There is a key difference going from "convergence" to "conformity": circular reasoning.
People who need to sell the thing they invent, tell themselves that "convergence" is the best way there. Customers like familiar traits. So they design for intentional convergence, and that is the essence of conformity.
99.99999% of keyboards are the same physical layout. This layout is a standard that conforms to the general shape of an 1800s typewriter. There is no functional need to continue this. There was no natural convergence toward this shape: it's so bad it often injures the user! The only driver is the value of familiarity to the act of selling a keyboard.
With the advent of cheap 3D printers, we have seen a surge in hobby keyboard design. These people care so much about getting a better keyboard shape that they are willing to spend hundreds or thousands of dollars to get there. What they want they cannot buy; this was especially the case 5-10 years ago. The very existence of this group proves that traditional keyboards are the result of conformity, not convergence.
I would dare say that 99.99999% of keyboard users expect a similar layout. Sure, it converged out of conformity and familiarity, but so what? This is part of the added value to me, I can learn to type in a particular layout and I'm good to go in any other. This is a crucial part of functionality in fact.
You are precisely the consumer demand that drives conformity here. You are real! That doesn't change any of the other things I just said: in fact, it exasperates the issue, because it isn't wrong to target your familiarity as a reliable selling point.
The problem is that it's also a good thing to try create new designs. I have a keyboard that you would hate to use. That's how I feel about the traditional design you prefer. We can both be served by the market, but only if designers have enough confidence to do so.
You wouldn't even care about this if you had originally learned to type on my keyboard. If that was the ubiquitous thing you started with, then we wouldn't even be having this conversation. Had the progression of keyboard design been free to explore other options, it might have naturally converged on something more ergonomic; which would serve both the need for familiarity and the desire for ergonomics.
We are already seeing ergonomic keyboard designs converge to 3 shapes: Dactyl-like extreme ergonomics that place keys as close as possible to the resting fingertip; flat compromises like ergodox that can use a PCB; and grid layouts like planck that optimize the difficulty of conceptual memory layouts when using many layers. All three place keys in columns, so they are pretty easy to switch between. The only real mystery left is the thumb.
Most of the discomfort involved in switching from the typewriter layout stems from how obscure and unwieldy it is relative to anything else. Like a Who from Whoville, we look at the elegant hammer - as a replacement for the familiar Whabam - with distaste.
For almost every product out there, a "better" version is available that is not the most common one. This is simple economics: the improvements you describe are not free, which is why they are not mass produced. If people judged that these improvements that you describe are worth it to them, they would pay. The fact you could buy such a keyboard demonstrates that conformity or convergence or whatever you want to call it did not inhibit you from getting what you want.
Where I think you get it wrong is to assume that everyone has your level of interest or applies the same logic as you to every item in their lives. Hell, I'm sure you don't. So if you are happy eating from suboptimal mass-produced forks which are just SO much worse than these $500 custom-made ergonomical marvels that the average man just simply doesn't know about... welcome to the peasant club.
They weren't even minimally produced until recently!
If any "better versions" were actually a real part of the market 20 years ago, then you could buy a laptop with one today. You can't. In a $140 billion market, there is not a single laptop manufacturer producing laptops with an ergonomic or ortholinear layout. That can only be the result of conformity.
> The fact you could buy such a keyboard demonstrates that conformity or convergence or whatever you want to call it did not inhibit you from getting what you want.
Not 4 years ago, but it certainly did 10-20 years ago. They simply didn't exist back then.
And it's not a binary, either: just because there is one company mass-producing one does not mean the problem is resolved. We are talking about one of the most commonly produced objects in the world. Practically every single computer (bigger than a hand) that has ever been made has a traditional typewriter-style keyboard!
And the fact I (and many others) an willing to pay >$300 for one is clear evidence that the market for ergonomic keyboards is incredibly diluted. If I could buy a poor quality one for $30, I would do it in a heartbeat. The average traditional typewriter-style costs well under $1 to produce. Starting that kind of production scale would take a huge investment.
The fact that no one has tried once in the last 30 years to mass produce a cheap non-typewriter keyboard design is proof that conformity, and not convergence is the deciding factor.
> I would dare say that 99.99999% of keyboard users expect a similar layout.
Why don't they expect the best one? Seems like the 99.99999% of users are fooled by monkeys and bananas on ceiling experiment. I can not name that amount of people just idiots but if your upper-home row consist a word "typewriter" I consider your choice as definitely not wise.
Yes. Exactly as what we are having, no extra keys or extra something, we are talking about different placing of same keys so I surprised to see this question.
