Stephen Tyler

The greatest obstacle to discovery is not ignorance – it is the illusion of knowledge.

October 25, 2011
by stephen
Comments Off on John McCarthy, father of AI, has died aged 84

John McCarthy, father of AI, has died aged 84

John McCarthy, inventor of the LISP programming language and a pioneer of Artificial Intelligence, died a few hours ago.

He coined the term “Artificial Intelligence” in 1955, and went on to found departments to study AI at Stanford & MIT.

The LISP programming language is the second oldest computer language still in use.  The oldest, FORTRAN, spawned the procedural languages BASIC, C etc., that are in widespread use today.  Each year brings new variations and derivations.

LISP, however, was different.  It was inherently complete.  The syntax did not need to be altered to accommodate new programming constructs or methodology.  It could be extended, in LISP, to accommodate such developments as garbage collection, object-oriented programming, functional programming, logic programming, type checking, concurrent programming, distributed programming and in fact any new programming methodology.  Indeed, most of these new programming methodologies were prototyped in LISP.

The LISP of the 1960s is immediately recognisable to a LISP programmer of today.  But all the computer languages of today owe many of their features to features that were first developed in LISP.


Skip to 16:37 for an interview with John McCarthy:

October 21, 2011
by stephen
Comments Off on Siri has the competition worried

Siri has the competition worried

The internet is awash with the sayings of Siri, the new voice activated assistant in the iPhone 4S.  But the competitors are also talking about Siri.

Android Chief says your phone should not be your assistant:

“Your phone is a tool for communicating. You shouldn’t be communicating with the phone; you should be communicating with somebody on the other side of the phone.”

Microsoft’s President of Windows Phone says talking to your phone isn’t super useful:

He affirmed that users can indeed talk to their Windows Phone handsets, but that the kind of implementation seen in Siri isn’t “super useful.”  He also — oddly, we must say — noted that WP7’s voice implementations rely on Bing, which harnesses “the full power of the internet, rather than a certain subset.”

[Ed: On Siri, “Bing this” will search Bing for “this”]

So why the public criticism? My guess is that Siri has made both Google and Microsoft very nervous.  Apple has just introduced a new layer that sits between the user and the internet search engines.  Instead of users querying Google (and occasionally Bing) directly, Siri gets to decide what queries are answered by its partners (Wolfram Alpha, Yelp etc), leaving only the untargetted (and less valuable) queries to be passed onto the traditional search engines.  Apple can now grab (and monetise) all the valuable queries for themselves.

Fast Company seems to think so too. “Why Google and Microsoft are bad mouthing Siri”:

Siri could gum up Google and Bing (and Yahoo) ad revenue.

Google, of course, thinks it should be king of all search, which is why it’s been launching its own search facilities for everything from cheap flights to patents, and now Apple’s app (which has earned huge press coverage) has gone and supressed access to its search beneath a friendlier, more user-pleasing interface. … Microsoft is nervous for the same reason–loss of traffic, and thus ad revenues that go alongside this.

Apple’s also now able to gather a huge database about how its users query the web–and it’s a data-rich “semantic query” one, which could let Apple leap ahead in optimizing its own efforts at Net search technology. And it’s not beyond the pale that Apple could find a way to monetize this layer of data and language-based interactivity in the future, possibly tying in to iAd.

If I was Google or Microsoft, I would be worried too.  Something just appeared out of nowhere that threatens to render their technology irrelevant.

October 19, 2011
by stephen
Comments Off on Lemony Snicket Comments On Occupy Wall Street

Lemony Snicket Comments On Occupy Wall Street

Lemony Snicket, author of the popular children’s books “A Series of Unfortunate Events” and a movie of the same name, wrote a series of Thirteen Observations made by Lemony Snicket while watching Occupy Wall Street from a Discreet Distance:

Money is like a child—rarely unaccompanied. When it disappears, look to those who were supposed to be keeping an eye on it while you were at the grocery store. You might also look for someone who has a lot of extra children sitting around, with long, suspicious explanations for how they got there.

The other 12 observations are just as witty and cleverly composed.  Well worth a read.

