As narrow as possible, but not as stripped as possible

[Mal-2]

I've been working on my own keyboard layout, with certain design goals in mind (also available here):

• It should be no wider than a 60% keyboard, to minimize the distance the mouse gets pushed out -- no matter which hand I mouse with (because I can and occasionally do switch)
• It must have dedicated function keys, because while I care immensely about the width, I don't care nearly so much about the depth. This rules out practically all 60% keyboards.
• The stagger is standardized at 1/2 key per row, none of the usual half-quarter-half or even less uniform staggering. This increases comfort when using the overlay modes.
• No tiny keys (smaller than 1 wide or tall), for cost and simplicity -- aside from the toggle switches in the upper right which are not really keys.
• Status LEDs in the relevant keys themselves. Dedicated space for status indicators is a waste.
• Periodic breaks in groups of keys I have to reach for, to facilitate identifying them by touch alone. That means retaining the standard groove (though narrowed) between clusters of four F-keys.
• Both QWERTY and Dvorak legends on keys, except where they are unchanged. (A and M are treated as moving keys even though they don't actually move, because they are functionally similar to keys that do, but the number keys are treated as non-moving.)
• All overlay modes should be an improvement over dedicated keys. To use the cursor keys, there is a momentary thumb tuck to enable the overlay, then the rest of the left hand remains in its usual position (and yes there are two down-arrow keys in the overlay, one for normal WASD-style and the other for 8-direction retro gaming). To use the number pad, there is a thumb tuck and the right hand moves up one row. I would like to have done it without any shift of position, but that leaves no place to put the 0 and . keys.
• No keys omitted, no matter how little I think I might need them.
• Minimal learning curve for someone else to use the keyboard. This means that other than evening out the stagger, most typing keys (and modifiers) don't move. The spacebar gets tiny to make room for the Curs and Num keys, but should still be quite reachable when the left hand is in WASD position for gaming.
• Hardware Dvorak. I don't want to change the OS configuration of someone else's machine just to use my greatly preferred (Dvorak) layout.
• All function toggles are on top, not DIP switches hidden underneath. This is what the four white buttons at upper right are: locking push-toggle switches. The upper two toggle whether or not the Curs and Num keys are allowed to lock, or are reduced to only-when-pressed modifiers. The lower two are the QWERTY/Dvorak switch and the Win-key disable switch. I don't really need a Win-key disable switch, but there is room for four switches and other people do sometimes demand this.

In normal (toggle switch not engaged) mode, the Curs and Num keys will act both as toggles and as modifiers, depending on how long they are held. If it's a tap (and I'll have to experiment a bit to find the optimal threshold for determining the difference but I'm guessing it's around 600 ms), then the lock state toggles. If held down longer than that, they will override the current lock state regardless of what it is, down being "use overlay" and released being "don't use overlay". Thus I can "clear" both overlay settings back to default by holding them down for a second or so, no matter what the current status is. I don't have to look, and I don't have to care.

The Dvorak legends would be in a different color (probably green), but I have no way of getting that out of the Keyboard Layout Editor.

I needed a seventh row to squeeze in dedicated (if somewhat inaccessible) PrtSc, ScrLk and Pause keys and my toggles, so the eight Macro keys are just the best use of the remaining space I could think of.

I've also decided that if I'm going to go to the trouble of designing and building everything from the ground up, there's no sense in cutting corners on keys. Thus I've planned to use PBT and dye sublimation for all keys, and the color choices reflect this.

Now for the big question: is there anyone who commercially executes build designs like this? If not, I know I have the skills to assemble something like this, but I'm not so sure about my ability to program it.

Also, would anyone else be interested in such a design, or am I a rare bird who isn't pleased with F-keyless 60% keyboards but still wants minimum width?

[Mal-2]

Oops, I meant the stagger is fixed at 1/4 key per row, not 1/2.
I still want somestagger, because (1) we're used to it, and (2) a right-handed mouser still has the keyboard slightly to the left of center, making such a stagger feel natural.

[Muirium]

Looks like a one of a kind board to me. Sourcing caps will be a real challenge. You'll probably need to get WASD keyboards to make them for you. (Great customisability, mediocre quality. I don't know ther printing limitations because their stuff doesn't interest me.) You'd have to build this yourself, and modify Hasu's TMK keyboard firmware to support the various modes.

All advanced stuff really, you've leapt in the deep end! Have a look in the Workshop > Threads of Interest to see how construction is done.

[7bit]

If you can live with row 3 modifiers, most keys should be no problem.

