In her blog, Shelby mentioned a book about how to read books. Which sounds silly on the surface but really does make some sense once you think about it – most people don’t give much thought to this, and just sit down and read one page after the next. Can this be optimized? Is there a better way to understand what you’re reading more fully?
For me this brings to mind a concept I learned about in a Great Texts course, called the “Hermeneutic Circle.” The idea behind this is that, in theory, it is impossible to fully understand any written work – in order to understand the significance of the work as a whole, one must understand the parts, but in order to understabd the significance of the parts, one must understand the the work as a whole. One apparent solution to this problem would be to read everything at least twice, but according to the circle that doesn’t really solve anything because you never gained a genuine understanding of any element of the work in the first reading.
This concept and dilemma of a person not being able to see all elements of something simultaneously isn’t unlike the window metaphor discussed in class, so it does call to mind the kind of augmentation Engelbart talks about.
Nelson, on the other hand, favors computer systems that are very economical to learn and use – sort of like an assistant and not quite like the helper who can do things you truly aren’t capable of that Engelbart seems to envision – he includes examples of how flow charts and such could be used to analyze complex data. Indeed at some level this could be (and is) useful, it’s great for a simplified overview of something, and sometimes even illustrates or communicates a concept better than a traditional format could. However, I think that, like much of the Nelson reading, which may or may not present his ideas fully, he fails to acknowledge the possible downsides of his own ideas and the advantages of current systems over what he is proposing – he pretty much dismisses the education system as it is right now, and accepts his own version as a seemingly flawless solution. I think ease of use has a limit and that it shouldn’t be a priority many areas – a complex system that is only useable with substantial training and not intuitive in the least could, after all, be faster and more powerful than a version that is easier to use; the latter may save time initially, but an intial investment in training for the less intuitive system could more than compensate for this.
To illustrate what I mean by this I’ll try to turn one of Nelson’s examples on itself. On page 324 he writes about a control module on a Computer Assisted Instruction system, explaining how it is not organized in a logical manner that categorically complements the functions of the different buttons. He then presents a possible approach for making the user’s options clearer, by dividing the buttons into three sections. This makes for more logical organization than clustering all nine buttons together, but in doing this the module has also become more difficult to use in a sense – the buttons are spaced further apart, such that it may require both hands, additional hand movement, or at the very least looking down at the module in order to use the buttons. How is this better than just memorizing the locations and functions of the buttons, such that one could eventually be able to effortlessly use the control module with one hand resting in place?
Using Nelson’s logic I could propose a keyboard that would provide “clearer options:” put all the vowels in one column, and put the consonants on the other side, categorizing them into rows with continuants first, followed by plosives, all categorized alphabetically. This might seem like a logical, categorical way to break down the keyboard if you’ve never used one before, but in practice it’d just be stupid. Granted, the QWERTY system was developed mostly from the goal of reducing typewriter jams and is only easy because we’re used to it. (Though QWERTY is the most commonly used, the Dvorak Simplified Keyboard configuration has proven to be the fastest and most efficient.)
Another example in which Nelson’s theory of simplicity doesn’t hold true: it’s very easy and intuitive to use the start menu in Windows to find a program. But this isn’t the fastest way to do anything. The fastest way to execute an action in Windows, if you know how, is to use a combination of the MS Dos prompt and hot keys – you don’t even need a mouse for just about everything. This way isn’t simple, it requires a lot of memorization of commands and hot keys, but it’s much faster for executing complex tasks if you need to do them regularly. Additionaly, you can do things in a prompt that you can’t do in Windows normally, so you’re actually losing substantial functionality from simplifying everything.

