Luke Breuer
2010-05-20 18:44 UTC

two kinds of thought
Thomas Edison, inventor of the [long-lasting] light bulb, is known for saying this:
Genius is one percent inspiration, ninety-nine percent perspiration.

I am not sure what the percentages are today, but I sincerely believe that it is possible to use technology to reduce "perspiration time", and therefore increase "inspiration time". Now, these terms are not very well-defined; Josef Pieper, in Leisure: the Basis of Culture uses two terms which are more clear (p28):
  1. ratio, routine, or repetitive thought (pronounced "rah-tee-oh")
    Ratio is the power of discursive, logical thought, of searching and of examination, of abstraction, of definition and drawing conclusions.
  2. intellectus, mature, or creative thought
    Intellectus, on the other hand, is the name for the understanding in so far as it is the capacity of simplex intuitus, of that simple vision to which truth offers itself like a landscape to the eye.
I like to think of ratio thought as gathering information, processing it, and turning it into puzzle pieces. Intellectus thought, on the other hand, is fitting those puzzle pieces into a whole. My question to you is this: how much time do you spend doing #1, vs. #2? Do you think there is a way to reduce time spent on #1?
work you didn't know you were doing
Any time you use a computer and want to tell it to do something in a language it does not understand, you are forced to break that task down into smaller pieces, until the language you are using is compatible with your computer's programming. This full problem is known as natural language understanding and is a very complex one, but what if we were to break it down into smaller pieces? There are two ways to do this:
  1. When you break down a task such that a computer can accomplish it, you usually do so with multiple levels of "hierarchy". What if, instead of eliminating that hierarchy, you could simply remove one level?
  2. Some individual tasks are made much harder than necessary, often by lack of uniformity: because many people all present information or functionality in different ways, you have to translate your idea into one or several of them.

An example of #1 is searching and aggregating: let's say I want to obtain a histogram of my MP3s, by play time to the nearest minute. Except perhaps for specialized software tools, I would have to do quite a few things manually. Instead, I would ideally want to say:
  1. generate a histogram,
  2. based on the items (MP3s) in these collections (folders),
  3. by extracting this attribute,
  4. and performing this function (rounding to nearest minute) on the attribute
I ask you to imagine a software system that lets you do all of the above in less than thirty seconds. Given the type of "speed-up" this example demonstrates, what else might be possible?

An example of #2 is following references in academic papers. To do so, one needs to:
  1. scroll from a citation to the reference
  2. copy the reference text
  3. search for it online
  4. possibly enter in credentials (because most academic papers are not free to access)
  5. download and open the referenced paper
  6. find the referred-to section of that paper
The Internet was supposed to turn all these steps into a simple point-and-click. What happened? What other operations require this much needless effort?
problems to write about