deskthority

My idea for a new keyboard is online: http://www.quirky.com/ideations/339699

In short: It combines the modifiability/changeability of touchscreen keyboards with good user experience of physical keyboards.

You can vote it if you like it...

Who would this be marketed for?

E Ink Optimus Maximus? Hopefully with better switches. There's a catch: for colemak/dvorak, an IBM/Unicomp keyboard with curved plate and identical (swappable) shaped keycaps does the same thing.

If we ignore the slightly wider keys, the other snag is that the Optimus Maximus keyboard uses ML switches: transparent keycaps slide down around the static OLED panels. How do you feel about Cherry ML? Do you think there's any way to put screens over proper full-travel switches such as Matias Quiet Switches or Cherry MX? I guess it would just mean very tall keys. Your page does not have any sensible diagrams of complete keyswitch construction.

Also, your layout is NOT ISO or JIS compatible: your big-ass enter key loses a 1 unit position consumed by the ANSI portion of enter.

I'm kind of torn to be honest. The downside to changeable keys is that you have to keep looking at the keyboard for everything. If I want the Photoshop eyedropper, I don't want to have to keep looking at the keyboard for a picture of the eyedropper. The tool palette already tells me that I just press "I", and I can do that without looking down for the the I key, so I gain little by paying $5000 for a keyboard that replicates Photoshop's tool palette.

Also, unless you pull of colour displays, you won't come close to the beauty of Cherry doubleshot English/Russian keycaps, and a set of those wouldn't set anyone back five grand either :-P

The idea is cool in itself (Art Lebedev's, not yours), but I think it's confined to the realms of Star Trek technology for the time being, although Star Trek went all touch panel anyway. If you just want a touchable Photoshop palette/Unicode entry panel/gaming button grid, your best bet by far is a small (maybe 5") OLED touchscreen on a tilt stand that you can place next to the keyboard. Higher up position for ergonomics, trivial construction, not tied to key switch size, doesn't need to replicate perfectly good keyboard technology, etc. You could even implement it as an app for the new iPad Mini, and just make a tilt/height stand with dedicated Bluetooth transceiver and USB cable for your computer that presents a HID keyboard (for numpad mode/Photoshop control etc) plus maybe a custom device if you want to do anything more clever. And you can take the iPad away from your desk when you want to use it in front of the TV.

Your page does not have any sensible diagrams of complete keyswitch construction.

That's true. I'm not engineer nor graphical designer and I don't know yet what the final product will look like. The current images are just rough suggestions.

The downside to changeable keys is that you have to keep looking at the keyboard for everything. If I want the Photoshop eyedropper, I don't want to have to keep looking at the keyboard for a picture of the eyedropper. The tool palette already tells me that I just press "I", and I can do that without looking down for the the I key, so I gain little by paying $5000 for a keyboard that replicates Photoshop's tool palette.

Why do you have to keep looking at the keyboard then?

How do you learn Dvorak or Colemak without first seeing the layout? It's possible, but difficult.

Also, people have different ways to learn things. If you learn without ever watching the keyboard, that's fine, but not everyone does, I'm afraid.

What keyboard costs $5000? Optimus Maximus? The price of Optimus Maximus doesn't imply the price for other similar keyboards.

Also, unless you pull of colour displays, you won't come close to the beauty of Cherry doubleshot English/Russian keycaps, and a set of those wouldn't set anyone back five grand either

Colour is not a problem. We can have colour E Ink (over 4000 colours, I think) that needs only a small fraction of the energy your state-of-the-art LED/LCD displays need...

You could even implement it as an app for the new iPad Mini, and just make a tilt/height stand with dedicated Bluetooth transceiver and USB cable for your computer that presents a HID keyboard (for numpad mode/Photoshop control etc) plus maybe a custom device if you want to do anything more clever. And you can take the iPad away from your desk when you want to use it in front of the TV.

That's cool, but that is not the vision I have in my mind...

Daniel Beardsmore wrote:[…] The idea is cool in itself (Art Lebedev's, not yours), but […]

As a matter of fact the idea is much, much older and such keyboards have in fact been built as early as in the 80s:
http://deskthority.net/marketplace-f11/ ... -t915.html

Apart from that, I agree with your analysis. In the end it would mean having a diferent keyboard layout for each application you run.

rodtang wrote:Who would this be marketed for?

