This article covers tools and tricks to help make your daily work
easier. It will hopefully enlighten you as to how to make your
computer environment faster to access, allowing you to get work done
faster and with less effort. Specifically, it covers my work
environment and efforts I have made to make working easier.
I'll be the first to admit I try to put minimal effort into most
things. However, do not confuse "minimal effort" with a lack of caring, or
laziness to achieve a goal. It is a desire to do the least amount of
work and have the most amount of useful work produced.
For those who have taken calculus, you might recall the "maximization
and minimization" types of problem sets. You know, Farmer Bob has
1500 feet of fence and wants to make the largest area with it. Well,
forget about calculus and Farmer Bob, here. The idea still remains
though, you have resources and demands. That brings me to
economics, the science of choice.
Economics studies choice. You, as a worker, have finite resources
(time, energy, patience, budgets) and potentially infinite demands. You
may have the time and energy to work on something, but it just isn't
working out so you run out of
patience. Whenever you run out of
a resource it becomes worthwhile to step back and analyze why you ran
out of said resource, hopefully to come to a conclusion that, in future
situations, will conserve your resources without hindering output.
Interfaces, of any kind, are almost always a topic of heated discussion
among geeks and nerds alike. You'll find there are many different kinds
of computer nerds out there. Some are called "power users" and stick to
more, so-called "raw" computing environments. Some folks like full
desktop environments such as Windows, KDE, GNOME, or OS X. Others are
in between. There are also many kinds of mice, keyboards, and other
input devices.
Everyone uses mice, right? What it really comes down to is how much
do you use your mouse? Some fanatics may even wonder why you need a
mouse when wonderful things like keyboard shortcuts exist. So let's
lay it out like it is: The mouse is a pointing device. You use it to
indicate where you want actions to occur, usually with a
click of some sort. How do you know what the consequences
of your
click are going to be? Certainly you can't use a
mouse with no eyes, right? How much can you do with a mouse while
your eyes are closed?
If you are in a Desktop Environment (Windows, KDE, OSX, etc), try
this exercise using only your mouse (read this through once first):
- Open a new browser window (not a tab)
- Move your mouse to the bottom left-hand corner of the screen
- Close your eyes
- Try to close that new browser window
- Go back to the the other (this) browser window
- View the page source
Were you able to do it? Chances are if you were in Windows and your
new browser opened up full-screen, then you should be able to do
atleast the first step, but what about OSX? or KDE? In case you
failed, here's how it can be
done in Windows:
- If the new browser window opened up maximized, then you are
in luck! Windows has a GREAT feature that you can click at the
far, top-right corner of a maximized window and it will close it.
If you've never noticed this, this allows you to close windows by
aiming for an easy target, the top-right corner of the screen,
and clicking. Kudos to Windows for this feature.
- Assuming you got that far, the next trick is to figure out
how to view the page source. With the mouse, there's two options.
The first is through the main menus (View->Page Source in
Firefox). The other is by accessing the context menu for the
page. The page is a large area you can probably hit blind. Right
click, now what? Aim down and hope for the best. Right?
The problem with mice is that it requires precise hand-eye
coordination, or you end up orbiting your intended target until you
slowly zero-in on the location you intend on pointing at. UGH! My
favorite example is the Windows calculator (or xcalc for X11). These
calculators have a similar look-and-feel to normal calculators. If
you try and "type" numbers using the mouse and the button-based
number pad, you're doomed to slowness and inefficiency. Use the damn
keyboard.
What mouse-usage comes down to is finding a mouse that suits your
style of use. There are many kinds of mice: trackballs, touchpads,
trackpoint (nipple), standard mouse, etc. Try as many as you can,
figure out which one feels right for you to get the job done best.
Don't get me wrong, I love my mouse. I does a great job of pointing
at things. However,
muscle memory
is infinitely superior to that of hand-eye coordinated efforts. Why?
If you can touch type, it's because you have trained yourself such
that your fingers "know" where the keys are based on where your
fingers are. The keyboard requires no hand-eye coordination if you
can do this. The mouse almost always requires hand-eye coordination
in most environments.
Notice I said the keyboard requires
no hand-eye coordination.
That's right. zero. You may not have noticed, but most (all?)
keyboards will have raised indicators on the positions of the
keyboard where you should place your left and right index fingers.
If haven't noticed this, lift up your hands and look at the 'F' and
'J' keys on a standard US keyboard. The bottom of the keys should
have little nubs. This helps you find the home row without looking,
simply brilliant.
