Talk-Talk is a fraud

It seems that I just can't pick a good mobile company. Or maybe, they are just all the same and use obfuscation to confuse and rip people off.

After probably 15 years I decided to give up on O2. Their contracts are extremely expensive compared to anyone else an their shop outlets have turned in to useless money wasting phone selling depots. You can't go to a shop to get help with your device or your account. I couldn't even buy a screen protector for my phone that I bought from them at the shop.

So, since I get my broadband from Talk-Talk, and since they had some good deals for existing customers I switched my phone accounts to them - three of them. AND ALL THEY SOLD ME WERE A LOAD OF LIES.

Now if you have a teenager, you might know that having a cap on your account is of utmost importance, so before I switched mobile provider I went online and used their chat facility (and I think I might have called them too) to make sure that when you run out of data or call time you would not be able to make any further calls or use data until you have gone online and purchased some more if you wanted to do so.

Today, however, I found out that this is not the case. After, my daughter overspent £18 on the account they decided to freeze my whole account. That is, not just the one phone ... no, they freeze ALL four accounts (the fourth one is the free one you get with your broadband/phone contract).

So as one does I resolved myself to having to waste the next couple of hours on the phone with a Talk-Talk representative.

The first thing they tell me is that Talk-Talk apparently have a usage limit and a capping limit. But actually they don't have a capping limit because, unlike I was led to understand in the first place, they don't place a cap on the traffic to your phone. So you can happily overspend without knowing it. The usage limit is the limit after which they will suspend your contract. So if you overspend because they are incapable of capping they will suspend your account. However, since they only get usage report every 24 hours, you can happily overspend in 24 hours and then they'll suspend your account when they receive the info. This reminds me of this story http://news.bbc.co.uk/1/hi/world/africa/8248056.stm.  So Talk-Talk admits to being slower than a carrier pigeon. We can put a rover on Mars, I can buy things in China and get it delivered within 3 days, I can transfer money to someone anywhere in the world in seconds, but Talk-Talk can't get an up to date usage report in less than 24 hours.

So basically there were three issues I wanted resolved. Firstly I wanted my usage limit to be the same as my capping limit. But since Talk-Talk can't get an up to date usage report they are incapable of implementing that and I had to sit and listen to a 15 minute advisory session of how to switch your data usage off when you are not using your data. Trying to get the message across that I was not in need of the tutorial on how and when and why to switch data off just led to contention. So I sat here chewing on my desk in an attempt to not burst into tears out of frustration.

The second issue was that I wanted the already accumulated charges of £18 waived based on the fact that I was miss sold a product. It wasn't so easy but after some time the Talk-Talk rep agreed to this and he agreed to unsuspend my accounts.

The third issue was that I wanted to be given the opportunity to cancel my contracts with Talk-Talk without any early cancellation charges because I was miss sold a product. To this the rep wouldn't agree. I pointed out that, since he has already agreed to refund £18 he has admitted to the product being miss sold and therefore I should be given the opportunity to get out of the contract without being penalised. He also insisted in telling me what the cancellation charges for all four phones were going to be, despite the fact that I kept telling him that I am not interested in hearing the cancellation charges because I do not wish to pay them. What I said to him was that if he cannot resolve this issue he should give me the name and/or address of the person to contact about this and I'll sort it out with them.

At this point the rep decided that he was no longer going to waive the £18 and that he wasn't going to unsuspend my accounts either.  I pointed out to him that he has, if he maintains this position, just made himself guilty of lying and dishonesty. The fact that he is now going back on something he has agreed upon, in my books, makes him and the company he represents dishonest and liars. Obviously he wouldn't agree to this statement but interestingly enough, now that I think about it, he never disagreed with the statements either and thus I can take this as acknowledgement of the fact that both him and the company he works for are guilty of being dishonest to clients. He also didn't disagree with my statement that he was now merely going back on his word because he was on a power trip.

I now have to wait until Tuesday when he has talked to his Chief Technical Officer to decide what is going to be done. So his power trip continues.

So let me take the opportunity to warn you against Talk-Talk's questionable practises. This is not the first time I have been lied to blatantly by Talk-Talk. Some years ago I demanded and early termination of a contract because I was distinctly told that I would get a static ip for my broadband provision, just to be told after installation that Talk-Talk did not do static ip addresses.

When something like this happens once, you can probably put it down to ignorance on the rep's part but when it happens a second time you start questioning the company's integrity.

TALK-TALK, I'M QUESTIONING YOUR INTEGRITY.

