Tag: data

Oyster Complexities

Something I learned this week about the complexities of managing London’s smart-card ticketing system,

Starting with 700 stations, Oyster does not know the destination when you tap in. So the system has to hold 700 times 700 possible fares.

Then there are alternative routes, where you have to allow for people doing ‘weird and wonderful things’. As a result Oyster allows for up to 32 different routes between any origin-destination pair.

Then there are peak and off-peak fares. Here Oyster has built in ‘stretch’ which could accommodate up to 16 different time bands over the day.

Add in adult, child plus youth, senior, unemployed and other concessionary fares and the total number of combinations in the system comes to 216 million fares. That is, of course, irrelevant to the customer, who knows that, however he or she travels between Zone 1 and Zone 2, the peak fare will be £2.90.

Source: Modern Railways magazine, March 2016, p73

Data Data Everywhere

Google knows all, but how much does it really know?

Since Google announced the Gmail service, there must have been a small rainforest of paper written about the privacy implications of scanning email. It’s never really bothered me (which you should not read as I don’t care about it – it just doesn’t bother me right now), and my lovely Gmail account remains somewhat underused at the moment (oh, and I do have a spare, if you’re interested and give me a good enough reason to send you the invite).

Anyway, Hublog has an interesting list of things Google knows about you(plus an interesting follow-up). Froogle is not mentioned there, but thanks to this link, they could believe I’ve just searched for dog beds twice in 10 minutes. If they infer that I have a dog (or am about to get a dog), they would be misreading this data.

Lots of organisations have access to lots of data about you. That should never be a problem. Personally, I am on CCTV for about 10 minutes after I have left my house until I arrive at the office. Anyone who cared to could tell you quite a lot about me by observing my travel patterns daily. My bank knows where I spend my money, and my doctor knows about my illnesses (although I have been to other doctors that my main doctor is not aware of).

The point is that mountains of information about me are out there. The problem is that none of it constitutes a full picture of me as an individual, and that is the real danger. Recording the data is not a problem until somebody tries to look at it. In itself, that’s not a great problem until somebody acts on an event because of that data. However, inferring something about me because of it is not a good idea.