Reviewing “The Thirteenth Floor” as a Computer Scientist (Can We Do IT?)

Standard

The Thirteenth Floor (1999)

Craig Bierko, Gretchen Mol

Rating : 9 / 10

The Thirteenth Floor

(WARNING) If you didn’t watch this movie I recommend that you watch it first before reading this blog because of the HUGE spoilers that will follow :D.

The movie tells Computer scientist Hannon Fuller has discovered something extremely important. He’s about to tell the discovery to his colleague, Douglas Hall, but knowing someone is after him, the old man leaves a letter in his computer generated parallel world that’s just like the 30’s with seemingly real people with real emotions. Fuller is murdered in our real world the same night, and his colleague is suspected. Douglas discovers a bloody shirt in his bathroom and he cannot recall what he was doing the night Fuller was murdered. He logs into the system in order to find the letter, but has to confront the unexpected. The truth is harsher than he could ever imagine, he finds out that his world is as well a virtual one and he himself is a computer program as all his world which was created by a REAL human world software company (IMDB).
I watched this movie a couple of weeks ago and I just can’t stop thinking about it. Not only because it is a wonderful and new concept (in its time of course) but also because it is very related to my research work as an Artificial Intelligence researcher in the university. My brother recommended me this movie because he watched it and the idea fascinated him as itself. Of course, the concept of a world within a world within a world …. etc with none of them aware of the above (larger) world is visited in the ASCAP award winner movie “Horton Hears a Who” (2008) but the thing here is that each world is totally created by individuals in the upper world. After I watched the movie my brother (not related to computer science) gave a trivial hint about the possibility of creating this in real world. I didn’t answer him at the moment but after a while I thought of it seriously. Is It possible to create “The Thirteenth Floor” in reality? 
To Clarify this to the normal readers, I will gradually explain 3 AI concepts related to the movie (probably the movie creators themselves didn’t have these concepts in mind when they created it and that gives them more credit of imagination abilities): Strong AI, Artificial Life and Automatic Programming. 
But before we discuss these concepts we have to clarify what is meant by AI (Artificial Intelligence) to readers who have never heard of it or misunderstand it. 
First of all, before I start explaining (or simplifying) AI to you, you should know that AI isn’t something unusual or impossible to achieve. AI applications are around you everywhere and you even use a bunch of them without realizing that this is an AI application. I said this because many people (non-computer science people) whom I have discussed AI concept with them were very puzzled and skeptic to the idea. Artificial Intelligence is simply the science to create machines that think, decide and act as a human BRAIN does. Why I capitalized BRAIN? You will know very soon. 
The idea is not that new to humans. Creating humanoids (something that has the appearance of humans) and thinking machines is as old as civilization itself. In Greek mythology itself we find some thinking machines like Talos (The bronze man who protected Europa from pirates) created by the god of fire, invention and technology Hephaestus. Not only that but we can find in the Lie Zi text that a mechanical engineer “Yan Shi” has built a life-size human-shaped figure of his mechanical handiwork to the King Mu of Zhou (1023 – 957 BC). There are more examples of AI in old mythologies but we are not going more into that as the previous examples must have cleared my point. Now to the modern concept of AI. AI Science (in its current form) was founded at a conference in Dartmouth College in the summer of 1956. So, now we understand what AI REALLY means in general. Before some of you (new ones to the AI concept) protest saying we CANNOT simulate a human brain as it is a miracle from God and it is only accessible to humans, I will state some examples of popular AI applications in our everyday life and get back to that ‘miracle’ issue later. 
The most popular and market dominating AI applications are Computer Games. You must have played some game in some years of your life (even if you’re not  a gamer like me) but all of us did. Doesn’t matter if it is an old, 2D game like Mario, Contra, Captain Tsubasa or Jackal or a huge 3D open/closed-world games like Assassin’s Creed franchise, Halo, Call of Duty, Pro Evolution Soccer. In all these games you can find computer agents (that you don’t control) which play against or with you and making decisions on their own without any input or parameters from you directly to them. They just record what you do with your player or character and make the perfect decision that will achieve their goal. 
Another application to AI are Expert Systems. Expert System is programmed using AI methods to simulate a human expert in a certain field. Rice-Crop Doctor is an expert system developed by National Institute of Agricultural Extension Management (MANAGE) to diagnose pests and diseases for rice crop and suggest preventive/curative measures. The rice crop doctor illustrates the use of expert-systems broadly in the area of agriculture and more specifically in the area of rice production through development of a prototype, taking into consideration a few major pests and diseases and some deficiency problems limiting rice yield. Professor Heather Christine published a paper about an expert system to educate “Introductory Data Structure Course” . One more example of expert system is in environment domain is the hybrid expert system, GIS and simulation modelling for environmental and technological risk management particularly called RTXPS and developed by K. Fedra and L. Winkelbauer from Environmental Software & Services GmbH. The system was actually an integration of a real-time forward chaining expert system and a backward chaining system as the Decision Support System framework using simulation models and Geographic Information Systems or GIS. RTXPS was based on the results from the international research project called HITERM which was funded under the European ESPRIT technology programme for high-performance computing and networking (HPCN) for decision support.
