Depending on your perspective, with recent developments, it’s tempting to think that Net Neutrality has debate has changed radically
But in reality, in the background, the Net Neutrality debate may be maturing
Here is why
I have attended IMS conferences for years and they have been predictably boring
Mostly we see the same faces and the same companies (mainly Operators) who complain of the same issues (OTT).
Absent from the debate are any real web companies
So, I was very surprised to see facebook and Mozilla speaking at the IMS world conference
This is a good development
To me, it shows that behind the rhetoric – we may be actually engaging in real dialogue to create useful services
webRTC plays a big role here – and I am a fan
And the interesting aspect is both facebook and webRTC are platforms
Which is good for wider innovation i.e. not just the players themselves
The sessions are
A NEW WAVE OF COMMUNICATION SERVICE INNOVATION:
14.00 IMS and WebRTC: What does it mean and where is it going?
09.35 WebRTC: Because the web is the platform
• How browsers and operators can work together
• End user value of real-time communications
• P2P and privacy aspects of WebRTC
• Case study of how firefox can be integrated into IMS networks
Jean-Baptiste Piacentino, Firefox Director Browser Communications,
12.05 IMS and RCS integration and next generation carriers expect from Facebook.
• How can we evolve new architectures?
• Is Facebook really an OTT threat?
• Future deployments
Graham Bright, Mobile Partner Engineer, Facebook, Ireland
We are also seeing a more complex discussion ex Has the FCC Chairman Solved the Net Neutrality Quagmire?
All these are good developments and in the end reflect a growing reality that everyone is ultimately focused on serving their customers in the best possible way
This is excellent and I recommend you read it
The concluding para
One major test will begin on Feb. 7 at 11 a.m., EST, when Hola’s users will try livestreaming the Olympic Opening Ceremonies from a variety of foreign websites, bypassing the vast system of electronic borders seeking to hem them in. Those who succeed will be watching up to eight and a half hours ahead of most Americans, since NBC is declining to carry the ceremony live on TV or to stream it at all, presumably to boost broadcast ratings in prime time.
There will be pageantry and spectacle and most likely a flurry of bizarre and over-the-top performances. Then the Olympians will parade around Fisht Stadium, a stirring display of athletic excellence and a testament to the human spirit, putting aside ethnic and national differences and coming together for a common purpose, borders be damned.
I am attending the state of the net conference in Washington.
If you are attending – happy to meet. Please contact me at ajit.jaokar at futuretext.com
This is a not your usual anti-facebook scare mongering
The very first sentence is a classic and sums the entire problem “The problem with facebook is – that it is keeping things from you”
and then later “facebook is using it’s filtering power to make money”
and later again “we are all advertisers”
I agree with this analysis .. and my page has more than 40,000 fans – Concepts of Programming Languages .. I hope someone at FB is listening to this
We used to advertise on FB .. but not recently ..
Ovum claims predictibly that Operators’ M2M opportunities revenues will more than treble over the next five years
This is interesting since its about M2M and not IOT
‘M2M’ (machine to machine). M2M is mainly a Telecoms industry term.
M2M implies a ‘machine + radio’ at least at one end with a cellular modem.
IOT has a wider goal of making intelligent devices using radio technologies (but not necessarily cellular).
IOT is expected to be in the range of 50 billion devices by 2020
By 2020, we are expected to have 50 billion connected devices
To put in context:
The first commercial citywide cellular network was launched in Japan by NTT in 1979
The milestone of 1 billion mobile phone connections was reached in 2002
The 2 billion mobile phone connections milestone was reached in 2005
The 3 billion mobile phone connections milestone was reached in 2007
The 4 billion mobile phone connections milestone was reached in February 2009.
So, 50 billion by 2020 is a large number
Thus, M2M is a subset of IOT(Internet of things).
For example healthcare has the following wireless technologies – Bluetooth, MICS, 802.11, IEE 802.15.4, 802.11e, Zigbee, Whitehart, Wibree, ISA100.11a, 6LoWPAN , 6lowpan, 3G/GPRS/GSM.