Conway's Game of Life written in Processing

When I feel a bit "down in the dumps" and I don't fancy getting up to do anything, I program. I think it is because it allows me to cut off from the world and focus on one tiny thing. So I have been meaning to look into Conway's Game-of-Life, but never got around to it. The past few days I have been struggling to focus on writing my thesis and ended up writing one in Processing. It's nothing spectacular, but quite fascinating to watch the different patterns grow (or not). Here is my code:

001 int size = 100; 002 int cellsize=4; 003 int displaysize = size * cellsize ; 004 int[][] world = new int[size][size]; 005 int[][] newworld = new int[size][size]; 006 int r1 = (int)random(size); 007 int c1 = (int)random(size); 008 boolean pause = true; 009 color alive = color(0, 0, 0); 010 color dead = color(255, 255, 255); 011  012  013 void setup() { 014   size(displaysize, displaysize); 015   // initialise world with zeros 016   for (int x = 0; x < size; x++) { 017     for (int y = 0; y < size; y++) { 018       world[x][y] = 0; 019     } 020   } 021  022   // start life 023   // gun 024   world[39][39]=1; 025   world[40][39]=1; 026   world[39][40]=1; 027   world[40][40]=1; 028  029   world[49][39]=1; 030   world[49][40]=1; 031   world[49][41]=1; 032   world[50][38]=1; 033   world[50][42]=1; 034   world[51][37]=1; 035   world[51][43]=1; 036   world[52][37]=1; 037   world[52][43]=1; 038   world[53][40]=1; 039   world[54][38]=1; 040   world[54][42]=1; 041   world[55][39]=1; 042   world[55][40]=1; 043   world[55][41]=1; 044   world[56][40]=1; 045  046   world[59][39]=1; 047   world[59][38]=1; 048   world[59][37]=1; 049   world[60][39]=1; 050   world[60][38]=1; 051   world[60][37]=1; 052  053   world[61][36]=1; 054   world[61][40]=1; 055  056   world[63][35]=1; 057   world[63][40]=1; 058   world[63][36]=1; 059   world[63][41]=1; 060  061   world[73][38]=1; 062   world[73][37]=1; 063   world[74][38]=1; 064   world[74][37]=1; 065  066  067   // oscillator Blinker 068   world[10][10]=1; 069   world[11][10]=1; 070   world[12][10]=1; 071   world[9][11]=1; 072   world[10][11]=1; 073   world[11][11]=1; 074  075   //DieHard   076   world[20][10]=1; 077   world[21][10]=1; 078   world[21][11]=1; 079   world[25][9]=1; 080   world[24][11]=1; 081   world[25][11]=1; 082   world[26][11]=1; 083  084   stroke(dead); 085 } 086  087 void draw() { 088   for (int c = 0; c < size; c++) { 089     for (int r = 0; r < size; r++) { 090       // if there isn't life, make the point white; 091       if (world[c][r]==0) fill(dead); 092       // if there is life make the point black 093       if (world[c][r]==1) fill(alive); 094       //if (world[r][c]==1) System.out.println("life at " + r + "," + c); 095       rect(c*cellsize, r*cellsize, cellsize, cellsize); 096     } 097   } 098   if (!pause) makeworld(); 099 } 100  101 void makeworld() { 102  103   // initialise newworld 104   for (int x = 0; x < size; x++) { 105     for (int y = 0; y < size; y++) { 106       newworld[x][y] = world[x][y]; 107     } 108   } 109  110   // Determine life in the new world 111   for (int c = 0; c < size; c++) { 112     for (int r = 0; r < size; r++) { 113       // initialise neighbours 114       int neighbours = 0; 115       //check surrounding life 116       for ( int ic = (c-1); ic <= (c+1); ic++) { 117         for ( int ir = (r-1); ir <= (r+1); ir++) { 118           if (((ic>=0)&&(ic<size))&&((ir>=0)&&(ir<size))) { 119             // don't include self 120             if (!(ic == c && ir == r)) { 121               // if there is life increment neighbours 122               if (world[ic][ir]==1) neighbours++; 123             } 124           } 125         } 126       } 127  128       // apply game of life rules 129       // if there is life 130       if (world[c][r] == 1) { 131         if (neighbours < 2 || neighbours > 3) { 132           newworld[c][r] = 0; // Cell dies 133         } 134         // if there isn't life 135       } else {// i.e. world[c][r] == 0 136         if (neighbours == 3) { 137           newworld[c][r] = 1;  // Cell becomes alive 138         } 139       } 140     } 141   } 142  143   // initialise newworld 144   for (int x = 0; x < size; x++) { 145     for (int y = 0; y < size; y++) { 146       world[x][y] = newworld[x][y]; 147     } 148   } 149 } 150  151  152 void mousePressed() { 153   if (pause) { 154     pause = false; 155   } else { 156     pause = true; 157     System.out.println("pause"); 158   } 159    160 } Java2html

ODEs with Java using the Apache Commons Math library

It drives me nuts when I struggle to get my head around something just to, when I eventually do, discover that it is not really as complicated as the documentation makes it out to be. Simple examples are what I need. Here is an example of how to implement the Lorenz System. Not that I really know much about the Lorenz System but it is always a simple system to use when you want to try out a new way of solving differential equations. More information about the Lorenz System can be found in Wikipedia.