IBM’s Watson is a miracle of its own. In February 2011, in a Jeopardy! quiz show exhibition match, IBM’s question answering system,Watson, defeated the two greatest Jeopardy champions, Brad Rutter and Ken Jennings, by a significant margin. 
Watson in Jeopardy
Similar to Watson is iPhone’s most popular application Siri, which is a program that replies to your talking whether its a question or a normal dialogue. Of course, it is not the same as Watson of course but it deserves its credit of AI discussion.
Our last example will be in the field with the most number of expert systems developed which is Medicine. PEIRS (Pathology Expert Interpretative Reporting System) appends interpretative comments to chemical pathology reports (Edwards et al., 1993). 
So, now you know WHAT AI is and that AI does exist for you skeptics out there 🙂
Now, let’s get back to the subject, The Thirteenth Floor. 
Our first concern is “Is it possible to create computer programs that really really think, decide, interact and feel exactly like a human?”. We saw in this movie computer programs (human likes created by Hall and Fuller who turn out in the end to be computer programs themselves) that act and feel exactly like humans. To answer this question we need to explore the first concept “Strong AI”. 
Artificial Intelligence field is divided into two major areas of research: Strong and Weak AI. Weak AI researchers (The majority including me) tend to build applications that act like human in a specific area and no more. Like all the examples I mentioned earlier. Games intelligent agents are intelligent in that they act and think like humans to beat the other player in this game. An agent program in PES will not be able to play against you in COD and vice versa. Each intelligent program is programmed only to function in its environment and to do a specific task. I don’t want the word ‘Programming’ confuse you as it is not conventional programming that tells the program exactly what to do and how to do it, AI programming just tells the program what is its task or goal. We – AI researchers and programmers – don’t tell the program the steps to solve the problem. We just train it to solve it itself without any help from us. Another example on Weak AI is in Expert Systems. In the examples we mentioned, The Rice-Crop Doctor cannot answer a medical or educational question. And as well, PEIRS cannot answer an agricultural question. 
Strong AI researchers tend to reach an AI system that matches or exceeds human intelligence. They believe they would reach a phase in which not only a computer program will think and act like a human but also to understand what it is doing. To clarify this, Let’s take the Watson example. We have just known that Watson is IBM’s brilliant invention to understand human natural language and interpret its meaning to get the right answer for the question. As an example one of the clues given to Watson in Jeopardy was “William Wilkinson’s ‘an account of the principalities of Wallachia and Moldavia’ inspired this authors’s most famous novel” for which Watson got the right answer “Who is Bram Stoker?”. Ok, he got it right, that’s true. But did he understood it? I mean what happened IN Watson is these exact steps:
  1. Watson got the input (which is the clue) from Alex Trebek.
  2. It analysed it using AI and NLP algorithms developed by IBM to interpret its category and meaning.
  3. It used DeepQA algorithms to find the correct answer.
  4. it found the answer and produced a sound and complete question “Who is Bram Stoker?” using AI algorithms as well.
These are the steps more or less, but does all this means that Watson understood what he is solving? Simply, NO!!
Watson doesn’t understand or comprehend who is William Wilkinson nor the fact that he is a person. Watson doesn’t understand what Wallachia and Moldavia are. Watson doesn’t understand what an author is. Watson doesn’t understand who is Bram Stoker. Watson doesn’t even understand that he is in a competition. He KNOWS all that but doesn’t UNDERSTAND it. And yes, there’s a big difference between knowing something and understanding it. 
To explain the difference we will use the american philosopher John Searle’s thought experiment “The Chinese Room“. Searle writes in his first description of the argument: “Suppose that I’m locked in a room and … that I know no Chinese, either written or spoken”. He further supposes that he has a set of rules in English that “enable me to correlate one set of formal symbols with another set of formal symbols,” that is, the Chinese characters. These rules allow him to respond, in written Chinese, to questions, also written in Chinese, in such a way that the posers of the questions – who do understand Chinese – are convinced that Searle can actually understand the Chinese conversation too, even though he cannot. Similarly, he argues that if there is a computer program that allows a computer to carry on an intelligent conversation in written Chinese, the computer executing the program would not understand the conversation either.
And here is the difference, in Weak AI researchers try to make the computer programs know what to do without our interference but in Strong AI researchers seek to make the programs know, understand and maybe feel what they are doing.