Keyboards have been pretty much the same since QWERTY came around.
You'll find small differences between manufacturers: Sun keyboards
have control/caps switched, backspace shifted down one row, tilde
elsewhere, escape where tilde used to be. Apple makes some keyboards
with keys half-height keys you almost have to look for before
attempting to strike them (Requires hand-eye coordination,
UGH!). Rarer keyboards may vary wildly, but most you'll
find are similar.
There are also two kinds of keyboard "keys" you'll find. There's
the common, big keys found on standard desktop keyboards. There's
also laptop-style keyboards with keys that use "shallow scissor"
keys, which are much shorter have much less travel required to
press than standard keyboards. Some keyboards you'll find have
"hotkeys" on them that do magical things such as tweak sound volume
levels, open your mail client, etc.
You have a mouse, you have a keyboard, what now? Graphical User
Interfaces have existed for almost 25 years, and still nobody agrees
on which is the best. The graphical environment provides you
wonderful toys such as a mouse cursor, and colored icons in your
file manager. Most graphical environments have fun little widgets
like buttons, checkboxes, textfields, windows, icons, etc.
Your graphical environment probably provides you with a blended
interface relying on both keyboard and mouse interaction.
Unfortunately, many things can only be done with the keyboard and
others can only be done with the mouse. Good graphical interfaces
will provide both keyboard shortcuts and fast mouse interfaces to
let you, the user, decide which you want to use.
Many graphical environments that use the X window server have some
neat innovations Apple OSX and Windows lack:
- Grab-and-splat: Applications will have a grab-and-splat style of
mouse-oriented copy-and-paste. First, you "grab" text by selecting
it with the mouse. This activity almost always copies the selected
text to the clipboard, so forget needing to hit
Control+C or fumble around looking for a "copy" option.
The "splat" part refers pasting by using a mouse click to indicate a
paste action. It's always been the middle mouse button in my
experience.. This lets me copy and paste using only a few clicks
and no keyboard interaction.
- Easy window movement: Most window managers
I've used allow you to "grab" a window ANYWHERE simply by holding
the
ALT key and dragging a window with the left mouse
button. OSX and Windows make you hunt for the title bar to drag it.
This is my favorite feature, ever.
- Easy window resizing: Again, many window managers allow you to
again "grab" the window anywhere. This time, if you use the
right-mouse button (or middle mouse in GNOME) while holding
ALT you can drag the window in any direction and it
will resize the window. Simple and fast.
- Mouse-focus: focus-follows-mouse is another of my favorite
innovations not found easily OSX or Windows. This means that
whatever menu your mouse is hovering over gets focused, meaning you
can type in it. This doesn't mean the window gets moved on top of
others, just that the keyboard events go to it. This is so
useful.
- Bring Forward/Send Back: Windows are in layers. You can put a
window on top of this browser window, right? Many window managers in
X allow you to send a window to the bottom layer by using
alt+rightclick. You can bring a clipped window forward
by using alt+leftclick
Nothing quite makes me angry like having to hunt down the title bar
or the tiny "drag me to resize" widget you find at the bottom-right
of many OSX applications. No one should have to be that precise with
the mouse to do simple actions such as resizing or moving a window.
I wonder why nobody at these big companies has discovered how useful
these features are. Think about it, what does alt+rightclick+drag do
in a window in OSX? Nothing, right? Why not make use of it? I should
mention, however, that Windows XP appears to support mouse focus, or
sloppy focus, (older versions may too) by flipping some
options in the registry. TweakUI should be able to enable this for
you.
I mentioned
patience being a resource earlier in this article.
Patience is how much irritation you are willing to put up with in your
computing environment before taking a baseball bat to your monitor.
Your mileage may vary, but personally, patience is the first thing I
lose, and it's the first reason that causes me to find less irritating
activities. This becomes a huge problem when the activity that is
wearing on my patience is one that has an impending deadline! When a
deadline is bearing down on you, you cannot afford to be detered by a
lack of patience for the computing environment you need to be using to
complate said activity.
Bringing back the
Principle of Least Effort, I am wholely
willing to spend resources (time) when I have nothing better to do.
This is called boredom. I love learning new things, discovering new
tricks, etc, so when I have some free time to kill, I often kill it
learning new tricks or enhancing my personal computing environment.
This make any required future tasks consume less of my patience,
allowing me to work better and longer on tasks I need to get done.
