Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Comment Re:or not... (Score 3, Insightful) 569

Executive Summary: The mouse is faster than the keyboard.

Or not.

Here is the article where Tognazzini describes his test. Tognazzini writes:

The test I did I did several years ago, frankly, I entered into for the express purpose of letting cursor keys win, just to prove they could in some cases be faster than the mouse.

Note, "cursor keys", not "keyboard".

I typed in a paragraph of text, then replaced every instance of an "e" with a vertical bar (|). The test subject's task was to replace every | with an "e." .... The average time for the cursor keys was 99.43 seconds, for the mouse, 50.22 seconds.

Never mind the absurdity of reporting the times to four significant digits. He said, again, "cursor keys", not "keyboard". He had the users move the text cursor with the arrow keys alone, from one "|" to the next.

Here's another way to do it, using the keyboard. Got your stopwatch?


Six seconds, independent of the length of the paragraph or number of changes. (That's ed(1); "ed is the standard text editor".)

Even if you constrain the user to move the cursor to each "|", one by one, the keyboard is faster: for instance, in vi(1), "{/|^[re" and then repeat "n." But why would you make the user do that? That's not just ignoring the utility of the keyboard, but of the computer itself. So the mouse is faster than the arrow keys at performing task X forty-two times? If you use the computer as a fucking computer instead of crippling it to the level of a typewriter, then you don't do it forty-two times; you do it once. Tognazzini's test suffers from Mac System 6 tunnel vision.

It might be argued that automated repetition defeats the true purpose of the test -- that it isn't about replacing "|" with "e" forty-two times, that that isn't a real-world editing task but just a stand-in for forty-two different tasks.

Better for the keyboard! A keyboard does have keys other than arrow keys -- it has keys that bear the very same characters that appear in text. There is an obvious correspondence between a character on the keyboard and a character in the document, one about as "intuitive" as you can get. This lets the user press the keys to locate the corresponding character in the document, either individually, or sequentially to magically form composites we call "words" that have meaning within the user's task.

Using the keyboard, the user can have the computer find the correct location, rather than being forced to do it himself, visually, with the possibility of error. What if Tognazzini's test had not involved finding the vertical bars, which are visually distinctive in text, but, say, replacing "blue" with "green" throughout a ten-page document? How many instances would have been missed? Do you want to cut the blue wire, or the green one? Are you sure?

(Oh, I'm sorry. Did I say "|" was visually distinctive? Here you are, user: take your mouse and change every "|" in this Helvetica paragraph. Don't touch any "I" or "l" or "1", though.)

The mouse ignores the semantic content of the characters and symbols, words and keywords, blocks and sentences.... It even ignores the symbols themselves; it wanders haphazardly over a picture of the document (a static picture, if you're lucky; ever try using a mouse to select something that doesn't hold still because the window is being written to?)

Revised Executive Summary: The mouse is faster than the keyboard that has nothing but four arrow keys, when errors don't matter.

Slashdot Top Deals

Too much of everything is just enough. -- Bob Wier