Cellular is just one of them ..
Now, what is Ovum claiming - Ovum outlines operators’ M2M opportunities as it forecasts revenues will more than treble over the next five years
a) Over the next five years, M2M revenues will grow to reach $44.8bn, with more than a third coming from Asia-Pacific.
b) New forecasts* from Ovum reveal that revenues will grow slightly more slowly than connections, reflecting the increasing competitiveness of the market and the extension of M2M into lower-value applications.
c) Total global M2M connections will more than treble from 106.4 million in 2012 to 360.9 million in 2018, at a CAGR of 22.6%.
d) There will be growth across all regions, but it will be fastest in Asia-Pacific and the Middle East and Africa. Revenues in Asia-Pacific will grow to almost $15bn, at a CAGR of 26.5%, between 2012 and 2018.
e) The most important industry verticals in 2018 will be healthcare, manufacturing, and energy & utilities, which are forecast to generate revenues of $7.9bn, $7.1bn, and $7.0bn respectively by 2018.
f) “This is not a forecast of the ‘Internet of Things’ but rather of managed and paid-for connections over public mobile networks,
g) “For telcos there are really two opportunities: to stand back and provide the connectivity for M2M services, or to roll up their sleeves and get involved with the end-to-end provision of solutions.”
h) The first is a smaller opportunity but is much more straightforward for telcos to address. The second is much bigger, but involves new skills and competencies, and defining new kinds of relationship with systems integrators (SIs) and software development.
Most of these reports are written for Telco audiences.
So, they follow a now familiar pattern
a) Show that there is a big market .. (X billion after Y years)
b) Claim connectivity is minimum and thats the low hanging fruit ..
c) Claim services are the key but Telcos need different skills to take advantage of services (or those Big Bad OTT guys will take it all!)
So, no surprises here!
Why Asia Pac? esp with healthcare, energy and utilities driving the uptake?
I am not sure ..
But overall .. a sense of Deja Vu
For real innovation – see IOT, Kickstarter etc
As a child – I disliked IQ tests
Not that I was very bad at them – but because people short-sightedly tend to use IQ tests as measures in themselves for selection and admission
That process often excluded me ..
As an adult – I realise that my unique form of ‘intelligence’ if you will – lies in co-relating existing ideas to create new ones ..
IQ tests cannot measure that skill ..
However, IQ tests are not REALLY a measure of intelligence at all – (despite the views of some who think that they are) – because you can learn to pass these tests ..
In other words, there is an element of skill involved
So, it’s nice that I can use my form of intelligence of ‘hacking IQ tests’ to apply to another domain – i.e. learning programming
Mastering IQ tests is really often a matter of understanding common patterns in these tests
I mentioned in Women and Coding – Do we need a different approach – the idea of learning programming like a ‘Chess master’ i.e. understanding common patterns for programming and understanding how they relate to the big picture
This is a technique we are using in our trials
For an analogy from IQ tests – Joseph Perla’s blog post is a good example – where he identifies common patterns in IQ tests and how they can be solved
So, we teach programming by
a) How to read code (which gives you a big picture)
b) How to identify patterns in code
c) and then how to take a ‘systems view’ through the feynmaps techniques we are developing
Note I am not speaking about software patterns – rather I want to study programs and create common patterns as a way to learn programming..
As usual, comments welcome
Image source – Wikipedia
Over the last year, as part of my edtech start-up Feynlabs – I have been involved in developing new ways to teach Computer Science and Programming.
We have been primarily focussing on kids and schools including teachers.
Everyone agrees it’s a major issue – and a lot is at stake in getting young people to code.
As the father of a very motivated 10 year old who accompanies me on the Feynlabs journey with his own insights such as the lemon Pi – I am personally motivated in this quest.
Over the last year we conducted trials in UK, Europe and USA to develop the Feynlabs methodology – specifically the feynmaps techniques (more on this later).
We also conducted trials with around 400 teachers online – which are still ongoing ..
So, some interesting insights over the last year
An interesting secondary observation of this work was - that teaching women to code may need a different approach ..