[bhtooefr]

And WASD is lasered ABS, IIRC.

That said, I'm not seeing anything other than maybe the spacebar that wouldn't be available in DSA profile, and not insisting on doubleshots helps a lot here... ignoring that cluster in the upper right. You wouldn't necessarily need the window for the LED, you could just rely on backlight spill effect to get it lit.

Also, given this forum's Eurocentricity, might want to add .25 U to the right side to better support ISO layouts without going below 1 U on enter. That said, the layout I want to do is 14.5 U wide at standard stagger (a standard 60% is 15 U), and I'd just go for ANSI enter for everything.

[Muirium]

But what about the legends? Does SP do custom printing for single orders? Far as I know, that's WASD's forte.

[bhtooefr]

For single orders, you'd have to go for WASD, but I thought dye sub MOQs were low? (Actually, I thought Unicomp tried individual order dye sub, but found that the method they were using was too low quality...)

[andrewjoy]

that is cool

needs a big ass spacebar tho

like this

http://deskthority.net/resources/dream-layout/15319

seriously i want that layout so so much

[Mal-2]

And WASD is lasered ABS, IIRC.

That said, I'm not seeing anything other than maybe the spacebar that wouldn't be available in DSA profile, and not insisting on doubleshots helps a lot here... ignoring that cluster in the upper right. You wouldn't necessarily need the window for the LED, you could just rely on backlight spill effect to get it lit.

Also, given this forum's Eurocentricity, might want to add .25 U to the right side to better support ISO layouts without going below 1 U on enter. That said, the layout I want to do is 14.5 U wide at standard stagger (a standard 60% is 15 U), and I'd just go for ANSI enter for everything.

The cluster in the upper right would probably be completely non-standard (as far as keyboards go) switches. They'd be more like the power button on old-school hardware that is actually disconnected from the mains (that is to say, actually off) when you switch it off, only there would be four of them. Also, if this proves impractical, they may just get replaced with ordinary toggle switches, either sliders or nice steampunk levels. Those would be relatively easy to mount to the case.

As for width, this one clocks in at a whopping 14.25 U. I was serious about "as narrow as possible". Admittedly the lack of provisioning for an ANSI enter key might be a problem, but frankly I wasn't thinking about non-U.S. layouts. I can give this more thought, though it would break some of the feng shui of the current layout. If I am willing to go to 3/4 width for the ~` and |\ key, i can trim this still further to 13.75. Both of the top two rows have room to be compressed by 0.5U and still have channels everywhere I think they should be, and the macro keys could easily be done in groups of four rather than two. The Ctrl keys would go back to being 1.25 U instead of 1.5 U, which would probably bother nobody, and the CapsLock would be reduced by the same amount, meaning the two could still be swapped.

That's another thought... perhaps the fourth switch should swap Ctrl and CapsLock rather than locking the Win keys out. This would require one extra LED on the board, but the power to them could be routed through that toggle as well (they're practically all double-pole) so only the properly located LED would light up.

It looks like what I'm going to do for the moment is chop up a couple existing rubber dome keyboards and experiment. As my proof-of-concept, I've been running with my mouse pad on top of the cursor pad and keypad. Obviously this is stupidly inconvenient and has my mouse sliding downhill, but it's enough to prove that yes, even the width of the cursor section needs to go. Even the keys down the right side of most compact keyboards needs to not be there -- not just for width, but because this then makes the narrowness of Backspace and Enter an actual problem. It's much less of an issue if there is nothing to the right of them. You can't hit the wrong key if it's not there, and you'll clearly know if you completely whiff on the key.

For single orders, you'd have to go for WASD, but I thought dye sub MOQs were low? (Actually, I thought Unicomp tried individual order dye sub, but found that the method they were using was too low quality...)

Ah well, then a prototype might well have to be ABS and laser. That's not a huge issue, because if it takes off at all, that should be sufficient to cover the minimum necessary to order dye sub PBT.

that is cool

needs a big ass spacebar tho

like this

http://deskthority.net/resources/dream-layout/15319

That's the one thing you clearly aren't going to get out of this keyboard. I think that spacebar may exceed 14.25U all by itself.

[Findecanor]

Sourcing caps will be a real challenge. You'll probably need to get WASD keyboards to make them for you.

WASD Keyboards does not have some of the sizes for the rows you want.

Maybe some set in Signature Plastics DSA profile has some of the keys in the right sizes.
The right Shift is 2.5 wide, and 2.5 wide key is not available in DSA profile, and otherwise quite unusual.

