Conclusion / synthesis of the keyboard poll results


What would you need Compose for, now that acute / grave / circumflex / diaeresis (umlaut) / tilde are covered and any language-specific letters you need could be mapped efficiently on the AltGr layer?

Even with a dedicated Compose key, it's still 3 keystrokes for just one letter. Dead diacritics are still more efficient (2 keystrokes), and the built-in éèàçüöäß or locale-specific AltGr symbols are even more efficient (just 1 keystroke, not counting the modifiers which you can do efficiently with the shoulder buttons without moving your thumbs).

I'm not against a single-width space, but let's not keep running in circles here, the decision has been made.
I need compose for compose, compose is better than string-together compose, simple as.

The decision has not been made. Can we stop going in circles and pointing fingers at eachother. We all benefit with doing this right.
The decision for a single vs a double width Spacebar key has been realized. The Spacebar is two key lengths long. Compose is firmly set on Fn + Spacebar, which reassuringly is a very very easy, almost like pressing one key, combo. There's also a hint of at least one User definable key for the layout, which could also be remapped by the individual to Compose. Where there's a will there's a way.  :)
 
Should the Space button have nothing on it, Compose symbol or leave it like it is?

You cant move in a straight line from button space → button because with noncentered doublewide space that makes it button → space → space → button.
I'm going to let it go, except this line. That's a non-issue that I don't care about.
 
There is actually a symbol for space, which is used on most compact keyboards.   ␣ (Unicode U+2423)

I see no reason not to use that symbol,  though I would also be happy to have it labelled "Space" like on the Pandora.

-Neelix
 
Saber: Fn + Spacebar is  "very very easy" reassures me of nothing, it isnt as easy as a dedicated compose, which is why it is worse. You lose not only a configurable button to a sub-par compose, you lose the Fn+ onto that button.

Dualwide                                                                                   Singlewide

Compose less tactility and function                         Dedicated compose, better in every way.

Fn+button to the right of space used                      Fn+button to the right of space available for anything

The decision on space has not been made. First it was doublespace, then it was single, then it was double, and now its

"we will most probably stick with double space" thats a quote from ED.

Its hard to draw a consistent conclusion from the arguments you can make to support doublewide.

Singlewide is just flat out better.

You can always remap anything to your likings

The main goal is to find a layout that fits as good as possible for ALL people out there, not for certain tasks.

Other insight

And latest: "The keyboard layout should be almost finished here now as well!"

When the injection-molds are made either with one less or one more button, is when the decision is made.

Natsu: If you dont care about efficiency (or any other benefit, or sets of benefits) why wouldn't you want them for the people that do care? If its irrelevant to you, making it better for more people still benefits you. If not in an egalitarian way, then certainly by the premise of selling more devices.

Neelix: We went through that discussion some time ago, cant remember what the conclusion was, but a space key, single, without anything, works for even the most novice of novices in my testing. I specifically tested this by asking them to say how they would write a sentence.

I opted for no prints on space since that is a cleaner look. It is limiting into other areas to have a dualwide space with no labels.
 
Last edited by a moderator:
If you are going to put a secondary function on Space and have it labelled, (e.g. compose) then you need to also indicate somehow that it is also the space key.   (Especially if it were only one key wide)  The aforementioned space symbol in the primary key colour would accomplish this nicely.

As for the width of the space key, that decision was made and announced months ago, you're just in denial.

-Neelix
 
What would you need Compose for, now that acute / grave / circumflex / diaeresis (umlaut) / tilde are covered and any language-specific letters you need could be mapped efficiently on the AltGr layer?
I'm an artist. I want to make pretty pictures using obscure symbols in text while i befuddle and bedazzle people with my prose.


I'm never going to use it for typing other languages anyway... but don't give me a pallet and say 'You have a great selection of colours here, but if you want monkey bottom pink or silly simon yellow, you will have to remap the olive green or daffodil yellow'


I want ALL the colours!!!


Yes, I know it's just one extra keypress.... but that's the difference between having to wash my brush everytime I want those obscure colours, or having a brush that's ready to go! When inspiration strikes, it makes a big difference.


Plus, a singlewide space will give me ANOTHER 2 BRUSHES!!
 
Neelix: The thing is, if its one key wide, you dont have to put two functions on it. Or, one function, and one meta-function to be precise. It solves itself.

My personal preference would be blank space, and compose symbol on the compose button for improved visibility and less clutter.

A blank button saying nothing says "space" a lot more clearly than a doublewide button with either "|_| compose" or "         compose"  on it.

I know it was, so then I discussed it with ED. Im not saying he always says its ok, im saying he does both. As of late, he said what i quoted earlier. So its a fair point to bring up.