We explored some of the Weak AI real-world applications, but where are Strong AI applications? The answer is “There aren’t any”. Strong AI can only be found in science fiction novels and movies like I-Robot, Wall-E, A.I. Artificial Intelligence, The Matrix, Tron, The Terminator, Short Circuit, RoboCop, Transformers, Bicentennial Man, … etc. 
So, if on some day Strong AI guys found a way to overcome all the problems that face Strong AI then we – for sure – can create computer programs like the ones in “The Thirteenth Floor”. I’m not saying it is impossible (being a weak AI researcher) but it is very very difficult because in Weak AI we are simulating the human BRAIN (See? I didn’t forget that 🙂 ) but in Strong AI we are trying to simulate the human MIND. 
What is the difference? If you want to know it in much detail, you can read “Then the brain became a mind” book by Amr Sherif but this one is only in Arabic. An alternative may be “The Mind and The Brain: Neuroplasticity and The Power of Mental Force” by J. Schwartz and S. Begley. Simply, a brain is a bunch of neurons and electrochemical processes that perform different functions, but a mind is what defines our consciousness and ourselves. It has the meaning of things we interact with in every moment. Brain is Knowing but Mind is Understanding. 
That cleared the first question “Is it possible to create computer programs that really really think, decide, interact and feel exactly like a human?”. 
Now, let’s move on. Suppose that we created a computer program that can act, think, understand and feel exactly like a human. Then, can we create an entire world of these programs with other objects like buildings, cars, phones, …. and other life forms like animals, plants, germs, ….. etc?
This is the role of our second concept, Artificial Life. A-Life is the simulation of any aspect of life as through computers, robotics or biochemistry. A-Life concept was named by Christopher Langton, an american computer scientist in 1986. Artificial life imitates traditional biology by trying to recreate some aspects of biological phenomena. The term “artificial intelligence” is often used to specifically refer to soft alife (software created artificial life). 
There are some working A-Life simulators like Avida, Breve, Creatures Game, 3DVCE, … etc. And in games also we can see some sort of A-Life like in Assassin’s Creed Game, where you can find animals, birds, plants, objects (buildings, streets, … etc) and everything you can interact with. But not only you, other agents (computer characters) in the game also interact with this life. 
ac4_2388056b gsmarena_001 Assassins-Creed-III_2012_04-13-12_001 600-1370772810_acivbf_screenshotsp_e3_pirateparty_130610_4h15pmpt 468px-Assassins_Creed_3.Screenshots.262 Assassins-Creed-3-24 Assassin_s_Creed_4_13745304255247 ac4pax_06
(Excuse me I’m a big fan of the game, couldn’t help myself 😀 )
If I managed to get your interest in A-Life and you want to learn more about its methods you can keep up-to-date with these journals MIT journal of A-Life and International Journal of A-Life Research
So, now we answered the second question “Can we create an entire world of these programs with other objects like buildings, cars, phones, …. and other life forms like animals, plants, germs, ….. etc?”. Yes we can. 
Now, to the last piece of the puzzle, in the movie humans create an artificial life (Fuller and Hall’s entire world) and this one was able to create another virtual A-Life on their own (the 30’s world). So, here is the question, Can intelligent computer programs create (program) other computer programs on their own? Can programs AUTOMATICALLY create other programs? 
That leaves us with the last concept in our bag “Automatic (Generative) Programming”. 
Despite the different definitions of Automatic Programming through the years, now it means simply to “write code that writes code“. In other words, to create computer programs that are able to generate other programs automatically without (or with the minimal) need to a human programmer. It may not be the same idea as in the movie but it is a start. Expert developers should know that there are already some automatic programming in YOUR programs (like automatic buffering, memory management, … and other low level tasks). Currently the Automatic Programming trend is to just input a very high-level specification of a program to automatically generate its source code. But this is just a start, the progress in this area is still in its first steps. More about Automatic Programming can be found here at its course in Texas University.
At last, we have defined the three concepts (Strong AI, A-Life and Automatic Programming). So, let’s try to get a conclusion out of all this. Can we create something similar or near to “The Thirteenth Floor”? The answer is: if we someday managed to create Strong AI, then with the help of A-Life and Automatic Programming we can create worlds like the ones in the movie. 
Thank you for reading this very long article, and of course I welcome all your thoughts and comments.

The Thirteenth Floor (IMDB)

2 thoughts on “Reviewing “The Thirteenth Floor” as a Computer Scientist (Can We Do IT?)

  1. Arlee Bird

    This is all so beyond me that my head spins when I start thinking about it. The video games amaze me and even some of them seem so complicated to me that I never get far in them, but then again I don’t play them much at all.

    Love films like 13th Floor and I could watch these films repeatedly. For now I’ll enjoy them and let the smart guys like you ponder what to me are almost imponderables. Now I need to go back and watch the Matrix series again.

    Arlee Bird
    A to Z Challenge Co-host
    Tossing It Out

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s