Cell phone security problem

It's been over a month since I've lost my HTC Sensation smartphone.  I wasn't  too overwhelmed with grief, however, because it was actually just a piece of high-tech plastic. Still, I had to spend some time changing my passwords for all the applications such as e-bank, data storage facilities etc. etc. And yes, I also miss my Ibiza bikini pics that I have not backed-up and that somebody else can see now. That's the most annoying part.

All of these could not have happened if I were cautious and had used a security key to protect my phone's content. I have never ever done this, because I am too lazy to memorize the key, and the need to enter it every time I want to access your phone really irritates me.

However, it is better than changing passwords and sharing the photos of my not-that-killer abs with some random dudes.

When I got my new ringing gadget, that is iPhone 4S (no link needed :-)), the first thing I did was to introduce that security key. The iPhone device has the only option, that is the 4-digits key, while other modern smartphones, such as Samsung Galaxy Note, offer the possibility to choose between the key and a geometric pattern.

I've noticed that the majority of my friends choses patterns, and I wondered, why? Are the geometric keys visually more appealing than numbers? When it comes to safety, pin-codes are definitely safer. The number of combinations one can compose of 4 digits is equal to:

A pattern can be composed of up to nine positions, but each point must be adjacent to the previous one and non-repeating. I've asked some of my friends, at random, how many digits do they employ in the security pattern. The result is: from 2 to 4. People do not seem to tend to bother to invent 9-positions keys.

These pattern keys turn out to not only be a measure of safety of one's information but also to breed a nice combinatorics problem. To my mind, the latter is of more value :-).

Recently, as spending a day at a beautiful Costa Brava's beach and having that kind of view in front of myself:

I had nothing to think of. So, this problem came to my mind, and I decided to estimate the odds of breaking a pattern-key. For a precise comparison, as a pin-code has 4 digits in it, I have calculated the number of all possible combinations of 4 adjacent points that could form a sequrity pattern. For 4 points, it was pretty complicated to do this analytically, so I've opted for a straightforward calculation. Which led me to this kind of tree:

(image from my Moleskine, perdon my handwriting).

From here, the number of combinations can be deduced for 2,3 and 4 adjacent points.

The result is: 488 combinations for adjacent only points with no repetitions. For comparison, if not only adjacent points could be used, there would be 9*8*7*6 = 3024 combinations.

Resume: 4-digits number keys are safer than 4-points sequrity patterns.

To compare the complete safety of the number key VS the pattern key, all the combinations of 2,3,...,9 points must be sumed up and the resulting number should be opposed to 10000. To do so, I'd probably straightforwardly compute the number of combinations from 5 to 9 by simulations. A program is needed for that, of course. The sum of all the combinations might be close to 10000 or even superior.

However, common sense tells me that people won't probably bother as inventing keys of more than five positions. All in all, I don't regret that iPhone does not provide the opportunity to use pattern keys.
   

Hong Kong's air pollution hits a two-year record and, I so want to know more

How fun is that? Lately, I've been tremendously interested in Asian air pollution research, and I have been convinced that Hong Kong is a place that hosts a top notch findings and methods in this field. As I am going to China this September, I am very much willing to get to know people that carry out air quality research there and see how they apply their methodology.

Two days ago, I have contacted CAN to inquire whether I could join them as an unpaid intern for some two or three weeks. No reply has followed yet, however. And today I have found out that Hong Kong is being suffering from the worst air pollution in two years. CAN's website provides an interactive air pollution map, which now looks as follows:

As you can see, there are many unhappy faces on this map.

On August, 1, CAN has published a press release stating that air pollution in Hong Kong exceeds the guideline values provided by WHO and avoiding that people go outdoors.

I've used the twitter search to find out the exact numbers regarding yesterday's and today's concentrations, but so far there are no such reports yet. Also, I have not succeeded to find daily listings of contaminants concentrations at the monitoring stations.

I wonder, which spatial prediction models are of common use in Hong Kong to evaluate the region's air pollution? I can count some 25 monitoring stations on this map, an the CAN's July press release reports that the study has been has been based on the information from 14 stations. The number of the monitoring stations is within the average, so geostatistical methods should be probably successfully applied. This amount of stations for the area of 1.104 sq. km is comparable to the placed in Barcelona's metropolitan region: some 49 stations for the area of 3.218 sq. km.

Land-use regression models should be probably used, too. A recent study by people from the Hong Kong Polytechnic University reports on the analysis of PM2.5 concentrations in Hong Kong. The analysis is based upon the data obtained by a spectroradiometer (satellite data) and further integrated into a GIS database with an account to meteorology. A linear regression model has been used to fit the concentrations.

In a nutshell, as a statistician, I understand that yesterday's and today's values are outliers, and I would like to see how they correspond to the fitted distributions for pollutants, as well as how they concord with the predicted pollution surfaces.

I wish someone could share this knowledge with me.

And, the last but not the least, I am tremendously sorry for the residents of Hong Kong that are suffering from high contamination right now. I very much respect they bravery to create an alert and draw public attention to the problem. For instance, in Moscow, which is my home pueblo, no such alerts happen. Although, Moscow is a very, very polluted city, however, there is no much fuzz about the pollution problem there, and it should be voiced and handled ASAP.