My writings can be found on sites other than this one. Content hunters are invited to read on for discussions about why people think it’s natural for us to have assumed that the Sun went round the Earth — and to view what’s probably the first Powers-of-ten type video, where we zoom from an atomic scale to the edges of galaxies.
One of my courses this semester is Programming Languages, covering their fundamental histories and differences. Something I direly need to open my eyes and jolt me out of Java fanaticism. I particularly enjoyed playing around with Scheme, a dialect of Lisp. Rather than selfishly leaving the precious resources somewhere in a dark corner, I wrote this article to help get you started with Lisp as well.
RSS feeds are everywhere now, and for a good reason! It can be a really powerful tool for those who want to stay savvy, but don’t have much time to browse all of their favorite sites (yeah, I just described the entire population of Earth there).
Here’s an overview of what RSS is, how you can use it to save your precious time and also some additional nifty tricks I frequently use (top secret stuff).
A new school semester has dawned and once again I find myself scuttering about to prepare for what’s ahead. It’s funny, nevermind the fact that it happens every year, surprise never fails to rear its horns when I suddenly realize it’s almost winter again. Jump started by the usual strong cup of coffee and inhalation of deadly nicotine, my day is now full of people with bad English accents pimping out Powerpoint slides. And I’m sitting there thinking how the learning process could be different.
There’s something very important you must always try to do when interviewed: Ask to review it before it prints! This may not be possible at all times. For example, it tends to be a lot easier to get a copy if the article is long and scheduled for publication with a few days notice. Nevertheless: ASK. It’s amazing how often quotes can get scrambled, your statements misinterpreted, ripped out of context, and so on and so forth. But it doesn’t always have to be bad to allow the reporter to have his way making the story sound more exciting on the expense of its accuracy.
The general media and science are truly an odd couple. Important details often get lost in translation, and sometimes the reporter himself is lost in the mazes of our architectures. In my run-ins with the media, I’ve come to understand a thing or two on how the general media can help you, and how to deal with them for maximum benefit to yourself, your project, and to the readers. Here’s a some basic advice for young scientists, making their way up the academic ladder.
When working on any software system there are various challenges. When a system is entirely new, there’s a ton. Here’s a short and generalized description of what’s involved in the process. In the interest of keeping this non-technical, let’s use some analogies of adventure and countryside (that we’re ruining with dams to power our systems, ironically). A system is a specific place that you want to go to, like Grandma’s Secret Hillside Bakery. The path that that leads to it are parts of the program you have to create.
Everybody can agree that the number of teachers versus students significantly effects the quality of education. With a high number of students to each teacher, the courses have to be adapted to the group and less attention payed to each students characteristics, strengths and weaknesses. I’ve been aware of this pesty fact through my own experience of school and consider it a noteworthy problem of modern education. So, let’s mass produce teachers to come to the rescue.
When I wrote the laws for the Icelandic Society for Intelligence Research (ISIR), Iceland’s first A.I. associated which I founded in 2006, there was one law that I was particularly fond of. Law nr. 13. The Fifth Member of the Board. It’s a particularly interesting law, many find it creative — others scary, but most find it a bit silly. Those who find it silly are usually basing their opinion on their own imagination and on the image that Hollywood has created for artificial intelligence. Which is not what I envisioned when I wrote it.
The 13th law is written in Icelandic, but can be translated as follows.
(i) A majority vote from the board can approve an artificially intelligent agent to serve the duties of the fifth boardmember in any way the board sees fit. The voting right of the fifth boardmember is decided by the human members of the board, but the agent will always maintain a right to provide advice or input during decision making.
(ii) The fifth boardmember has the same rights as a human boardmember to suggest changes of laws or code of conduct. It is the duty of the board to assess these suggestions as if it was presented by a human.
To the people shaking their heads, this law is real — reviewed and approved by government officials. I believe it is the first governmentally approved law in the world to actually account for an artificial intelligence in a management position. At this point, judging from previous responses I’ve got, most will probably be thinking "Damn, that’s cool", "Terminator is coming" or "What a nutjob". I suspect that the majority will be thinking the last option. So, if you bare with me, I’ll explain the thought behind this idea; you might find that it’s not as spaced out as you think.
Precursor to Advanced Automation
Only 30 years ago all companies, organizations, stores and corporations did their financial transactions and business deals using typewriters and pens. Files had to be manually typed up, calculations had to be mentally accounted for. There were no Excel documents that automatically calculated the annual growth rate, weekly estimates or taxes. There was no Oracle or MySQL database to store this vital data that kept your company alive, either. Everything was manual. Automation in this sector was only for producing products on an assembly line, or the automatic "jam release" of your typerwiter.
It wasn’t until January 1971 that the first portable calculator was introduced. It weighed 1 pound (too large to carry around casually) and cost over $300. Yet, it’s introduction produced a worldwide wave of Oooh’s and Aaahhh’s. Shortly thereafter there came devices you could take with you to the store to calculate the total price of purchased items …. in your pocket!! Ooooh, aaaahhhh. I don’t have to go deep into the rest of history, most people know how the computer has revolutionized our way of living. My point is to make it clear how the world of business changed with the advent of computing mechanisms; how recently this revolution began and how computers make it possible for companies (especially) to mow through thousands of transactions, calculations and a high number of other tedious subprocesses without lifting a finger. These are processes that have been integrated into your spreadsheet programs, your databases and your operating systems and have paved the way for faster production, innovation and evolution of human society.