The first thing to do is to create a class containing the equations. The Lorenz system consists of the following three equations:

Doing this using the Apache Commons Math library means that we have to create a class that implements the FirstOrderDifferentialEquations interface. This interface requires two methods to be implemented, which are computeDerivatives and getDimension:

01 import org.apache.commons.math3.exception.DimensionMismatchException; 02 import org.apache.commons.math3.exception.MaxCountExceededException; 03 import org.apache.commons.math3.ode.FirstOrderDifferentialEquations; 04  05 public class Lorenz implements FirstOrderDifferentialEquations { 06  07         double sigma = 10.0; 08         double beta = 8 / 3; 09         double rho = 28.0; 10         double X, Y, Z; 11  12         @Override 13         public void computeDerivatives(double t, double[] y, double[] yDot) 14                         throws MaxCountExceededException,                             DimensionMismatchException { 15                 X = y[0]; 16                 Y = y[1]; 17                 Z = y[2]; 18                 yDot[0] = sigma * (Y - X); 19                 yDot[1] = X * (rho - Z) - Y; 20                 yDot[2] = (X * Y) - (beta * Z); 21         } 22  23         @Override 24         public int getDimension() { 25                 // TODO Auto-generated method stub 26                 return 3; 27         } 28 }

Java2html

The next step is to write the code that will call the integrator. There are various integrators. For this application we'll use the ClassicalRungeKuttaIntegrator. You will also need a step handler to capture each step of the integration:

01 import org.apache.commons.math3.ode.FirstOrderDifferentialEquations; 02 import org.apache.commons.math3.ode.nonstiff.ClassicalRungeKuttaIntegrator; 03 import org.apache.commons.math3.ode.sampling.StepHandler; 04  05 public class LorenzMain { 06  07         public static void main(String[] args) { 08  09                 ClassicalRungeKuttaIntegrator rk4 =                             new ClassicalRungeKuttaIntegrator(0.01); 10                 FirstOrderDifferentialEquations ode = new Lorenz(); 11                 double[] y = new double[] { 10.0, -2.0, 50 }; // initial state 12                 StepHandler stepHandler = new                             WriteToFileStepHandler("lorenz.csv"); 13                 rk4.addStepHandler(stepHandler); 14                 rk4.integrate(ode, // equations 15                                 0.0, // start time 16                                 y, // initial conditions 17                                 100.0, // end time 18                                 y); // result 19                 for (int i = 0; i < y.length; i++) { 20                         System.out.println(y[i]); 21                 } 22         } 23 }

Java2html

The step handler has to implement the StepHandler class. For the step handler below I created a constructor that takes a filename so that the results can be written to a csv file. Obviously you can do whatever you want with this data:

01 import java.io.File; 02 import java.io.PrintWriter; 03 import java.util.ArrayList; 04  05 import org.apache.commons.math3.exception.MaxCountExceededException; 06 import org.apache.commons.math3.ode.sampling.StepHandler; 07 import org.apache.commons.math3.ode.sampling.StepInterpolator; 08  09  10 public class WriteToFileStepHandler implements StepHandler { 11         ArrayList<String> steps = new ArrayList<String>(); 12         String filename; 13          14         public WriteToFileStepHandler(String filename) { 15                 this.filename = filename; 16         } 17         @Override 18         public void handleStep(StepInterpolator interpolator, boolean isLast) 19                         throws MaxCountExceededException { 20                 double t = interpolator.getCurrentTime(); 21                 double[] y = interpolator.getInterpolatedState(); 22                         String onestep = "" + t; 23                         for (int i = 0; i < y.length; i++) { 24                                 onestep += ", " + y[i]; 25                         } 26                         steps.add(onestep); 27                 if (isLast) { 28                         try { 29                                 PrintWriter writer =  30                                 new PrintWriter(new File(filename),                                     "UTF-8");                                     31                                 for (String step : steps) { 32                                         writer.println(step); 33                                 } 34                                 writer.close(); 35                         } catch (Exception e) { 36                         } 37                         ; 38                 } 39         } 40  41         @Override 42         public void init(double t0, double[] y0, double t) { 43                 // TODO Auto-generated method stub 44  45         } 46  47 }

Java2html

Last but not the least I read the file with R and created the graph: