It’s Thursday, and it looks like I’m back with another tech-related whine.
I’m a quicksilver user, and I have to say that I adore this program more than just about any other on my computer, and it’s one of the reasons why I’m always so insistent about using my computer: it’s downright unsettling to use a computer that doesn’t have QS on it.
What’s quicksilver? It’s a nifty little application launcher that lets you interact with your computer via text inputs. Do a google search and you’ll learn way more than I could really hope to tell you here. Anyway it’s awesome, and it’s free. If you use a mac, get it now. If you don’t there are some things that are close in some of the functionality, but the truth is that so much of what makes QS so good, is the fact that it draws on all kinds of unity that already exists in the mac, and makes all that connectedness appear at your finger tips without having to futz through menus and folders and so forth.
So this got me thinking: modern computing (ie. the last 20 years) has been centered around the GUI (graphical user interface) that represents file structures via pretty pictures. So most people are used to interacting with their computers via pictures and what not. (You UNIX nerds, be quiet!) And this is really good for letting people intuitively figure out how to make a computer work. It’s the reason that most people don’t have a real problem moving between Macs and PCs these days. The down side: it’s really fracking slow for a good deal of what we do.
There are a couple other pieces of this mind puzzle that I really don’t remember the source for, and I’m sorry for that, but they added to the mileux of this whine.
The first is that I saw an article that noted how in the hands of a skilled user the address bar of Firefox can function like a rather smart command line: If you type in a collection of words, depending on your settings Firefox will take you to the first google search result (which is often what you wanted) or it will take you to the google search result for those terms. You can augment this by adding extra codes in the settings, so that prefacing a string of terms/search operators with a key (like wiki for wikipedia) will perform a certain kind of search. It’s pretty nifty, and to be honest I think it was an article in a blog that I glazed over because I’m not using firefox these days (it’s all about Camino for me!)
The second thought was that someone mentioned, again in a blog or podcast, that if you watch “power” users even if they’re not working in a command line environment don’t tend to use the mouse very much. It’s inefficient, it takes too long, it can be hard on your wrists, and it’s a lot easier to hit control/comannd+C, control/comand+V to copy and paste than it is to interact with the edit menu. The truth is that most of the commonly used features have associated commands and by just studying this, you can use the mouse a lot less frequently.
I don’t want to give up all GUI features, and go back to olde school command lines, mind you, as there are some things that just work better with mouse actions: Web browsing. Document Preparation. Graphics manipulation. Sound Editing (sometimes). And so forth. But file manipulation? Task switching? Text editing? Command lines are way better for that.
Do I have a solution? Of course not. I think something a lot like quicksilver but that had the permanency of the menu-bar in OS X would be ideal1. I think the real challenge though is finding a way to make this kind of human-computer interaction seem as intuitive as pointing and clicking. I think with some smart technology it could be done.
do you have a whine about technology? send it in or leave a comment. I’ve written this for two weeks in a row, and I didn’t quite intend this to be another series, so if you like it, it might be up to you to continue…
best, tycho
-
I mean what I seem to be calling for is a re-imagination of computer interface, and I think the window paradigm is something that might need rethinking, as part of this process. Because if we have a list of running/active windows/programs/applications that is easily accessible from some sort of text input, do we really need to have such a confusing graphical representation of the fact that our computers are doing all of these things at the same time? Arguably not. Just a thought. ↩︎