October 15, 2011
by stephen
Comments Off on Cooking the Books

Cooking the Books

Benford’s Law states that numbers for many real-life data sources have a distribution of leading digits that follow the following equation:

P(d)=\log_{10} \left(1+\frac{1}{d}\right)

In other words, the first digit of an arbitrary number is the digit “1” around 30.1% of the time, with successive digits appearing less frequently, culminating in the digit “9” around 4.6% of the time.

As a simple exercise, consider the numbers from 1 to 20. The leading digit “1” appears 10 times (1, 10..19), the digit “2” appears twice (2, 20), and the other digits appear only once.  As we expand the range of possible numbers, the distribution of leading digits converges on the Benford Law distribution.

Not all real-life sources follow this distribution.  For example, the distribution of eyes per person, or children per family do not follow this distribution because the range of possible values is quite narrow.  But numbers that can vary over many orders of magnitude, such as leaves on a tree, or ants in an ant nest, or income figures for a large corporation, should all follow Benford’s Law very closely.

In Benford’s Law and the Decreasing Reliability of Accounting Data for US Firms, Jialan Wang examined data from 20,000 firm’s SEC filings, with some interesting results:

Deviations from Benford’s law have increased substantially over time, such that today the empirical distribution of each digit is about 3 percentage points off from what Benford’s law would predict.
While these time series don’t prove anything decisively, deviations from Benford’s law are compellingly correlated with known financial crises, bubbles, and fraud waves.  And overall, the picture looks grim.  Accounting data seem to be less and less related to the natural data-generating process that governs everything from rivers to molecules to cities.  …   And it’s just one more reason for investors to beware.
So there is strong evidence to suggest that accounting statements are becoming less reliable over time.  Very scary stuff, and it does not bode well for the current financial crisis.  Wang promises more analysis of the factors affecting deviation from the expected distribution, such as deregulation, changes in accounting standards, and measures of corporate governance.  Stay tuned.

October 6, 2011
by stephen
Comments Off on Steve Jobs has passed away

Steve Jobs has passed away

  • Creator of the personal computer.
  • Creator of the first windowed operating system to reach the market.
  • Creator of the first mouse to reach the market.
  • Creator of the computer-animated feature-length movie.
  • Creator of the portable music player integrated with a music marketplace.
  • Creator of the modern touch-screen smart-phone.
  • Creator of the first popular tablet.

Steve Jobs. 1955-2011.

May he rest in peace.

September 11, 2011
by stephen
Comments Off on Gen Y plays, but Gen X pays

Gen Y plays, but Gen X pays

Flurry Analytics have published an interesting report on the different demographics for App revenue and App usage, “Mobile Freemium Games: Gen Y Plays, but Gen X Pays“:

Players aged 25-54 account for 77% of the revenue, but only 44% of the usage of the mobile apps surveyed. The most lucrative players are actually aged 35-54, generating almost 9x the revenue/usage compared to the 13-17 year olds. I’m not sure how the ages of players were measured, but there is a distinct possibility that many of the players aged 17 and under may have lied about their age, making the real demographic differences even stronger.

But is it sensible to focus marketing efforts only on the 25-54 year olds?  Perhaps not.  Although younger players are not contributing much revenue directly, they may serve an important role in the game.  These time rich players can provide the man-power to construct and populate the online-world, creating a more interesting environment for the time-poor but more lucrative older players.  But they could also be turning the older players away, if trash-talking and anti-social behaviour is allowed to flourish.

Anyway, the survey completely demolishes the view that gaming is only for the young.

September 11, 2011
by stephen
Comments Off on Moral hazards of plagiarism detection

Moral hazards of plagiarism detection

In my earlier post, Unintended consequences of plagiarism detection, I discussed the relative merits of allocating scare resources to pursuing bad students, versus enriching the learning environment for the good students.