I heard this from an educator (and I paraphrase)
If you give a two boys a tennis ball and ask them to throw it against a wall and catch it (one who drops first loses) ..
They will happily play that ‘game’ all day ..
But if you try the same exercise with girls – they will ask –
i.e. Whats the point of this exercise ..
This is actually a profound observation – with some implications for learning to code
Over the last year – working with so many people – we also observe the same.
So, instead of making changing colour/design of sites (ex pink) – how about tying it to a cause?
Hence the image on this blog of the environment (image source – EU Digital Agenda)
So, we are now announcing a brief free online trial – essentially an adaptation of our work in schools – to teach women to code
I will post more on this later – assuming there is interest – but here are some key aspects – all of which we have been trialling over the last year
a) How to read code
b) Understanding Big Picture first
c) Understanding meta patterns – think like a chess master – who sees many ‘matches’ before and can draw common patterns for the next move
d) Coaching and Programming (focussed improvement)
e) Pi (physical computing)
f) the feynmaps techniques– more on this later
g) linking to cause ex environmental issues (as opposed to boys being targeted to competitive etc) – see above
i) Small steps – over a longer duration – 6 months
j) tutor led (one on one)
If you are interested – please email me at ajit.jaokar at futuretext.com
I am very much aware that this issue has been in the news over the last week and I hope our approach – based on our insights from trials – can add value
Later it could be a new program/kickstarter etc – not yet decided – so welcome feedback
I will be at Mobile World Congress 2014. If you want to meet up/feature on my coverage for MWC 2014 please email me at ajit.jaokar at futuretext.com
Here is what I have been on with Feynlabs.
We call it feynmaps – because my original motivation was feynman diagrams
What does feynmaps address
In theoretical physics, Feynman diagrams are pictorial representations of the mathematical expressions governing the behaviour of subatomic particles. In essence, a visual way to represent complex interactions
In the teaching of Computer Science in schools, I wanted a simple, visual way to unify concepts with practice. If we consider the Pi, Arduino etc (doing) .. often we focus on the doing .. but not on the concept.
So, feynmaps is a simple, visual way to represent Computer science interactions to students, to include HW/SW and Algorithms holistically and unify practice with concepts
This is work in progress .. so I will post more about it later. But here are some initial ideas:
1) I am drawing on the ideas and notations from UML state diagrams(more specially Harel state charts). But UML state diagrams are too complex for learning and they focus on precision and completeness of a specification. At a learning stage – we want flexibility, simplicity and ‘back of the envelope’ mechanism
2) So, the idea is to create a simplified state diagram where we are able to model computer science processes across HW/SW/algorithms across the stack and tie the concepts, practice and code together
3) The topmost level would be based on a physical process (ex a
4) The time dimension could be horizontal and the stack dimension could be vertical
5) We want to focus on minimalism
6) Note flowcharts are not the same since they don’t cover a change of
7) The main diagram could be simply two end points node Input objects and output objects – Human, software, sensor, device, algorithm and sets(collections)
8) The lower layers could focus on implementing the concept across the stack
9) The code example could be the lowest layer
10) Schematic example
An analogy in UML
The bank ATM but the model also shows also the processes underlying it ie serving customer is the main process but behind it are others that customers do not see but are useful to model and understand as a system
source UML state diagrams
ex temp sensor can be modelled as an interaction using Pi/Arduino etc and then if statements and for statements which use to poll the GPIO pins as a lower level / composite process
So, my proposed approach is to address from two sides
a) which programatic concepts to model (bottom up)
b) which processes to model (top down)
and then draw it out -
That will be a start.
I will take a back of napkin approach although it will use UML like visuals (subset of UML)
Its a modest start but by releasing in public domain and open sourcing it effectively feynmaps can evolve over time
I am already collaborating with various schools and educational institutions and comments welcome at ajit.jaokar at futuretext.com
To recap, the idea is to represent any Computer Science process from a teaching perspective in a simple, back of the envelope manner and tie hardware, software, algorithms and code in a visual manner
Comments welcome at ajit.jaokar at futuretext.com