A few months ago, I finally managed to purchase a keyboard like the one I described in a previous post
The keyboard model is called Iris and it can be purchased from keeb.io
While I love using it, most people who have seen it think of it as a practically insignificant hobby. I differ, so this post is my attempt to explain why it can be useful to pretty much everyone, even those twho are already well adjusted to their present keyboards.
I have been told that my previous posts were too long and the few who know about my blog lost interest mid-way through the articles. While I can’t drastically decrease the size of the post because of the nature of the subject, I have included lots of diagrams to explain my points. I hope this makes it more readable.
So let’s consider this keyboard:
It’s the most common configuration in compact laptops today. The key colours explain the ‘official’ technique of hitting the keys:
|dark green||stretching the index|
|yellow||pinky and stretching the pinky|
|grey keys||moving the hand out of ‘home’ position|
|white keys||too far away to hit comfortably|
Let’s notice a few reasons why this configuration is sub-optimal:
- The weakest finger (pinky) hits the most keys. Even worse, it is used to hold down shift and control (using more force than usual) while other fingers hit alphabet keys as part of a key combo. This is not just a theoretical problem - pinky injuries are common amongst authors and coders.
- The layout is asymmetric - while the right hand has a natural angle of approach, the left hand needs to be contorted to if one wants to use it in the official typists’ technique. One can chose to ignore the official technique (I use my left index to hit ‘c’), but there certainly are benefits to the official technique.
- The usage pattern is also asymmetric. While the left stays more or less in the same neighbourhood (albiet contorted), the right hand moves a lot - to hit backspace and delete, to access cursor keys, to hit enter, and for the mouse.
- Some of the most often used keys are the least accessible. Ctrl, Backspace, Delete are located at the extremities of the board, and Enter is relatively inaccessible as well.
- Some rarely used keys such as Caps Lock and semi colon are very accessible.
- The strongest finger, the thumb, is used only to hit a couple of keys (albeit, often used ones).
- Inconsistent key sizes - the giant space-bar takes up a lot of easily accessible space. Most people hit the space-bar in the same place (Notice the wear out pattern on old space-bars and you’ll know what I mean). So a lot of easy access space is unused here. Similarly other large keys such as Shift, Caps and Enter use more space than needed.
I am now going to show, step by step, how we can get rid of all these issues while bringing in new features that will improve the keyboard.
Step 1: Symmetry, key size consistency, breaking up the space-bar
- We reduce all keys except thumb and cursor keys to the same width.
- We use the increased space is used to move the columns apart and make the board more symmetric. The left hand can now approach the board at a natural angle.
- The space-bar is broken into 4 keys, 3 of which are not yet assigned functions.
Step 2: Bring inaccessible keys closer
- Get rid of the function keys and access them through a thumb activated function layer instead. So, Func+a = F1 and Func+s = F2 and so on…
- Move the numbers from their normal positions to another thumb activated layer. So, Num+m = 1, Num+, = 2, Num+. = 3, Num+j = 4 and so on… We create a number-pad on the right half of the board. This actually enables faster number entry, and frees up the dedicated number keys. The symbols that are accessed using the number keys will be dealt with later.
- Move Esc, Backspace, and Delete to the erstwhile number row, where they can be accessed with lesser stretch and by stronger fingers.
Step 3: Reduce pinky usage.
- Move Ctrl and Shift from the left side, and Shift from the right side to the erstwhile number row. So they are now accessed by the stronger ring and middle fingers than the pinkies.
- We move Enter to the semi-colon location, where it can be hit very comfortably by the pinky in its home position.
- We remove all pinky keys that need more than a key width of stretch (except for the top row - we’ll use these sparingly). This removes the remaining symbol keys. We access them instead using a layer activated by a ‘Sym’ key. We leave it next to the Enter key where it can be accessed easily. We note that this needs the pinky (a weak finger) to hold down a key - an undesirable feature. We might move it to a different location later.
Step 4: Move the cursor (arrow) keys closer
- We get rid of the arrow keys and move them to a thumb activated navigation layer instead. So Nav+j = left, Nav+k = down, Nav+l = right, Nav+i = up. Note that this resembles the arrow cluster closely, and actually gives us full height keys instead of half height ones.
- The layers are getting complicated now, but the concept is clear. We have 4 layers - Navigation, Number, Symbol and Function. We will describe the layers elsewhere.
- We move the right control key to the erstwhile number row. In its original position, it is was very convenient to use with the arrow cluster. Since the arrow cluster has moved, there is no need to leave it there any longer.
Step 5: Further pinky optimization and ergonomics
- We move the Sym layer key from the undesirable pinky location to the now empty thumb key (the right ctrl used to be here)
- We move the rarely used Caps Lock key to the top row.
- We move the ‘halves apart as far as they can go. This allows for a user’s shoulders and hands to be used at a more natural angle (more ergonomic). We get rid of the Caps Lock key and the old Sym key to allow us to move the keyboard halves a further key width apart. The qwerty row pinky-stretch keys are reduced to half keys, or may be removed completely.
Step 6: Optimising thumb key placement and utilisation of empty space
The Sym key and Func key positions are exchanged since symbols are used more often than the function keys.
The empty space in between may be used for a touch-pad and extra buttons, or for a much bigger battery. The touchpad in the above picture is 4 keys wide. For reference, the touchpad on my Dell XPS 13 is 5.5 keys wide.
The biggest apparent benefits are reduced pinky usage and better utilisation of the easily accessible keys.
The most apparent disadvantage is a steeper learning curve. But I would argue that this is nothing that better key labels and smart layer configuration can’t solve. I’ll tackle both in another post. I’ll also explain how users can use their existing muscle memory or minimise learning new ones.
Another criticism could be that common keys like the arrow keys that needed one key press now need two. While this is true, I’d argue that accessing layers through (opposable) thumb keys makes this much easier than moving hands around and then losing the finger position on the home keys.
A third criticism is that certain combinations such as ctrl+left are not possible. Again, this is something that I tackle in my post on layers. I hope to demonstrate that this keyboard configuration makes such combinations much easier.
There are further paradigms we haven’t explored here - exact key arrangement in rows and columns; and the arrangement of the alphabets. I’ll get into that in subsequent posts.