyping platform.

[andywe]

Quote: (re Arduino) but you can hardly build a curriculum around it....
I take your point, however, there are a number of interactive art degree courses where a large part of the course is built around Arduino and Processing (the IDE) ...
If you look at the Robotiq web site ... navigate to the LeTry section ...
I have failed to get much enthusiasm at the school level, but several parents have taken it to heart ...
The notes and teaching materials need improving and I am working on these ...
I have am also working on teaching modules where the processor is an Arduino Nano board ... which does fit onto the breadboard plus HBridge to drive the motors ... (you need to use the version with two small breadboards) ... the LeTry is used widely in Taiwanese and Far Eastern schools ... is cheap, and effective ...
Quote: You keep banging on about an electronic prototyping platform ..
If I have caused offence, it was not intentional ... without inputs and outputs most computing devices are very limited ... with the development of all kinds of new sensors many interesting applications become possible ...
IMHO if the RaspberryPi becomes just a very cheap Linux PC with a small form factor it will achieve very little ... In order to be interesting computer systems need to be more interactive .. in order to do this they need to acquire enhanced sensors .. and enhanced "self organisational" capabilities ...
This raises all kinds of moral and ethical questions (potentially as these technologies are still very much in their infancy ... )
Do you have any suggestions for a Computer Science curriculum for primary school, secondary school and beyond ...
I have been toying around with a variety of approaches ...
Maybe this is not the forum in which to discuss such topics in detail ...
Would you like me to set up a forum, do you have any outlines of what you would consider an interesting starting point for say a primary school computer science curriculum, or a secondary school computer science curriculum?
Have you looked at any of the recent articles published by the ASE (Association for Science Education) on teaching science at primary school level, and on the use of computers in this context ?
Have you sketched out any thoughts on cross-curriculum aspects of computer science teaching ?
I am a bit wary of "National" and "Curriculum" in close proximity ... especially after over a decade of "pronouncements by the blessed Blair" and "his associates" .
My guess is that curricula should evolve .. and that means that, sometimes, worthwhile attempts, such as e.g. the Nuffield projects on maths and science teaching ... fall short of what they set out to achieve ...
Talking of "visionary aspirations" ... I have a great deal of sympathy with Margaret Boden's vision of computers as a tool for building models and knowledge exploration see e.g. "Artificial Intelligence And Natural Man" ... and as means for stimulating variety and diversity as opposed to 'a means of generating "wealth" in the "banking sense" and imposing "conformity" ... '

I would hope that a modern computer science curriculum would include extensive sections on such aspects as privacy, civil liberties, controversies regarding software patents and, human rights ... areas that can be severely harmed by "misuse" of computer technology ...
I would also like for subjects such a bioinformatics and semantic web technologies to be included in such a curriculum ...

I still, like most contributors to this forum, have to get my hands on a RaspberryPi device,
in the meantime, my experiences and the ideas that spring from them are based on
- working with Beagleboard, TinCanTools hammer, and related devices
and teaching courses that use these devices .. and via this means gaining some small appreciation as to their capabilities and limitations ...

[jamesh]

Quote from andywe on December 15, 2011, 20:24
IMHO if the RaspberryPi becomes just a very cheap Linux PC with a small form factor it will achieve very little ... In order to be interesting computer systems need to be more interactive .. in order to do this they need to acquire enhanced sensors .. and enhanced "self organisational" capabilities ...


I'm not sure that IO is necessary to make a computer interesting. Maybe to you, but not to many others It's quite possible to do interesting things on a Linux box. Adding GPIO's just make it more interesting, to some people.

That said, Gertboard should make GPIO and hardware control much easier to use.

[andywe]

I was thinking about much more than GPIO ... being able to turn a few relays on and off, to flash some LEDs in pretty patterns is pretty standard stuff ... you don't need a Gertboard to do that, even though the design looks cute and it will be fun to tinker around with it ...
I was thinking along the lines of the whole plethora of I2C based sensors, of USB based widgets, of internets of things, of NFC, Bluetooth .. and beyond ...
Look at the ePuck for example ... to see what's already been possible for quite a few years now ...
Look at the CMUCam ...
Look at Microchip's relatively cheap PIC32 starter kits and the various addons for these kits ...
Now 6lowPan and nice connectors on a RaspberryPi .. that would truly be something for the future ...