O'Really?

December 22, 2014

2014 vs. 1964: Numbers speak louder than words

It’s that time of year when people look back at over the year that was 2014 (1-5). The place where I work, celebrated it’s 50th anniversary. Colleagues put together a little booklet of facts and figures with an some accompanying web pages to mark the occasion. My personal favourite factoid compares computing in 2014 with 1964. The Atlas Computer represented the state of the art in computing in 1964, and today that crown is held by SpiNNaker – a very different kind of computer.

fifty years of computing

50 years of computing (and pipe-smoking is lesson common around computers)

Sometimes, numbers speak louder than words, so here is a numerical comparison of Atlas (1964) with SpiNNaker (2014):

Feature (see this) Atlas Computer (1964) SpiNNaker (2014)
Size A very large room 19 millimetres square
Transistors 60,000 1,100,000,000
Instructions per second 700,000 3,600,00,000

One way of looking at this data is to say, based on the the instructions per second, SpiNNaker is around ~5000 times faster than Atlas. But what is probably more interesting is that SpiNNaker (which is due for completion in 2015) is expected to be used by neuroscientists and psychologists, as a platform to study problems such as Alzheimer’s disease – something that would have been impossible (and unthinkable) only fifty years ago [6,7]. Wonder where the next 50 years will take us in 2064?

References

  1. Anon (2014). The most-read Nature news stories of 2014 Nature DOI: 10.1038/nature.2014.16550
  2. Morello, L., Abbott, A., Butler, D., Callaway, E., Cyranoski, D., Reardon, S., Schiermeier, Q., & Witze, A. (2014). 365 days: 2014 in science Nature, 516 (7531), 300-303 DOI: 10.1038/516300a
  3. Anon (2014). 365 days: Nature’s 10, Ten people who mattered this year. Nature, 516 (7531), 311-319 DOI: 10.1038/516311a
  4. Katherine Maher (2014) What did the world make 100 million edits of in 2014? Wikimedia blog
  5. Hand, E. (2014). Comet Breakthrough of the Year + People’s choice Science, 346 (6216), 1442-1443 DOI: 10.1126/science.346.6216.1442
  6. Furber, S., Galluppi, F., Temple, S., & Plana, L. (2014). The SpiNNaker Project Proceedings of the IEEE, 102 (5), 652-665 DOI: 10.1109/JPROC.2014.2304638

April 19, 2013

Will Academic Education ever meet the skills needs of the IT Profession? #BCSDebate

“This house believes that Academic Education will never meet the skills needs of the IT Profession” via #BCSDebate

“This house believes that Academic Education will never meet the skills needs of the IT Profession” via #BCSDebate

Here’s an interesting upcoming event: a debate on the motion: “This house believes that Academic Education will never meet the skills needs of the IT profession

Universities are failing to educate graduates with the skills we need – this is the oft heard complaint by employers of IT graduates. Does the problem start in school with the dire state of ICT teaching and assessment at GCSE and A Level? [1] Should academia be trying to produce graduates with only ’employable skills’ that have a shelf life of at best a couple of years? Are employers really expecting universities to produce a mature, rounded professional with 20 years experience straight out of university? Is it reasonable to expect Academia to bridge the skills gap when employers are not prepared to provide a robust career path for IT professionals?

Academia and the IT Profession seem to be out of alignment in a way that other more mature professional career paths are not. Medicine, law, accountancy and the teaching profession provide a clear path from university to the highest levels of those careers – not so in IT. The IT Profession’s skills framework (SFIA) is only a decade old, and IT is neither a regulated or statutory profession – perhaps employers ask and expect too much of Academia, when the IT Profession is still in its infancy.”

This deliberately provocative motion conflates Education with Training as well spreading further confusion about the important differences between Computer Science and Information Technology. There’s already been some debate, including this early response from Ian Sommerville at the University of St Andrews:

“Computing systems are now ubiquitous in all areas of our professional and personal lives – which means that are incredibly diverse from personal apps for your phone to remind you to exercise to safety-critical, world-wide air traffic management systems. The notion that there is a single body of practical skills that is applicable to all of these different types of system is ludicrous as is the expectation that university courses should attempt to cover all aspects of computing practice.”

That’s a view from academia, no doubt employers will probably have a different take on the motion. David Evans and Deborah Trayhurn will be supporting the motion, with opposition from Bill Mitchell and Kevin Jones. Whatever your opinion, the debate takes place on Wednesday 12th June 2013 from 6.30pm – 9.00pm at the Armourers’ Hall, 81 Coleman Street, London, EC2R 5BJ. You can book a place at events.bcs.org/book/577, more info on twitter at #BCSDebate.

March 15, 2013

Creating with the Raspberry Pi vs. Consuming Apple Pie at the Manchester Raspberry Jamboree

MiniGirlGeek

Thirteen year old Amy Mather aka @MiniGirlGeek steals the show at Manchester Raspberry Jam 2013

Last Saturday, the first ever Raspberry Jamboree rolled into town, organised by the unstoppable force of nature that is Alan O’Donohoe (aka @teknoteacher). The jamboree looked at the educational value of the Raspberry Pi (a $25 computer) one year on from its launch on the the 29th February 2012. Here are some brief and incomplete notes on some of the things that happened in the main room, aka ‘Jamboree Central’. The workshops and other events have been covered by Jason Barnett @boeeerb.

A key feature of the Raspberry Pi foundation (and the Jamboree) was neatly summed up by Paul Beech (aka @guru) who compared the Raspberry Pi to various Apple iThings. Paul’s view is that when it comes to computing, Apple gives you a “sandy beach, sunbed and cocktail” to passively consume digital content with while the Raspberry Pi gives you a “desert, knife and a bottle” to actively create new things (see his tweet below).