30 years ago the public would have been very suspicious of the idea that today there would be cash registers in stores which you could just wave a series of products in front of, in any order, and the cash register would all by itself recognize the product and calculate the price. Even the change left over — and maintain a correct inventory while doing so! Sure it’s just a barcode being scanned, or an RFID, but nonetheless — automation has come a long way, we’ve become so used to it that we often fail to see how incredible it is.
The fundamental difference between cashiers and politicians, is that politicians work with natural language. They work with images and graphs in various different formats. They don’t work with flat, standardized formulas and barcodes. We can’t represent a terrorist threat or the happiness of people with a simple formula. These are extremely complex issues that must be dealt with through a combination of descriptions, images, formulas, videos and other media.
Extending the Applications of Automation
Finally, this leads me to my point. Management, politics and beurocracy are extremely tedious. They deal with complex issues which do not only concern the general happiness of all humans, but their safety as well. With these responsibilities come tons of paperwork: As many different views and opinions possible must be gathered and the irrelevant ones filtered away in order to make a decision. How do we choose what is relevant to the decision? By collecting more views and opinions of what is relevant and then filtering the irrelevant ones away, and so on — *Shivers*.
Now imagine that politicians could wave their 300 page report on how to lower taxes in front a machine and the machine would produce a 10 page summary, with added recommendations of risks and additional factors that the report doesn’t account for. Moreover, imagine that a 300 page report could be generated by a computer automatically by having it listen to the debate of politicians during a meeting of congress, with references to related debates over the years. Or that tons of computers would work day and night reading and evaluating the past 50 years of political history, finding patterns, ways to cut prices or methods for reducing crime. A.I. has already been tried (with ok results) to find correlations between crime-cases over long periods of time, identifying patterns in data amounts too vast for a group of detectives to go through. We’ve all laughed at old laws that are still in effect but have been forgotten, what prevents us from using this technique to point out system faults, figure out how certain financial laws affect startup-companies over a period of 30 years, or to scan political history for information related to the latest city construction plan?
The examples don’t have to be so extreme, I simply enjoy thinking big. Taking a less futuristic example, the semantic web (Web 3.0) is all the rage these days: With only a bit better semantic-annotation (associating meaning with data to allow easier searching and machine processing) of political documents it would relieve the bureaucratic bottlenecks manifold. It doesn’t require a science-fiction based A.I. to improve or take part in the political process. It just takes a little automation.
Society is becoming increasingly complex. This is an accepted fact for the world wide web; and A.I. or machine automation is considered the solution to this problem. It’s a smart choice to let computers help handle the amount of data produced every day. And in fact, there are already A.I. technologies built into desktop computer software (databases, user interfaces, etc.) even though many don’t realize it. But for some reason, I never hear of explicitly directed research on how to use A.I. technologies for governmental purposes other than to kill people (war applications). The reason I wrote the 13th law of ISIR was to try and bring some attention to these issues, and to possibly inspire someone to do research projects or find methods for automating political and managerial processes. Not necessarily to see anything noteworthy happen in my lifetime, but to provoke thought and push things forward.
What does it all mean?
Will we see computers (in the next 30 years) make it possible for politicians to mow through thousands of motions, reports and a high number of other tedious subprocesses without lifting a finger? Thereby giving them more time to think about the big picture — about what people are suffering from, or which people are gaining something — instead of having to ravel in the headache of bureaucratic bottlenecks?
Well. We’re a long way from HAL 9000. Even though ASIMO is already running, I don’t dare to envision him running for president anytime soon. But there are still amazing things that can be accomplished without near human-level intelligence in machines. Increased automation gives us more time to think and find solutions, no matter in what form or what order of magnitude. I do (justifiably) believe that with a concentrated effort, A.I. and automation technologies can significantly improve the bureaucratic process.
Time. If only time could keep up with the birthrate of my ideas, then maybe I’d have a stronger feeling of accomplishment. Unfortunately, I get so many ideas that I only have time for a fraction of them — which results in a feeling of underachievement. Know the feeling? Of course, the reality of it is that I’m a workaholic and I am achieving things, thankfully always learning new methods to increase my productivity. The irony of it is that increased productivity means I have more time to get new ideas. So a feeling of underachievement remains the same. The curse of the creative breed.
So what can one do? Do you concentrate your gray matter on only a selected few ideas with the possibility of one of them taking off as something extraordinary — or do you spend time on as many as you can with the possibility that they’ll all fall into the shadow of eachother, each receiving limited attention and therefore never becoming a finished product?
I try to take the third approach, which is a combination of both: Don’t stick to a certain plan in these matters, maintain balance. Weigh each idea individually. Some of my ideas are long lasting projects while others receive limited attention and for the most part remain a written word, locked away in one of my many black books. This has worked out ok, aside from the problem I originally stated above on feelings — which can thankfully be partially overpowered by logic.
But a key element is writing down the ideas. As many as I can (I shudder at the thought of how many ideas I didn’t write down and have now forgotten … and have no idea I ever had). Often I read notes that I wrote a few years ago and find that I’ve outgrown them somewhat fierce — some even cause a small chuckle. Others remain intriguing, and this is actually a great way to identify how good your ideas are. If they survive a few years in a black book without execution, and are still good when you read them again — then they’re probably worth your precious time. By doing this you also automatically get better at identifying key elements of good ideas, and key elements of bad ideas. Gradually, you’ll gain a deeper sense of which of your ideas are good, a lot faster.
Next -
Get Think Artificial posts delivered via RSS, or get e-mail updates by navigating the links below
I first read about Twine today in a swedish newspaper...Think Artificial is a proud member of the 9rules blog community.
