http://www.microsoft.com/globaldev/tools/msklc.mspx

Thank you for the link! I do not remember seeing that before, but it was mentioned on a blog post (with several other keyboards, so there is a possibility that I missed it) that was in my browsing history. Google Desktop is interesting to have around, speaking of which.

Anyway, the keyboard does seem to have many attributes that I would like. However, I cannot see paying over one hundred dollars for a keyboard first of all. Secondly, I do not see Linux support. If there is no Linux support, then it’s pointless for me to try. Most important though is how it types. A review on Engadget I came across said that the typing experience on it was less than pleasurable. I guess I will stick with my old Packard Bell PS/2 keyboard. It types great, it’s durable, and it works across the board assuming I have my PS/2 to USB adapter with me.

It’s a nice toy, could probably make for a very interesting layout learning tool, with some specialized software. Though ultimately you’d be touch-typing anyway. Any blank keyboard should do.

As Roland pointed out above, a programmer isn’t always going to be programming, unless they are Richard Stallman. Thus, this would allow someone to have one layout for optimization in programming, and another for optimization in typing. The same principle would be even more useful when switching between languages. Then we could see layouts that no longer have to give English options. Greek layouts, for example, could be focused only on Greek characters. They would no longer have the need to be able to type in Latin characters too.

Oh well. Even if the technology were to exist by some odd stroke of luck, there’s no way that it will be widespread in the foreseeable future.

Fact is I use QWERTY at work, and have a Kinesis Contour at home that I regularly switch between Dvorak and QWERTY (for games). I don’t have the same speed that I used to have when I was a QWERTY or Dvorak only; but it’s only a matter of flipping the mental switch. I’ve been doing this since mid-98.

I would suggest if you were going to go this route to stick with one for a period of time, then after the new layout becomes ingrained, allow yourself to switch between the two. I found that having a keyboard with a very different physical feel (the Contour) made it easy to flip the QWERTY/Dvorak switch. Nowadays, I just have to think “….Qwerty.” or “….Dvorak.”

I ran a test as I was typing this to see how hard it is to flip the switch–not very–interestingly, there’s an element of feedback, where-in I see that I’m typing in the wrong mode and quickly, unconciously flip to the proper mode.

Specifically in regards to how either layout stacks up for programming: they both suck, just in different regards–IMO.

You are indeed right that some keys — most notably the semicolon, but also keys like ‘f’ — has quite a different usage profile in English than in programming languages, and that a keyboard optimized solely for a programming language would look quite different.

But the converse is also true, and I don’t think you would be very productive on a keyboard optimized solely towards for instance C or Java, unless programming is the *only* thing that you do, and there is no documentation, notes, correspondence and blog entries(!).

Consequently, I think you should consider the layouts in sum. Yes, the semicolon is placed at a more awkward place. But perhaps there are other improvements that make more than up for it?

My approach has been to start out with English (using the existing Dvorak) giving it first priority and then adjust all the other symbols to better match programming (by the same principles) second.

But there is, as you mentioned, other layouts that does otherwise. I believe that the Colemak layout[1] retains most of the symbols as they were (if you like that), and that the Arensito layout[2] claims to have taken programming more into consideration. I will leave it to others to comment on those.

[1] http://colemak.com
[2] http://www.pvv.org/~hakonhal/keyboard/

A keyboard has a lot of letters and numbers and symbols that need to be arranged, but not every key on a keyboard is equally accessible — this is where alternative layouts come in. Placing commonly used keys in easily accessible locations make the overall layout easier to use, but now one must decide which keys are more “common”. Semicolon, a prime example, is rarely used in English text, so an English optimized layout would place this key further away. On the flip side, it is much more common in certain programming languages, so this set would benefit from having it closer in. This creates a conflict between two sets, so an optimization towards one must come at an expense of the other.

So different keyboard layouts settle at various points of this balancing act.

I apologize if I led anyone to believe that my comments were meant to be an independent opinion; that was not my intention. To my defense I have to note that the link on my name goes to the homepage of the keyboard layout, and it has my name in it. However, I should have stated it explicitly from the start instead of assuming it would be known. I am most certainly biased!

Rather, my comments were meant to elaborate on the thoughts behind some of the trade-offs that have gone into the design. As the original review found them strange, and a lot of people initially do, I wanted to shed some light on their rationale and the alternatives.

In a way, the entire layout is a review of the Dvorak keyboard layout in conjunction with programming, because I created it out of the desire to do just that and the frustration of having no good alternatives available. For years I was the only user, and I would still use it even if no one else did. I do not make any money out of it.

My reasons for commenting on it is that I believe that anyone that consider converting to Dvorak (and I think there are mostly converts; I haven’t yet met any of those trained to Dvorak as their first layout), should hear from those who have gone down that path before them, because of the high cost of doing a thorough evaluation.

But then again, I am biased.

Not that I mind hearing some criticism from people that have taken a look at (Programmer) Dvorak and turned it down. In fact, I try to ask everyone I know that use it if they feel that anything could be improved, to see if I have missed out (or perhaps overstated the importance of) something.

In particular, I would like to hear the reasons for why “any optimization in favour of programming languages would now come at the expense of English”, from those who have that experience.

I think my time is up…