Try also to find out if you really would be comfortable with the Space bar so close to centre. I know I usually press mine in several places under N and M on a regular staggered keyboard.

[Mal-2]

Maybe some set in Signature Plastics DSA profile has some of the keys in the right sizes.
The right Shift is 2.5 wide, and 2.5 wide key is not available in DSA profile, and otherwise quite unusual.

Try also to find out if you really would be comfortable with the Space bar so close to centre. I know I usually press mine in several places under N and M on a regular staggered keyboard.

Looks like I have to do some research into what sizes are commonly available, though the spacebar will obviously be a custom job, or at the very least come from a Japanese-purposed keyboard.

When I type, my thumbs actually touch in the middle quite frequently. Striking the spacebar outside the area under (what are for you) V, B, and N would be quite unusual. (For me they're K, X, and B.)

Here, you'll get a kick out of what I'm currently using as a test setup to see if this is worth pursuing.


The one on the left is exactly what you think it is, a chopped Logitech Deluxe 250. Its circuit sheets are folded and taped behind what remains of the shell. The rubber dome HP keyboard has had all of the cursor and number pad column keys removed (and stored in a bag) so I can throw my mouse pad on top of it. The tilt is no longer severe enough to cause the mouse to slide downhill, and I don't have to worry about pressing down on the mouse pad.

It's a bit sad, as the Logitech was actually the better constructed keyboard, but they made the brain-dead decision to rotate all the home-bumped button shafts by 90 degrees which makes it impossible to freely move the keys around for Dvorak. I had to label four of them. (My Acer Aspire 5253 has a similar problem, only they rotated 180 degrees so I just have four keys mounted upside down.) The HP lets me move any 1U key anywhere I want, and it's native PS/2. It is not NKRO, but I've never had a problem with it.

The keypad doesn't actually have A and B or Esc keys. A and + are both really +, and B and Esc are both really Enter. It's just that the case provides two stems AND a stabilizer for these keys, so they always felt a bit sluggish. Separating them into two (functionally identical) keys is a workaround, and I had plenty of leftover keys. The same goes for the double 0 key, though at least in its case they both say 0 (or O).

[jacobolus]

The stagger is standardized at [1/4] key per row, none of the usual half-quarter-half or even less uniform staggering. This increases comfort when using the overlay modes.

Have you tried this? I have a couple keyboards with such layout, and it’s a huge pain to type on them. It’s just about as annoying IMO as learning a matrix layout, column-stagger layout, or symmetrical row stagger, but without the associated benefits.

[jacobolus]

I strongly advise you don’t make the numpad and other layers be modal with toggles, but put them on shifted layers instead. At that point, you won’t want to use the same-hand thumb as the layer shifter, since you need the thumb to use the bottom row of the numpad.

My personal recommendation is put the numpad shift on the left hand, and then center the numpad on the right hand so that 4-5-6 is on the home row J-K-L keys, with the spacebar for 0.

[kbdfr]

I wonder what you're willing to change on your keyboard just to have your mouse in a better position when there is a much simpler solution: get rid of your mouse and use a RollerMouse, i.e a pointing device placed in front of your keyboard and consisting of a bar which you can move sideways and rotate with your thumbs without changing the position of your hands.

setup03.jpg (56.23 KiB) Viewed 5607 times

[CeeSA]

you still have to leave your home row. Trackpoint is the better solution for me at working place

For gaming with a mouse it is a real benefit to have keyboard that is narrow.

[bhtooefr]

My concern with the RollerMouse would be how it'd do for gaming, and as I understand, it's got a bit of a learning curve.

Not that TrackPoints don't have a learning curve, mind you... (And, I've actually gamed on a TrackPoint before. I'm not great at gaming, but I'm not really any worse on a TrackPoint than on a mouse.)

[kbdfr]

No idea about how a RollerMouse does when gaming*, I use it for work and it is just great.

No precisions issues whatsoever,
and the learning curve is for sure nothing compared to that of the keyboard layout in the OP




*no [:vomit:] smily here, I fear

[Mal-2]

I wonder what you're willing to change on your keyboard just to have your mouse in a better position when there is a much simpler solution: get rid of your mouse and use a RollerMouse, i.e a pointing device placed in front of your keyboard and consisting of a bar which you can move sideways and rotate with your thumbs without changing the position of your hands.

setup03.jpg

I've looked into the RollerMouse, without actually being able to see one (they're not common). As noted, I don't game that much, but when I do, it has to be on the same hardware as everything else. I might go for hours and hours of extended bouts of typing (in between thinking of what to write, or deleting what I just wrote that sucked) and editing said text, to gaming for half an hour to a couple hours, to working on music in CuBase and Encore, both of which are quite mouse-dependent.

