Monday, December 16, 2013

Is the XBox One or Playstation 4 worth buying?

I've been wondering if it is worthwhile spending thousands of Rands (or hundreds of Dollars ...) on a next generation console, and I came to the conclusion that it may not currently be worth the money. The reasons for this, in my opinion, are:
  • The price of games: Is a computer game really worth R800?s think about this ... R800 is more or less equal to a month of DSTV, if you still waste your money on that (Yes Naspers, I am looking at you ...). If you are lucky it will cover a large part of an electricity bill. And then there is food: depending on what you buy this could be a week of groceries or petrol. Worst of all: buy 5 and you have spent R 4000!
  • Single player games tend to be linear: Is it really worth playing something that is pretty much on rails? If everyone will have the same experience, it makes more sense to just watch a game playthrough on YouTube. Multiplayer is about the only real usage scenario for me here because of the replay-ability.
  • It's a single purpose device: In the USA, you could use your XBox One to watch TV, as well as play games. But in SA it is just a gaming device. This happens even though it is a full PC and can probably run a full version of Windows 8.1. I don't want to spend between six and eight thousand Rand on something I will only ever use for computer games. It is also a DVD player, but if you want a DVD player there are cheaper options. This is also true for browsing the web.
  • Locked in ecosystems: I guess it could be argued that this is not a problem: most major games get released on most platforms. The thing that bothers me, however, is that the console manufacturer controls the distribution channel and can cut you off at any time. If you publish via this channel they will take a chunk of your money.
  • The price of the device, and getting ripped off because we do not live in North America: The PS4 has a recommended retail price in South Africa of R 6300. This is US$ 609 at current exchange rates!!! Overseas it costs US$ 400, or about R 4150. Someone is making a cool two thousand Rand per sale.
I may eventually get an XBox One due to its ties with the Windows operating system. Be that as it may, the price will have to come down to about R 3000 before I will even consider buying it. Game prices will also have to come down to a maximum of R400 per game.

Referenced Links

http://www.ign.com/wikis/xbox-one/Xbox_One_Hardware_Specs
http://www.ign.com/wikis/playstation-4/PlayStation_4_Hardware_Specs
http://mybroadband.co.za/news/gadgets/93691-playstation-4-launched-in-south-africa-price-details.html
http://www.pcmag.com/article2/0,2817,2419594,00.asp

Sunday, November 3, 2013

Google, Too Much Advertising and That Thin Red Line

The other day I was watching a series of videos on YouTube, when an extremely annoying beer add popped up. The average volume (which I set on my tablet) was automatically raised and I was forced to watch and listen to 5 seconds of it before I could "skip" it. After skipping it, I had to watch it again and again as I browsed to other videos, even though I already made it clear that I am not interested.

This raised the following question with me: When do we step over the line of too much advertising in a free service?

This is not a simple question, mainly due to the fact that Google is not just pushing ads, but also providing useful services. Their products include e-mail, entertainment (via YouTube), a software platform (Android), a means for programmers to distribute their wares (via the application store, or "Play Store"), a search engine and business applications. Not to mention this blogging platform that I am using.

This is an impressive range of services for a company that "only" requires you to watch advertising. The exact amount of money made from advertising can be found here: http://investor.google.com/financial/tables.html. According to this (and assuming I am reading the table correctly) they made about 12.5 Billion US Dollars in advertising in the third quarter alone.

So it would appear that there is a balance here between annoying the client and giving him services. The problem I am having with this is that they are approaching a thin red line where they will soon start loosing customers. A bit of Googling (with the search engine I pay for through advertising ...) showed me that I am not the only annoyed person:
So now I have to wonder, where is my thin red line, and when will they cross it? It has already been crossed for this blogger: http://kindofahater.com/post/16989050637/my-revolution-against-google-begins.

I was myself tempted to leave, however their products are still good enough to keep me enticed. I would considder paying them a monthly/yearly fee to make advertising go away but to keep their product. In the mean time I get the sense that they will continue to play the "numbers game" where they phase in as much advertising as possible while trying to lose as little as possible clients.

In the meantime I will think of this poem I first encountered in The Prince of Thorns when I see advertising on the net:
Yesterday, upon the stair,
I met a man who wasn't there.
He wasn't there again today,
I wish, I wish he'd go away...
It was written by Hughes Mearns and can be found here, free of charge: http://en.wikipedia.org/wiki/Antigonish_%28poem%29

Sunday, October 20, 2013

Crazy Idea: Lets build a "reality decoder"

Every once in a while, I have a crazy idea. I believe that most people who build things for a living or who work as researchers should at least try to come up with something crazy once in a while. So here is my Crazy Idea: Let's try to find a way to predict the future and take pictures of places we have not been to. In effect, I want to represent reality as a mathematical function.

1. What is reality in this context?

Many people have speculated about the nature of reality, but this is a very informal posting so I am not going to delve into the philosophical aspects of this problem. For my purposes, I view reality as a pattern. The reason for this is that there are entire fields in mathematics, statistics and computer science that is dedicated to finding, analysing and predicting patterns.

In short, the "reality pattern" is a mathematical function that takes the following parameters as input, and renders an image of what you would see there if you were to take a picture with a digital camera at that time/place:
  • Position in 3-dimensional coordinates
  • A "look vector", i.e. in what direction are we looking
  • A "up vector", i.e. which direction is up
  • A time index
This forms our basic problem statement: How do we find this “reality function”?

2. If reality is a mathematical function, how do we find it?

This is where my knowledge is somewhat lacking. From my training in computer science I can think of at least one method that could form the basis of a project that attempts to derive this function. I am not sure what the statistics/numerical methods approach to this would be, but I think that genetic programming would make a good starting point from a computer science perspective. 

Simply put, genetic programming allows you to take a set of data as input, and then "evolve" a program using a process similar to that of evolution. This program will satisfy a programmer-defined fitness function that would be used to eliminate "weak" programs and promote "strong" programs. A more detailed synopsis can be found on Wikipedia: http://en.wikipedia.org/wiki/Genetic_programming.

In the context of this problem, the objective would be to evolve a program that renders the image from section 1. The set of training data for the genetic program would be a collection of photos taken with a digital camera. For each photo, the location, look vector, up vector and time index will be recorded. A program will then be evolved that attempts to produce an output that is equal to the photo.

The closer the generated images is to the input photos in the training set, the higher the program that produced them will be rated. The evolved functions can also be tested against a separate data set.

3. Further steps and consequences

Building the system in section 2 may be possible; I have never tried it and have no idea of the challenges involved. If such a system were to be built successfully, it will have interesting social and economic consequences. On a personal level it would mean the end of privacy. On a larger social scale it will break almost every human system that emerged in the last 12 000 years. But then again, maybe people will just ignore it and go on with their lives.