David Harrington has written an interesting article, (Moral) Hazards of Scanning for Plagiarists: Evidence from Shoplifting:

Turnitin offers another product called WriteCheck that allows students to “check [their] work against the same database as Turnitin.” I signed up and submitted the early pages of Shoplifting. WriteCheck matched many of Shoplifting’s phrases to those of the New York Times articles in its library of student papers. Remember, I submitted them as a student paper to help Turnitin find them; now WriteCheck has them too! WriteCheck warned me that “a significant amount of this paper is unoriginal” and advised me to revise it. After a few hours of right-clicking and scrambling, I resubmitted it and WriteCheck said it was okay, being cleansed of easily recognizable plagiarism.

Turnitin is playing both sides of the fence, helping instructors identify plagiarists while helping plagiarists avoid detection. It is akin to selling security systems to stores while allowing shoplifters to test whether putting tagged goods into bags lined with aluminum thwart the detectors.

So whose side is Turnitin on? Perhaps it is a case of “If you can’t beat them, join them.”

September 11, 2011
by stephen
Comments Off on Shopping centre now recording all licence plates

Shopping centre now recording all licence plates

Westfield Bondi Junction in Sydney has just released a new iPhone App that lets shoppers find their parked car by entering the licence plate number. The App responds with an image of the vehicle, and directions on how to find that parking spot.

While that is all very convenient, it does open a whole Pandora’s box of other opportunities. Stalking someone? Just enter their license plate to see exactly when they visit the shops, and where their car is parked. Is your employee really sick, or are they just shopping? Wanting to locate a stolen vehicle? Just wait for it to visit the shopping centre.

When visiting a shopping centre, or driving down a public road, your car and its licence plate is visible to the public. But it is only visible to the public at that location. Any surveillance that is possible would be at great cost. By automatically recording this information and making it electronically searchable, the loss of privacy takes on a whole new dimension. It suddenly becomes possible to monitor everybody’s movements, at little or no cost. Does the public have a right to privacy in regards to where they went, when they went there, and what they did in public? What does it mean to have your every movement monitored not only by those physically adjacent to you, but by people located remotely, completely anonymously, and without your knowledge? Not just right now, but also your entire history permanently recorded for all to see and search at some future time. It is technically possible now.

Screen shots and more about the App can be found at Car Advice.

September 9, 2011
by stephen
Comments Off on iTunes U tops 25 million downloads per month

iTunes U tops 25 million downloads per month

Apple: iTunes U tops 600 million downloads:

In researching my recent story, iTunes U: Educating the world, Apple provided me with some updated figures for its free education portal on iTunes.

According to Apple, iTunes U has had more than 600 million downloads since it first launched in 2007. What’s even more impressive is that they’ve had more than 300 million in the last year alone — a testament to the growing popularity of the service.

So in the average month, iTunes U now has over 25 million downloads. It is interesting to consider why iTunes U has become such a success, when there are many alternatives like YouTube, Vimeo and others that could have hosted the same videos.

According to the report, more than 1000 Universities in 123 countries have active accounts. A University account in iTunes U is a much more formal arrangement than the concept of the account in YouTube. I’m sure Apple would not close an account the moment it received a complaint. And I am sure that Apple provides a lot more reports and statistics to the University than would be available from YouTube, and they would be able to speak to somebody at Apple if any problem arose. Alternative video hosting providers typically only offer a fully-automated, no human contact account that is great for individuals, but not ideal for large organisations. By targeting a specific niche (higher education course materials), Apple is now in a very dominant position.

The report also says that 30% of iTunes U traffic comes from iOS devices. No doubt that is because the process of watching lectures on iPad/iPhone/iPod devices is particularly streamlined. By subscribing in iTunes, new lectures can be automatically downloaded and placed onto the device overnight, ready for viewing on the bus or train the next day. No internet connection is required on the device while watching the videos.

Finally, Apple does not force advertisements or Apple branding upon the iTunes U videos. Nor does it charge the Universities to participate. It makes money from the service only when a student chooses to purchase an iOS device or Macintosh computer. Since the very beginning, Apple has worked hard to encourage the use of Apple products in education, hoping to nurture a new generation of people that were familiar with Apple. By encouraging university students to use Apple products, Apple is hoping that will result in continued usage when they enter the work force.

See how several Universities are using iTunes U: