I found this thread a bit late, but will add my voice. I've been using a Kinesis Essential daily for the past two years. These suggestions are roughly in order of importance to me.
1. Open source firmware (and even open source hardware)
- I see this as the most valuable feature. It should cut your development costs and since many of your customers are developers, I imagine it would be a selling feature too.
2. Regular Cherry function keys
3. Function layer shift key in the main keyset, not just on the foot switch
4. Trackpoint/trackball option
- I can see many users wanting only one or the other, or even wanting none at all. Trackpad is less critical as it is pretty easy to put an Apple Magic Trackpad or other off the shelf trackpad on top of the keyboard.
5. Detachable USB cable
- having connections for both mini and micro USB would be a bonus for those that already have tons of spare cables scattered around, but never seem to have one of each
6. Bluetooth option
- for acceptable battery life, it would require efficient firmware that puts the keyboard to sleep often, but also wakes up instantly. As an example, I have two wireless mice, both Logitech, and the older one has a lifespan of about a week, while the newer one has a life span of about two years. Some of that difference is probably the modern hardware, but I suspect that a lot of it is the firmware.
7. Blank keycap and/or Dvorak only option
8. Regular profile keycaps only
- if I recall correctly, the only odd caps are the tall thumb keys, the 1.25 keys, and the '3' and '8' keys. The 1.25 keys and the out of place number keys can probably be switched to regular caps by adjusting the tilt of the actual keyswitch (though I haven't spent anything on R&D to confirm this). The tall thumb keys might be replaceable by staggering a separate pcb into the thumb cluster. A little hard to explain in text, but instead of having the single PCB in each thumb cluster that is the shape of a 3x3 grid with the corner taken out, you could make a PCB the size of a 2x3 grid with a bit of an extra lip at the top edge, then you can make the taller keys on a 1x2 grid PCB with a bit of a lip at the bottom edge. Rest the smaller board on the larger one and solder some pins through, and you should be able to use the same profile keycap as the home and page up keys. I imagine this would add to your manufacturing costs, but perhaps it would save you some since the keycaps would be regular profiles.
9. Reduce the height of the keyboard
- I love the well shape (the lowest row is the only one that is a tad awkward to reach, but it is much less awkward than it would be if I tried to press the same keys on a traditional keyboard, so I'm happy) and the wrist rest area is pretty necessary to accommodate those raised keys, but I suspect that you could shave a bit off the bottom of the case under the key wells, even just by using lower profile rubber bumpers. I find that I can never find a chair that is tall enough to sit comfortably with my Kinesis on top of the desk, and the Kinesis doesn't fit in most keyboard trays either.
10. Increase the tenting angle slightly
- this might go against the height reduction in the previous point, but I can see it being more comfortable
11. The current Kinesis models are great for modding. There is lots of space inside and the key remapping makes things easy to customize. Keep this up. Additionally, I know that the current models have some unused keys in the matrix. If you could break out some pins to these blank parts of the matrix, perhaps with non-populated .1" headers, this would make life for the modding community much easier. Adding extra keys like the palm keys many users are adding would then be trivial, but could be left to the end user so they aren't forced on anyone that doesn't want them. You could even use this to your advantage in the future by offering different upgrade packages such as a new top shell with pre-installed palm switches.
12. Split option
- I would definitely consider chair mounting a split version. I liked the one suggestion of having each half be a separate keyboard, but I know that some operating systems can't handle dual keyboards very well (if I recall correctly, if you hold shift on one keyboard on a Mac, the second keyboard's keypresses are not shifted). This could also limit the possible features in an open source firmware such as an inverted number/symbol row since the press of the shift key would need to be known at the keyboard level, not just at the operating system level.
12. Headphone port
- perhaps not all that useful, but one feature I've wished for on occasion
13. Reduced price
- I'm pretty sure that the difference between the current Pro and the regular models in an additional $1 memory chip, a $0.50 toggle switch, and a different colour top case (and perhaps also a slightly different firmware). I'm not sure if the Pro significantly subsidizes the price of the regular model at this time, but I know that both models are prohibitively expensive for myself and others. I lucked out and got a like-new Kinesis Essential (a roughly 10 year old board, I think) for about 1/3 retail of a new board, otherwise I probably couldn't have bought it. I understand how retail works and that you guys need to make a profit in order to keep making these things, which we all want, but bringing the price for a basic model down from about $300 to $200 would make it immensely more accessible.
I've had lots of ideas (these and other), but really the fantastically unique and comfortable keyboard is what inspired all those ideas. If the next model were nearly identical to this one, I would still be very happy with it. Thanks for an awesome board, and I look forward to the new one.