First off, comparing keyboards and mice in this fashion is a bit misleading, because there are probably some tasks you can perform faster with a mouse, and some you can perform faster with a keyboard. So testing for a single task (which doesn't even appear in your quote) isn't saying much.
More to the point, I can say from my experience as a developer both on Windows and on UNIX, that using CLIs is much faster than using GUIs when coding. IMHO, this has two main reasons:
Since the keyboard is much more versatile, properly-written applications can take advantage of the many possible inputs. A perfect example is Vi, which utilizes most if not all of the keyboard's potential. In what other popular editor can you perform 'take the current code block, search for all the lines that contain the word XXX, and comment them out', in less than 20 keystrokes? This is power that just cannot be achieved with a mouse.
Using a mouse/keyboard combination in which you have to constantly switch between the two is very counterproductive, in the sense that it keeps breaking your line of thought. If you let your mind concentrate on one task for a certain period of time, it will eventually 'get into it', and you will start performing that task faster, with less errors, and in a more concentrated manner. This phenomenon isn't limited to computing -- it is well known among musicians, athletes, etc., and it also applies to coding. If you use only the keyboard, your hands don't need to move much, and your mind can 'get into' the typing. If, however, you have to send your hand searching for the mouse every 10 seconds or so, this cuts off your line of thought, and you never really 'get into it'. (some call this the Zen of Coding).
And then there's RSI, which is much harder to avoid with mice than with keyboards. But that's only if you care about a pontential disability which can prevent you from using keyboards and mice for the rest of your life..
I think you're looking at OSS from the wrong direction. OSS development starts when a programmer has a personal "itch" he wants to "scratch". Take Linux for example: You say it was a designed to be UNIX-like, which is correct, but completely misses the point. Linus developed Linux because there was no good OS for his i386 -- that was his "itch". The fact that the resulting OS looks a lot like UNIX reflects a design decision made early on -- not to reinvent the wheel, but rather to learn from many years of other OS designers' experience.
Linux is only one example of what drives OSS prjects ("The Cathedral and the Bazzar" includes a good account of a classic OSS project).
Now, in the last couple of years, with the MS vs. Linux Armageddon and everything, the "itches" have become more ideological and less personal. It's fair to assume that not everyone who contributes to KDE really needs or uses KIllustrator, KMail, KWord, etc. I'm certain many contributors do so to contribute to make Linux triumph over Windows. This is why you see so much of what you call "imitation" -- we're trying to provide a desktop alternative to Windows, and the simplest way to do it is to create similar applications on Linux. Also, if these applications will have the same look-and-feel as their Linux counterparts, migration for users will be easier. I'm not saying we should copy MS's IMHO ugly UI designs, but it's something to think about.
I think now it's easier to understand why we don't see more "innovative" stuff being developed. Something like.NET doesn't really scratch anyone's personal itch, so it isn't invented by OSS people.
Slashdot Mirror
More to the point, I can say from my experience as a developer both on Windows and on UNIX, that using CLIs is much faster than using GUIs when coding. IMHO, this has two main reasons:
- Since the keyboard is much more versatile, properly-written applications can take advantage of the many possible inputs. A perfect example is Vi, which utilizes most if not all of the keyboard's potential. In what other popular editor can you perform 'take the current code block, search for all the lines that contain the word XXX, and comment them out', in less than 20 keystrokes? This is power that just cannot be achieved with a mouse.
- Using a mouse/keyboard combination in which you have to constantly switch between the two is very counterproductive, in the sense that it keeps breaking your line of thought. If you let your mind concentrate on one task for a certain period of time, it will eventually 'get into it', and you will start performing that task faster, with less errors, and in a more concentrated manner. This phenomenon isn't limited to computing -- it is well known among musicians, athletes, etc., and it also applies to coding. If you use only the keyboard, your hands don't need to move much, and your mind can 'get into' the typing. If, however, you have to send your hand searching for the mouse every 10 seconds or so, this cuts off your line of thought, and you never really 'get into it'. (some call this the Zen of Coding).
And then there's RSI, which is much harder to avoid with mice than with keyboards. But that's only if you care about a pontential disability which can prevent you from using keyboards and mice for the rest of your life..Linux is only one example of what drives OSS prjects ("The Cathedral and the Bazzar" includes a good account of a classic OSS project).
Now, in the last couple of years, with the MS vs. Linux Armageddon and everything, the "itches" have become more ideological and less personal. It's fair to assume that not everyone who contributes to KDE really needs or uses KIllustrator, KMail, KWord, etc. I'm certain many contributors do so to contribute to make Linux triumph over Windows. This is why you see so much of what you call "imitation" -- we're trying to provide a desktop alternative to Windows, and the simplest way to do it is to create similar applications on Linux. Also, if these applications will have the same look-and-feel as their Linux counterparts, migration for users will be easier. I'm not saying we should copy MS's IMHO ugly UI designs, but it's something to think about.
I think now it's easier to understand why we don't see more "innovative" stuff being developed. Something like .NET doesn't really scratch anyone's personal itch, so it isn't invented by OSS people.