People who need different symbols, but can't find them easily.

People who want to learn Dvorak or Colemak easily.

People who need Unicode symbols (Unicode passwords, Unicode URLs).

Just about anyone.

Henri Heinonen wrote:Why do you have to keep looking at the keyboard then? ;)

Probably something to do with it having displays inside each key. If I wasn't looking at the keys all the time, having displays inside them would be somewhat redundant.

For Unicode entry, I just bind all the characters I use commonly to keyboard shortcuts, e.g. ctrl-alt-N for "→". Character Map is mostly OK for when I need something obscure.

webwit wrote:E Ink Optimus Maximus? Hopefully with better switches. There's a catch: for colemak/dvorak, an IBM/Unicomp keyboard with curved plate and identical (swappable) shaped keycaps does the same thing.

Which is exactly what I'm doing with my model F.


Seriously, while I don't want to crush the thread starter's spirit, this idea isn't new nor is it really innovative. The keyboard layout is still fixed, so people who want a Kinesis or flat datahand layout still need to have a custom-built keyboard. People who touch-type don't look at it for most of the time and don't need E ink. It needs software to change the symbols per application. If the keyboard has only one layout, the IBM keycaps or even old-school two-piece clear keycaps work perfectly fine.


Touch-screens don't have these problems; they can change the physical layout, update symbols on the fly, possible detect different pressures and maybe even detect when your hand drifts off track (read about FingerWorks' TouchStream keyboard and do some research about new input prototypes on tables). Their main problem though is price, fragility and the fact they're flat as a pancake. The trick lies in inventing the technology to get tactile feedback to locate the keys, and more feedback to know when a key is pressed and stays pressed. Apple bought out FingerWorks likely to keep their patents under lock and key, but research is still ongoing and that's what we need to look at. The keyboard presented here needs to be out now, or it might be dead on arrival.

JBert wrote: Seriously, while I don't want to crush the thread starter's spirit, this idea isn't new nor is it really innovative. The keyboard layout is still fixed, so people who want a Kinesis or flat datahand layout still need to have a custom-built keyboard.

It's pretty much impossible to design a keyboard that satisfies the needs of all the 7 billion humans on the planet. There will always be people who say things like "No! This keyboard is too ugly!", "I don't like it. It is not ergonomic enough!", "This keyboard has too many keys!", "This keyboard doesn't have enough keys!", or "I can invent a better keyboard in two minutes!"

I have had those ideas about ergonomy, but I didn't want to make this Universal keyboard "too weird". Ergonomic keyboards tend to be too weird and most of people seem to like the regular flat keyboard design.

Let's start simple and make things better step by step...

Hm. I think it's a nice idea but I wonder if it will be possible to actually make it without cutting down quality or making it super expensive.

On a more technical note. Unicode input would be cool but currently keyboards don't work that way, right?. Some drivers that accept full unicode from the keyboard would be ridiculously awesome for all our custom controller projects.

There is already cheaper tech than the one in Optimus Maximus: A single LCD screen with a low-profile scissor switch/rubber dome keyboard on top, with transparent chiclet keys. The point is that the scissors are on the outsides of the transparent keys instead of inside them.
Razer has a Star Wars-branded keyboard where this tech is used for only eight keys. Microsoft had a prototype of a compact keyboard where it was on all keys. MS loaned(?) it out to college students to make demo apps for it.
How about building a scissor switch keyboard that has leaf springs instead of rubber domes?

You can not get unicode everywhere just with a new type of keyboard. The common keyboard protocols on USB and PS/2 don't talk characters: they talk raw scancodes. Unicode entry is an issue for host software and which keymap file that you use.
Under MS Windows, you can already enter any unicode position by holding (left) Alt and entering the code on the numeric keypad, but not all programs support it and not all fonts have the characters that you want to type (Windows has had this for longer than they have had Unicode).
Linux does not have this, as far as I know, but the layer under Alt Gr is usually richer than under Windows in the standard keymaps.

Henri Heinonen wrote: What keyboard costs $5000? Optimus Maximus? The price of Optimus Maximus doesn't imply the price for other similar keyboards.

And there is a new version available for pre-order, $1086, Optimus Popularis http://store.artlebedev.com/electronics ... popularis/

It has the same problem as touch screens that you cover the displays with your hands. But if it was e-ink and cheap It would be cool but just as a smal feature not a usp.