I think the simplest answer might be an Evoluent Mouse-Friendly Keyboard to give me a keypad on the left, along with a separate USB keypad on the right. Then I remember that sometimes wrist and shoulder problems force me into using the mouse left-handed for various periods of time ranging from a day to several weeks. I generally will cease gaming at this point, because although I actually once had the fine motor skills in that hand, it is now too shaky. Were it not for that, just hanging the mouse out there would have solved everything, right? I was finding myself doing a white-knuckle grip on the mouse in order to stabilize it. If I have muscles already in tension, small random pulses don't really do much in practice. The problem with that of course is that it is extremely tiring.

Now I am finding that the problem is spreading, and the right hand is also getting twitchy. Yes, this was predicted. I need to get ahead of the curve and find ways to deal with that, and not overworking the shoulders with a mouse way off to the side is the first step.

Mind you, I'd be fine with multiple pointing devices. I am competent with a trackpad. I am competent with a TrackPoint. I wouldn't want to use either one for gaming, where I have become quite used to doing things like running while strafing, and compensating exactly enough to stay fixed on a target at the same time. This means having the mouse close remains a requirement no matter what my primary pointing device might be.

The only reason I even entertain the notion that what's best for me may in fact be beneficial to a broad class of people (thus caring about putting QWERTY legends on the key at all) is that I do actually do so many different things.

I do realize that the solution I've implemented so far is essentially the same as that offered by the Mad Catz S.T.R.I.K.E. 5, only without all the sharp angles and l33tness. That's why I was holding out hope that maybe something already exists.

Maybe what I should focus on is a combination cursor/numpad keyboard, with dedicated function keys, which could accompany any existing 60% board. Then I wouldn't even have to worry about multiple key legends, or layouts in general.

[IvanIvanovich]

Sure sometimes creating something can be fun... but in some cases it might just be better to compromise a little and take something 'off the shelf'. Have you seen Noppoo nano75? It has many of the features you want to implement in a very similar layout:

[Mal-2]

Sure sometimes creating something can be fun... but in some cases it might just be better to compromise a little and take something 'off the shelf'. Have you seen Noppoo nano75? It has many of the features you want to implement in a very similar layout:

I had, actually. The Choc Mini looks like a better fit, to be honest. Also high on the list is the Keycool 84.
I also think Noppoo's keycap legends are truly awful, but at least they're replaceable.

[Mal-2]

OK, I've decided maybe I should be attacking the other end of this problem. I can presumably find something off the shelf for the main body of the keyboard, like the Keycool 84. That would still leave me wanting a real, ambidextrous numpad. All the cheap (and most of the expensive) numpads fall short in one important way, there's ONE job for the thumb, and NONE for the left pinky if it's in the left hand. Also, many have double-size Enter and + keys where they could easily support two ordinary keys. Some split the + key into - and + to make room for a Backspace. This is a nice start, but why stop at four columns? In order to facilitate use by either hand, five columns makes more sense. A simple 5x5 matrix should be well within my ability to fabricate, and there won't be any weird-size keys.

But wait, what else is missing (without Fn layers) from a 75% keyboard? Oh yeah, PrtSc, Scroll Lock, and Pause. Not that I use the latter two much, but if I were to chop down a normal keyboard that doesn't do layers at all, they would go missing.

That leads to this design:


This is still well within my ability to build, and if 3/4 height keycaps for the top row prove impractical to source, I can expand the upper row to full height. It just seems like a waste otherwise.

Thoughts? This is accepting that maybe existing 75% boards are in fact good enough, and attacking the other side of the problem: getting numpad/cursor functions I like, that can be easily placed under either hand. I also could hack this together fairly simply on my own, using existing USB keyboard controllers for a prototype (meaning the orange keys would be hardcoded to something, rather than programmable as I would intend for a final product).

[jacobolus]

If you’re normally right handed but want to occasionally a numpad in your left hand, I recommend flipping the numbers (and everything else) right-to-left. It’s much easier to learn and makes switching between right and left handed numpads easy.

Since a numpad often doesn’t really need to be of minimal size, I personally think there’s little functional reason to keep a perfectly square grid arrangement either.