There are many questions you should ask yourself: How much do you
know about your preferred computing environment? What can you do
quickly? What can't you do quickly? Have you configured said
environment to your preference? If not, why not? The list goes on
and on, but the point is, you need to evaluate how you interact with
your computer and decide if you find it efficient. If it is not
efficient, find better means by which to interact. This means
experimenting with software tools, learning keyboard shortcuts,
tweaking interface options, and more.
Seeing as how muscle memory is faster than hand-eye coordination,
you should do your best to learn all the keyboard shortcuts you
can. This means reading documentation, taking mental notes of
keyboard shortcuts when using the popup and main menu systems, and
even making your own keyboard shortcuts if the program allows.
Using
Control+V to paste is always more efficient than
fumbling around with your mouse to find the paste button or
invoking a context menu and finding the paste option.
Mouse gestures are fantastic if your hand is glued to your mouse
most of the time. This is especially useful when you are using a
mouse-driven piece of software, such as a graphical file browser or
a web browser (such as Opera). I equate mouse gestures to the
rodent-version of keyboard shortcuts. They're efficient and can
quickly become muscle-memory reactions. Unfortunately, many programs
do not support them.
Opera is an
ad-supported web browser that supports mouse gestures by default.
Firefox has extensions you can install to add mouse gesture support.
Mouse gestures are just that, gestures you make with your mouse
pointer. A short horizontal movement to the left while holding the
right mouse button may indicate that you want to go
Back in
your web browser, for example. Mouse gestures are great, and can
often be combined with keyboard combinations; for instance, holding
shift while holding the left button and dragging the mouse upward.
The only problem with mouse gestures is that they are not easily
discovered, that means that you'll probably have to read the
documentation to figure out what the available mouse gestures are
and what they do. Fortunately, many programs that do support
gestures support configurability of those gestures. This means you
can make things like "copy text" and other common tasks execute when
you perform certain gestures. Fantastic!
There's problems with mouse gestures, however. They only seem to be
truely useful with standard mice, for myself anyway. I use a
trackpoint (nipple) mouse, so directional precision is somewhat
difficult to maintain when moving quickly. That means my attempt to
"click-drag right" may end up turning into "click-drag down-right"
depending on how error I make. Your mileage may vary.
Pie menus, also known as radial menus, are extremely useful elements
of mouse-driven software. I consider them to be graphical relatives
of mouse gestures. While mouse gestures are simply mouse movements,
radial menus have important graphical hints which make learning them
very quick. These types of menus are circular instead of your normal
rectangular, list-style, menuing systems. You are presented with a
circular menu with a few choices, if the item you chose has a
submenu, then selecting that submenu will bring another circle out
so you can choose elements from that submenu. The cool thing is most
implementations I've seen of this allow you to indicate submenu
selection by simply mousing in the direction of the submenu. Fast
and easy to learn? I like it.
What's a keyboard map? It's a table that your Operating System uses
to determine what key you pressed. Your keyboard sends a numeric
identifier for whatever key you pressed. This numeric identifier is
looked up the keyboard map table and is translated into a key press
that applications can recognize. This is a critical point to
understand. This map can be altered to make different keys act
however you please. One common alteration is to make the
Caps
Lock key act like a
Control key. Who uses
Caps Lock these days anyway?
Your tool/accessory options are limited to the kind of computing you
do. Certain software tools are available for X that aren't available in
Windows or OSX. The alternates are also true. This boils down to the
fact that if you want to make yourself more productive on a computer,
you're going to need to find the tools which give you a productive
edge.
I know of only two utilities that help you modify useful parts of
the Windows interface, they are below. If you know of more, please
let me know.
You can probably find more by googling for "windows tweaks" or
somesuch. I don't use Windows anymore, so you're on your own :)
I hate Macs, therefore this is a very thin section. Perhaps you
should email me with software/tool suggestions and descriptions?
-
xkeycaps - a
fantastic program that lets you remap keys on the fly in X11
and save the changes.
-
apwal - launcher
application that puts icons near your cursor when executed. Choosing
an icon executes it's program. Fully configurable, I like this idea
Again, I'll mention muscle memory is fast. So I tend towards
environments that lend themselves to activities that can be learned in
muscle meory. What do I use? Anything that's mostly driven by the
keyboard.
Right, so let's get down to it. What do I use in my environment?
-
The first thing I usually do when configuring an environment to my
preferences is changing the
Caps Lock key to act like a
Control key. Caps Lock is useless! Other keymap
changes may also occur: sometimes I map F1 to escape, for example. Or
right-alt to windows key.