Making apologist remarks against your own stated opinion is where denial comes into play.

Why is a compose included if its not needed? is all the rethorical questioning you need to back up a better compose being better. Fn-typing, or AltGr-typing of language isn't efficient, nor is it how typing works for the great majority of people. Wb knows this btw. Nobody wants to learn a new way to type for no good reason, ever.

I just like designing things, nesting up misconceptions is a lot harder.

Lets focus on the reality of the matter, and the actual A vs B comparison.
 
Last edited by a moderator:
What would you need Compose for, now that acute / grave / circumflex / diaeresis (umlaut) / tilde are covered and any language-specific letters you need could be mapped efficiently on the AltGr layer?

Even with a dedicated Compose key, it's still 3 keystrokes for just one letter. Dead diacritics are still more efficient (2 keystrokes), and the built-in éèàçüöäß or locale-specific AltGr symbols are even more efficient (just 1 keystroke, not counting the modifiers which you can do efficiently with the shoulder buttons without moving your thumbs).

I'm not against a single-width space, but let's not keep running in circles here, the decision has been made.
I need compose for compose, compose is better than string-together compose, simple as.

The decision has not been made. Can we stop going in circles and pointing fingers at eachother. We all benefit with doing this right.
Yes, but which letters/symbols are you going to Compose and why is the efficiency gain of a dedicated Compose key going to make a noticeable difference in your use case? (i.e. why would you not use remapping or the AltGr layer instead?)

I know you're not talking about things like Æ or Ø, because you would never consider to use Compose sequences for those. You're also not talking about letters with the common diacritics (acute/grave/umlaut/circumflex), because those are already better covered using dead diacritics. So what is it that you need a dedicated Compose key for?

In any case, if you really want to, you can assign a dedicated Compose key to keyboard Fn or keyboard Shift (provided you don't use those as a modifier, since they're already on a shoulder button). Or to Usr1 of course.

What would you need Compose for, now that acute / grave / circumflex / diaeresis (umlaut) / tilde are covered and any language-specific letters you need could be mapped efficiently on the AltGr layer?
I'm an artist. I want to make pretty pictures using obscure symbols in text while i befuddle and bedazzle people with my prose.


I'm never going to use it for typing other languages anyway... but don't give me a pallet and say 'You have a great selection of colours here, but if you want monkey bottom pink or silly simon yellow, you will have to remap the olive green or daffodil yellow'


I want ALL the colours!!!


Yes, I know it's just one extra keypress.... but that's the difference between having to wash my brush everytime I want those obscure colours, or having a brush that's ready to go! When inspiration strikes, it makes a big difference.
OK, so you want Compose to make pretty pictures with obscure symbols?

Most Compose sequences are of the form [Compose] [symbol] [Letter], where Symbol usually requires you to use the Fn modifier anyway.

Take the example of ā (a with macron), which is Compose - a.

Without dedicated Compose:

[Hold L2 (Fn)]   [space (Compose)]    [X (-)]    [Release L2]    [A]

With dedicated Compose:

[Compose]    [Hold L2 (Fn)]    [X (-)]    [Release L2]    [A]

The efficiency gain is very small to non-existent. And I would say that the use case of "making pretty pictures with obscure symbols" is not one we should optimize for.
 
^ Ok, i wont argue that... but what about all the other numerous benefits to had from a singlewide space?
I think the question is rather:

"What is the benefit of dualwide space in a thumb-oriented keyboard?"

I dare say the answer to which would be:

"None."

"Easier to hit."

In the context of writing? Why? You write letters too, space alone does not help.

In gaming? Baah... Emulated platforms which have space do not have 6 actionbuttons in games! C64, Amiga, AtariST: 1 lousy joystick button! 2 mousebuttons (but C64 has like 1 mouse title to begin with) --> just map space to actionkeys!

PC gaming? Yes, more action buttons in that if you emulate dualsticks, but most flight sims used those buttons for the same function as... yep, space and enter!!! --> map to actionkeys!

"Looks better?"

Personal opinion and counterargument: Bollocks!!! A symmetric keypad in a device like this is far more pretty and the dual wide space makes it uglier, not prettier. To each his own...

But this is all moot for 2 reasons:

1. I think most keyboard activists here actually would prefer singlewide, it's just that they are trying to see the dualwide as "not the end of the world" because...

2. ... it does not matter whether molds have been made - the case design was supposed to be done and if this is the only thing that would need fixing, then is it really worth a delay alone? Sure, if the design had something else too, maybe a rally to change this might be reasonable.

So: Especially Comradekingu. I think you're trying to convince the wrong people here - these guys, I think, would probably be all for the singlewide space, they're not opposed to it. You need to start bombarding ED, not these guys!?