For instance, I’d make a right-handed numpad in a similar physical layout to the right half of a hand-shaped ergonomic keyboard, along the lines of:


But even if you want to keep the standard grid, there's little reason not to add extra columns on the right and left and put the thumb keys in more accessible locations, rather than adding unreachable extra rows at the top.

(Just food for thought.. make whatever kind of layout you want

[Mal-2]

Incidentally, I have discovered a totally unanticipated problem. I can no longer Alt-#### for special characters because the Alt on one keyboard does not modify the keypad on the other. I will have to install a switch just for Alt on the number pad that I already have cut down. Obviously this won't be a problem in my proposed design (it has an Alt key), but it's something that had never even occurred to me to check into. Could this be because my primary keyboard is PS/2 and the extra is USB? If they were both USB, maybe Alt would carry over?

If you’re normally right handed but want to occasionally a numpad in your left hand, I recommend flipping the numbers (and everything else) right-to-left. It’s much easier to learn and makes switching between right and left handed numpads easy.

Actually I've left-numpadded before, and find it much easier to not reverse the layout. This is partly because left is still left and right is still right, no matter which hand it's under, and I've got "4 = left" and "6 = right" hardwired in at this point. If I reversed the numbers but not the arrows, that would confuse me when I go to use the numbers. If I reversed both the numbers and the arrows, that would massively confuse me when I went to use the arrows.

Also, being a musician, I do (minimally) play piano. The keys don't go in opposite directions to accommodate hands there, both hands have to operate on the same keyboard (or type of keyboard, if there is more than one). Thus, I'm used to things not being reversed for the left hand.

Since a numpad often doesn’t really need to be of minimal size, I personally think there’s little functional reason to keep a perfectly square grid arrangement either.

This I agree with in principle, but which hand do I optimize for? I input data with the right hand mostly, but would use it as a gaming control pad under the left hand. In order to favor neither (and be more approachable if made for sale), a grid remains practical. What you're doing is a bit like Novak fan-frets on a guitar -- it has significant technical advantages, but costs a lot more and requires retraining.

But even if you want to keep the standard grid, there's little reason not to add extra columns on the right and left and put the thumb keys in more accessible locations, rather than adding unreachable extra rows at the top.

It does matter. The more rows between the mouse pad (or primary keyboard) and the location of the hand, the further from center that hand is going to be. Granted, it doesn't matter as much as on a single keyboard, because the pad can be easily rotated to a comfortable position, but it still involves reaching.

[jacobolus]

Actually I've left-numpadded before, and find it much easier to not reverse the layout. This is partly because left is still left and right is still right, no matter which hand it's under, and I've got "4 = left" and "6 = right" hardwired in at this point.

Have you ever tried it with a reversed layout? It sounds like it would be confusing, but I promise you, it’s really really easy if you’re used to using your right hand. Basically, if you reverse the layout, you just pretend it’s your right hand doing the motion, and your left hand already knows what to do. If you can avoid consciously thinking about it, it “just works”.

As for arrows: well, the arrows on a standard numpad are in quite a terrible arrangement. I definitely wouldn’t recommend sticking with those, regardless of what layout you use.

Playing the piano is substantially different, because what your learn on the piano is how to play particular sequences of notes, with your hand moving all around, rather than just learning to use one specific finger for each note. But if you want to get more philosophical, a piano is actually a very suboptimal arrangement if you want to input arbitrary sequences of notes: but that’s okay, because the piano music we have is written assuming the shape of the piano and the typical limitations of the human body. Different key arrangements could definitely be invented that would make the piano a more capable instrument for making arbitrary music (but might not be as optimized for the canon of existing piano music). With a numpad, the problem is pretty constrained: type some arbitrary sequence of numbers. It’s not like there’s a large corpus of famous number sequences that need to be typed with particular technique using the left hand on a standard layout (right handed) numpad.

[Mal-2]

Have you ever tried it with a reversed layout? It sounds like it would be confusing, but I promise you, it’s really really easy if you’re used to using your right hand. Basically, if you reverse the layout, you just pretend it’s your right hand doing the motion, and your left hand already knows what to do. If you can avoid consciously thinking about it, it “just works”.

I have, actually. I found it incredibly confusing. The only keys I'd perhaps want to reverse when switching hands are 0 and . because of thumb tuck. However, the answer there seems to me to be three 1U keys: 0 . 0 and only use two of them whichever thumb you're using. Make . only mildly inconvenient for either thumb (instead of quite annoying for the right thumb as it normally is) without substantially affecting the usefulness of 0. I never understood why 0 is the double-width key instead of . (Yes I know about calculators with 00 and all that, but if that's such a concern, then don't put anything there if you don't want a 00 key -- and I can see why you might not want a 00 key on a device designed to generate single keystrokes. Even now you'd have to fake it, macro style or multiple scancode style.)