-
The next thing I do is get familiar software and configure it. If you
don't configure software you use commonly to suit your needs, then you
probably aren't using it as efficiently as you could. The first thing
when I login to a new system is copy my rc files (zshrc, vimrc, etc)
over.
Once those two tasks are done, I'm pretty much set to be comfortable
and efficient. However, you might ask what particular software and
configurations do I use? I'll start with vi.
- Comes standard in some flavor on almost every unix-like os
- Extremely muscle-memory friendly
- Very configurable (depends on flavor)
I've been using vi since I started using anything unix-like. This
was about 4 years ago. For those out of the loop, vi is a
keyboard-oriented text editor that comes in many flavors. The
principle behind vi is it's two-mode style of interface. In
insert mode, you can type into your document as you would any
other editor. However, if you hit Escape while in
insert
mode, it puts you in
command mode.
Command mode
presents the user with many different kinds of editing and movement
commands with single keystrokes. This gives you a rediculously
powerful editing system that is also very fast to use. The only
caveat is that there is a steep learning curve, becuase the
interface is far from intuitive.
If you spend the time to learn any vi commands you can, you'll begin
editing text faster and faster. The more commands you know, the more
effectively you can use the editor. The more often you use these
commands, the faster they are learned in muscle memory.
vi comes in many flavors. The most popular open source ones being
nvi and vim. Both are very configurable and allow you to change most
behavioral aspects of the editor. Vim has far more features than can
be described here along with very extensive documentation.
A nice benefit to knowing vi commands by heart is that many unix
shells sport vi-compatibility modes. Bash, zsh, tcsh, and ksh all
have vi-style input modes which let you line-edit with vi commands.
This is quite sexy and makes line editing in the shell faster.
Over the past 4 years of my unix experience, I've used many
different window managers. I started using WindowMaker then moved
to BlackBox and others. All of those window managers were very mouse
oriented and made much more efficient use of the mouse than Windows
ever has.
Enter ion. Ion is designed to be used primarily with the keyboard.
It's original implementation was based around "tiling" workspaces
and tabbed frames. Workspaces are split into frames that client
windows (xterm, etc) can be launched into. Frames can contain
multiple clients, discernable by tabbed title bars. Ion does what I
believe a window manager should do: manage windows. With the normal
framed workspace layouts, I never have to worry about where windows
are goign to show up. I never have to move windows around myself to
make them fit on the screen better, etc. Most of the effort required
in using a graphical windowing environment is handled automatically
by the window manager.
Ion has grown over the years and now supports things like floating
workspaces (they act like normal workspaces you're used to using)
among many other features. It's greatest benefit is that most of the
window manager's features are accessible through it's scripting
engine, Lua. It supports neat things like winprop kludges so you can
tell ion what it should do with certain kinds of programs and
windows.
Just about everything can be configured, from themes to
keybindings. The default settings are reasonably sane and learning
how to use it is fairly simple as there's a lot of documentation
online. If you browse through the keybinding config files that come
with it, you can quickly learn how to use this window manager.
Ok, so maybe I exagerated a bit when I said bears. At any rate,
Making effective use of your time depends wholely on the tools you
have available. There's a saying: "When all you have is a hammer,
everything looks like a nail." This is a very unfortunate truth.
Most of us just want to get the jobs done and move on with our
lives. The problem is, that this attitude is a hinderance. You need
to acquire the proper tools to get a job done to satisfaction. If
you are lacking in the proper tools, find them or make your own.
You'll enjoy free time later when you didn't half-ass something only
to have it not quite fit.
Many friends have pointed out that I "obviously" hate graphical
interfaces the only programs I use are console-based (with one
exception, Firefox). Whether I do or not is out of the scope of this
article. The point is, I use primarily console based applications
becuase the tasks I need to perform are most easily suited to my use
of these applications which happen to be console based.
There's ultimately a chicken and egg problem here. You don't know
what tools you need until there's a job to be done. Hopefully, your
OS comes with a standard set of useful tools. Unfortunately, many of
the fanboy fad Linux distros lack almost all useful tools by
default. I'm sorry if you're running one of them.
Learn the tools you need to use, and learn them well. This follows no
matter what hat you are wearing: student, full time employee, open
source developer, etc. The better you understand and can use the tools
you need to interact with, the more efficiently you are able to work.
Working more efficiently lets you work faster and smarter. This has
great benefits to yourself, your peers and your employer(s).