(edit additon: And CK, be rest assured, if you succeed and in December ED says "Pyra 2 months late because of single wide space", then I, personally, shall be annoyed at you, I doubt that means anything to you, but I really shall, big time, pissed, even ;-)
 
Last edited by a moderator:
I'm gobsmacked. Truely.
Would that you were, I'm afraid. There's nothing more indicative of the state of humanity that a group of supposedly intelligent people would prefer to sit in front of their computers arguing with eachother over the relative merits of the size of a spacebar.

D.
 
The only advantage of a wide space key is a psycho-visual one: I must admit that it looks somewhat "friendly" and "comforting" (to the extent that these adjectives can be attributed to keyboards) to have a double-width space.

It does not add anything in terms of functionality, in fact it takes away some functionality: one less key, and harder to remap for e.g. dvorak or russian layouts, without any real gain in ergonomics or efficiency.

But it does have a non-negligible psycho-visual advantage. It says: "look, we are not cramped for keys, we had room for a big space key" (even though that's not quite true, but then again, even with one extra key we would still be cramped for keys).

If we would go for a single-width space at this late stage (and that's a very, very big if), then I would go for something like this:

http://www.keyboard-layout-editor.com/#/layouts/cba4f8e25726df399aa6d71e80806049

with Fn, Shift and Space on the left, and comma period and a dedicated key for / ? on the right in their standard M+3 positions. The remaining ascii punctuation goes on the bottom two rows (following a shift-pair convention) on the right half. There are three keys without an Fn label because we have an extra key; they could be filled with ñ £ ù or something.
 
I'm gobsmacked. Truely.
Would that you were, I'm afraid. There's nothing more indicative of the state of humanity that a group of supposedly intelligent people would prefer to sit in front of their computers arguing with eachother over the relative merits of the size of a spacebar.

D.
You seem to be trivializing it as a mere aesthetic.
If that was all it was I wouldn't care one jot.


Those of us getting behind the idea of singlewide space are arguing for increased FUNCTIONALITY.


... and functionality is the driving philosophy behind the Pandora and the Pyra, so this is far from a trivial matter.


Note also that for my personal use, it doesn't matter 'that' much.


But I can clearly recognize the huge benefits to be gained for the wider community, increasing the appeal of the device and most probably, sales.


I won't list all the benefits here, as they have been pointed out repeatedly already.


(Perhaps someone would like to gather them all into a concise bullet point list?)


I recognize that many would be pissed with another 2 month delay, but seriously... let's think beyond just ourselves... think what this would mean for the greater community and project as a whole.


Now waiting another two months to get it done RIGHT, I do regard as *trivial*.


Let's man up and do this properly guys. C'mon!
 
If we would go for a single-width space at this late stage (and that's a very, very big if), then I would go for something like this:

http://www.keyboard-layout-editor.com/#/layouts/cba4f8e25726df399aa6d71e80806049

with Fn, Shift and Space on the left, and comma period and a dedicated key for / ? on the right in their standard M+3 positions. The remaining ascii punctuation goes on the bottom two rows (following a shift-pair convention) on the right half. There are three keys without an Fn label because we have an extra key; they could be filled with ñ £ ù or something.
While I do agree that a single width space button would be better, I like that layout less than the one we have now. We've discussed this before, but having Fn and shift on opposite sides is a good thing. It makes it easier and quicker to type.
That said, how realistic is this discussion? I recall ED being a proponent of the double space button. Changing it would invalidate the poll and thus its interpretation. We would basically go back a few weeks (or even months) on layout design (most layouts were designed with the assumption of a double-width space). In addition case design would need to be altered, retested etc... That takes time and money.

I think end of this year is already impossible to achieve, such a change would move it back even further. This discussion is potentially endless, there comes a time when you just need to move forward.
 
We should rid ourselves of this troublesome space-bar-thing by installing a camera and use right to left head tilts for space and backspace.

Nods would be Tab and wild shaking about could be Enter/new paragraph.
 
I'm gobsmacked. Truely.
Would that you were, I'm afraid. There's nothing more indicative of the state of humanity that a group of supposedly intelligent people would prefer to sit in front of their computers arguing with eachother over the relative merits of the size of a spacebar.

D.
It's not the size, it's how you use it...
... urm, sorry...

I'm impressed anyone can keep up with these threads, there's so many of them, and they're all full of repetition. Although it probably consumes a lot of one's life; time I'd rather spend doing something useful.
 