Consuming Apple Pie on a sandy beach, with a sunbed and a cocktail

Engineering evangelist Rob Bishop used Apple Inc. to illustrate what the Raspberry Pi is about in his talk ‘one year on‘. Rob pointed out that a huge amount of effort at Apple Inc. is put into making Computing invisible and seamless. This is great if you’re consuming content on your iPad or iPhone, and what many users want – easy to use, with all the nasty internal gubbins tucked away, out of sight. This is tasty Californian Apple Pie, which many of consume in large amounts.

However, invisible computing is a problem for education, because it is difficult to demonstrate the Wonders of Computer Science (Brian Cox’s next TV series) with a device like the iPad.  Many of the internals of modern devices are completely inaccessible, and it’s non-trivial for budding young engineers to build anything very interesting with it particularly quickly.

In contrast, the Raspberry Pi can be challenging to setup, just getting the Operating System up and running isn’t always straightforward. However, there’s a ton of interesting stuff you can build with it: Nifty robotics, bionic bird boxes, musical hackery, twittering chickens, live train departure boards, internet radiossinging jelly babies and loads of other pideas. Try doing that with your iPad…

Creating with Raspberry Pi in the desert, using a knife and a bottle

Most of the jamboree focussed not on Apple but on the things that can be created with Raspberry Pi: the What and Why and When And How and Where and Who with keynotes from Steve Furber [1] and talks and panel sessions from:

A highlight of the jamboree was the closing keynote given by the thirteen year old Mini Girl Geek on what she’s been doing with her Raspberry Pi. MiniGirlGeek (aka Amy Mather pictured above) stole the show with her demo implementations of Conway’s Game of Life in Python. [update: see video below]

What’s interesting is that Conway’s Game of Life is used as an exercise for first year undergraduates in Computer Science at the University of Cambridge. So it’s great to see teenagers mastering the “knife” of Raspberry Pi, and reminds us that Raspberry Pi is no “sunbed and cocktail” but with a little patience, ambition and talent there’s plenty to capture the imagination of young people about Computing.

References

  1. Steve Furber et al (2012). Computing in Schools: Shut down or restart? Royal Society Report

July 13, 2012

Animation 2012: Computer Science for Schools

Animation 2012 at the University of Manchester

Computer Science as a subject in mainstream UK secondary education is in a pretty sorry state [1,2,3] but it’s not all doom and gloom. While many long suffering school children are being force-fed a nauseating diet of Excel, PowerPoint and Access others are enjoying a nutritious platter of Raspberry Pi, Hack to the Future and Animated fun.

Here’s a brief report on one of these tasty appetisers: Animation 2012, a UK schools animation competition now in its fifth year.

The day kicked off with prizes being awarded for the animation competition. To get a flavour of the creativity and skill involved, you can see winning examples online.

Following the prize giving there was a carousel of activities which included:

Animation 2012 was great fun for all involved, congratulations to all this years winners, hope to see you again next year. There were 526 Schools involved from across the UK, with 914 entries. 58 students were involved in the 35 winning entries from 31 different schools. Thanks to Toby Howard, all the organisers, supporters (Google, Electronic Arts and NESTA) and associates (Computing at School, CS4FN and BAFTA young game designers) for putting on an impressive show.

References

  1. Steve Furber et al (2012). Computing in Schools: Shut down or restart? Royal Society Report
  2. James Robinson (2011). Eric Schmidt, chairman of Google, condemns British education system: criticising division between science and arts, The Guardian
  3. Keith Stuart (2011). Michael Gove admits schools should teach computer science: education secretary recognises the failings of ICT courses, The Guardian

April 28, 2011

Are machines taking over the planet?

TastyTalk of machines taking over the planet is the stuff of science fiction but if world domination was just a simple numbers game, some machines have already “taken over” from their human masters.

One machine, the particular brand of computer processor found inside all iPhones and lots of other electronic devices, has been quietly spreading around the globe at a phenomenal rate. There are some interesting statistics on just how many of these processors are out there published in an interview with engineer Steve Furber [1]. Here is an excerpt from the interview:

“Around the end of 2007, the ten-thousand-millionth ARM [Advanced RISC Machine] had been shipped, so there are more ARMs than people on the planet. I believe production is currently running at about 10 million a day. It is projected to rise to about one per person on the planet per year within two or three years”.

Those numbers highlighted in bold (emphasis mine) are completely mind-boggling. As humans, we are outnumbered by just one brand of machine! Of course, they are just lots of “dumb” computer chips with no intelligence. But Furber suspects that:

“there’s more ARM computing power on the planet than everything else ever made put together” [1]

So if you could find a way of using all these processors at once, maybe they’d become magically self-aware in a neural network [2,3,4,5]? Cue ominous Terminator theme tune

References

  1. Jason Fitzpatrick (2011). An interview with Steve Furber Communications of the ACM, 54 (5) DOI: 10.1145/1941487.1941501 (since 2007, numbers have risen to 10 billion in 2008 an another one billion in the first quarter of 2011 alone!)
  2. Steve Furber (2011). Biologically-Inspired Massively-Parallel Architectures: A Reconfigurable Neural Modelling Platform Lecture Notes in Computer Science, 6578 (2) DOI: 10.1007/978-3-642-19475-7_2
  3. Steve Furber, & Steve Temple (2008). Studies in Computational Intelligence Computational Intelligence: A Compendium, 115, 763-796 DOI: 10.1007/978-3-540-78293-3_18
  4. An estimated one million ARM processors give you about 1% of the capacity of the human brain see the details of the Spiking Neural Network Architecture (SpiNNaker) project
  5. James Cameron, et al (1991) Terminator 2: Judgment Day (T2)

[Creative commons licensed picture of Terminator terror by Tasty by cszar]

Blog at WordPress.com.