I've just put all the tools away after adding an Alt key to the chopped Logitech Deluxe 250. Now I can do symbols again without having to call up Charmap. I still have no idea if the fact that Alt doesn't span keyboards is because of the PS/2 vs. USB thing and won't know until I have another USB keyboard to test with... which was supposed to happen today, but it didn't arrive.

[pietergen]

Regarding mirroring, I think both are good solutions. I say that with a mirrored layout you remember the 'motions', like "I type this sequence using index-ring-pinky". That sequence is then maintained on the other hand. When the right hand layout is not mirrored, but simply put on the right hand, you have the benefeit of the same 'locations'. Like "ζ is left of σ, above η and to the right of τ"

For the record, I am right handed but have been mousing left handed for 10 years (this succesfully stopped RSI in my right shoulder). I do this with a right handed mouse. The benefit is that I can use the mouse with my right hand as well, and I also can use the mouse of other people (90% are right handed).

I think that a left handed mouse would be easier to learn though, when I started it was confusing to left click with my middle finger. And to switch to index finger left clicks when i put the mouse on the other hand. But you learn this in a few weeks

Short version: when you switch often beween right hand and left hand, I say a mirrored approach is best. Forced by other factors (the wish to use any mouse on both left and right hand, without changing PC settings), I used the 'non-mirrored approach' though, which is harder, but still very possible, to learn. I you never switch layouts, it matters even less. Both approaches are good.

[Mal-2]

Here's my latest idea, inspired by both the Tipro and the Chicony KB-5581 (one of which I had, ages ago -- it fell apart). This is the QWERTY version, though of course I'd be using it as Dvorak. It makes no difference, really, as the two layouts differ only in their alpha areas and the last two keys on the number row.


And a keyboard-layout link because Binhoster could suck-start a Harley through the tailpipe.

It's 15U wide, because although I could shave a little bit off, this would come at the cost of using non-standard key sizes for the >1U keys. Just not using the (thoroughly overkill for my needs) Tipro housing and its expandable bus will cut about 1U off of each end.

The Comp Char key (in green) would be a Cherry MX Lock switch, wired to Alt. This would allow it to be flicked before starting a character composition sequence, and flicked again after. For everything else (except the spacebar) I probably want Browns.

There are no Fn overlays. There are no macro keys. There are no keys with non-standard functions at all, though there would be a few duplicate keys (namely Numpad Tab and Shift, and the locking Comp Char key would be a duplicate Alt).This will allow me to recycle an existing, non-programmable keyboard controller I already have lying around. It does supports multimedia keys, so I may end up throwing a few of those into the empty space, but I never have grown to like them. Why does this matter? Because I can actually build this, even if I have to use a non-printed circuit board and wire everything by hand. In order to do so, I do need to find out how (or even if) Cherry MX compatible 2U+ keys ride on two switches. The only 1.75U key is CapsLock, and I wouldn't mind just using a 1.5U key for that and leaving a gutter between it and A. (This would also make CapsLock and Ctrl interchangeable, and I plan to provide a switch for this function, along with Winkey-disable.)

[bhtooefr]

You are aware of how stabilizers typically work, right? http://deskthority.net/wiki/Stabiliser

Cherry and Costar stabilizers are the common types on MX boards.

[Mal-2]

I am aware of how stabilizers usually work. However, on the Tipro, it's set up with a switch every 1U, but sometimes there is a switch in the middle of a 1.5U space. (The QWER row, left and right. On the ZXCV row, left and right. In several places on the spacebar row.) I'm asking if there are keys prepared to accommodate this rather unusual arrangement. For example, left shift. In the U.S. layout, it has a 1.5U Shift key with a "blank" 1U key just to the right of it. I am guessing this is set up also as a Shift key, but I would want a single 2.5U key that covers both switches, not a 1.5U and 1U button next to each other that do the same thing. If such does not exist, and nobody custom-makes them, then this makes the Tipro considerably less desirable, especially at a $400+ price tag.

I figured that if keys DO exist for the Tipro, then I could very well use the same thing and skip the whole stabilizer mechanism.

[kbdfr]

More information here.

Edit:

[…] especially at a $400+ price tag. […

You can get used Tipro keyboard for just a fraction of that price. 7bit may still have some used KMXs and I some used MIDs.