If we would go for a single-width space at this late stage (and that's a very, very big if), then I would go for something like this:

http://www.keyboard-layout-editor.com/#/layouts/cba4f8e25726df399aa6d71e80806049


with Fn, Shift and Space on the left, and comma period and a dedicated key for / ? on the right in their standard M+3 positions. The remaining ascii punctuation goes on the bottom two rows (following a shift-pair convention) on the right half. There are three keys without an Fn label because we have an extra key; they could be filled with ñ £ ù or something.
While I do agree that a single width space button would be better, I like that layout less than the one we have now. We've discussed this before, but having Fn and shift on opposite sides is a good thing. It makes it easier and quicker to type.

That said, how realistic is this discussion? I recall ED being a proponent of the double space button. Changing it would invalidate the poll and thus its interpretation. We would basically go back a few weeks (or even months) on layout design (most layouts were designed with the assumption of a double-width space). In addition case design would need to be altered, retested etc... That takes time and money.


I think end of this year is already impossible to achieve, such a change would move it back even further. This discussion is potentially endless, there comes a time when you just need to move forward.
I think you're right. I'm very happy with the layout ED showed on Twitter so I suggest to move forward with that one, perhaps refine it a bit typographically and then consider it finalized.

In terms of functional layout it's perfect, and while an additional key for a primary slash or dash would make it even more perfect, I'm perfectly happy with the current level of perfection.

Compared to the Pandora, it will be a major improvement, and I already liked the Pandora keyboard. All annoying idiosyncracies are fixed, there are several new dedicated keys, the numbers/F-keys will be much more convenient, non-English languages will be much better covered, etc.

Can we conclude these keyboard wars in peace now, agree that it will be a major step forwards and agree to disagree on the remaining points?
 
Can we conclude these keyboard wars in peace now, agree that it will be a major step forwards and agree to disagree on the remaining points?
While I don't want to be the voice of dissension here, I'd endeavor to fix some things before the bus leaves the station, before we can't undo anything unforeseen by our hastiness.  B)

One point I'd like made is, aside from the hardware toggles and Insert/Delete, is how important I think it is to maintain independence with the gaming controls(everything above the Pyra key) from the QWERTY layers. North does it's thing and South does it's thing autonomously. If you're more inclined to be a thumb text entry oriented person(many are) versus an index finger oriented person this works out rather well than attempting to harmonize the two halves.  

A solution bandied about was to make Left Shift sticky, which is fine if it will be sticky by default(not confirmed yet), but it's also cumbersome as well to have BOTH sticky-type keys working the same right thumb. It's a bit of an ache for one digit to do all this labor. Plus, if you aren't an acclaimed apprenticed aficionado of the Shoulder trigger modifiers, braces ( and ) then become somewhat of a bugbear to get at, as the layout appraises now. 

Another um, tidbit, is some least used key placements, namely ° § € on Fn + q, w, and e. I can think of several monumentally more important symbols who would rather vacation in those hotspots. Placing those symbols in those prime spots puts \ | ~ unduly inward to reach at Fn + b, h, and j for a coder's device too. Some reshuffling and reevaluating of this "big U.S./German keyboard" Shifted pairs and familiarly symbol placement convention may be beneficial to our usage.

Last but certainly not least is brightness controls, on one key. From what I cognizant here display brightness increases by one press of this solo button while Shift + this button is decrease display brightness, but this isn't very or at all intuitive without the Shoulder button. Same with mastering the keyboard backlighting with Fn + this key. I have proposed using the Dpad with the keyboard Meta for brightness controls, which I see no problem with, but if all brightness attributes must be downgraded to one chintzy button then we should switch it to where the F11/F12 Usr1 button is so it's not on the same concordant side of the dual sticky keyboard modifiers. :)    

These suggestions above are only meant to improve the Pyra experience for all users. 
 
Last edited by a moderator:
Those of us getting behind the idea of singlewide space are arguing for increased FUNCTIONALITY.

... and functionality is the driving philosophy behind the Pandora and the Pyra, so this is far from a trivial matter.
The philosophy behind the Pandora was maximising CraigIX's profits. The Pyra is something different, hopefully. That remains to be seen; once bitten and all that.

But I can clearly recognize the huge benefits to be gained for the wider community, increasing the appeal of the device and most probably, sales.

I recognize that many would be pissed with another 2 month delay, but seriously... let's think beyond just ourselves... think what this would mean for the greater community and project as a whole.
You got me. We could SOLVE POVERTY AND WORDWIDE HUNGER! We could finally attain colonisation missions to our neighbours! Spread technology to the poorest to help them get to a level of affluence they deserve!

MY GOD GUYS, THIS IS AN IMPORTANT THING WE'RE DOING!

I am so excited to be a part of this.

D.
 
Back
Top