> Is 'best' compatible with what I learned without invested more time in learning something new?
Well, why (in ideal world) the kbd supplier has to consider the needs of not wise people? If you really have learned Qwerty you will not even noticed a different paint on the keys.
> Is 'best' something that's going to live far beyond it's usefulness based on it's cost?
It is best in the sense that the best side of knife for cutting bread is the sharp one but it is one of the worst ones for holding the knife.
> Is 'best' something you can't hand to someone else and have them use it?
Most people have spent their entire lives only seeing a few small deviations from the traditional typewriter shape.
That's what they had, so they developed themselves alongside it. It was hard work, and they don't want to do it again.
Not having to do the work of creating new muscle memory is a quality many people highly value.
Unfortunately for the rest of us, the assumption has been that there aren't enough people interested in trying something new to have a viable market. It would be a risky business venture: at the very least, difficult to raise venture capital. That is a self-filling prophecy: circular reasoning that demands conformity.
Thank you for your analogy. Demand shapes supply is a capitalism fairy tale. For most markets you just can't buy what you want. Even if you would be willing to pay 2 to 4 times. You have to buy the mass produced sameness. We do not choose what there is to buy. In 95% of products and shops the supply dictates the demand, often with a simulation of choice.
Look closely, for many applications there's only one or two products used nearly world wide. Often they are neither the best nor the cheapest. And don't get me started on the "choices" when it comes to smartphones.
>For most markets you just can't buy what you want. Even if you would be willing to pay 2 to 4 times.
This is a problem I see everywhere... People have no idea that mass production leads to order of magnitude reduction in costs or more.
In most products you buy today there is a 4X price range between low to high end products. When suddenly you're talking about 10-100X price range people go with why they can buy.
You'd think HNers would have a better idea on manufacturing costs.
> There is a key difference going from "convergence" to "conformity": circular reasoning.
> People who need to sell the thing they invent, tell themselves that "convergence" is the best way there. Customers like familiar traits. So they design for intentional convergence, and that is the essence of conformity.
This is completely tautological. You're just saying "it's not convergence, it's conformity" and your only argument is "because I say it is".
They're only the same if you're taking an extremely reductive approach to them... And at that point you could also say that "99.99999%%" of Smartphones are the same, wherever they're iPhones, Androids or windows from a dacades ago. They're all rectangular devices with a large screen and touch input after all.
And for the record, keyboards have massive size difference, in all dimensions. Keyboards actually have a way bigger variance then you'd find in phones.
Some are just a few mm high, others go as high as 5cm. There are 60% keyboards. Ergonomic keyboards.
Keyboards that're basically just a small nipple with which you're doing input by gesture.
the keys and actuation themselves are another topic with Rubberdome on cheap and trashy keyboards, mechanical with springs and other techniques, even some optical - though they're pretty gimmicky.
There are stylized key caps off all kinds, including some really cringe material like ahego keys from a few years ago. And actually quite pretty one's with mountains, clouds and areas stylized under resin [1]
Some are so small you can only use them with your thumbs
Keyboards don't inherently need to have the keys laid out like a typewriter. Most people are only willing to use a keyboard that is, because that is what they already developed their muscle memory for. That's the motivation behind the conformity I'm talking about.
A natural convergence would start with many competing options, and gradually boil down to what's popular; but keyboard shapes have only recently seen meaningful diversity be available at all.
I get where you're coming from and it'd definitely agree that the average keyboards has strongly converged on a standardized layout.
Nonetheless, there isn't a single layout. You got ANSI and ISO as the big branching standards with adjustments to bindings and added keys for various extra letters/modifiers depending on the localization.
And yes, i brought up the phone example to illustrate how they're only superficially the same. Your argument maps exactly onto phones after all, as they all have similar looks, input pattern etc.
> I get where you're coming from and it'd definitely agree that the average keyboards has strongly converged on a standardized layout.
They didn't converge: that was my whole point. They never diverged (significantly) from the original typewriter. All of the letter keys are in the same relative position.
Where to place function and return was a new question (the typewriter didn't have those), so naturally that aspect actually had variance until converging on a few standards.
Take one look at a dactyl, and you can see what I mean.
There is a key difference going from "convergence" to "conformity": circular reasoning.
People who need to sell the thing they invent, tell themselves that "convergence" is the best way there. Customers like familiar traits. So they design for intentional convergence, and that is the essence of conformity.
99.99999% of keyboards are the same physical layout. This layout is a standard that conforms to the general shape of an 1800s typewriter. There is no functional need to continue this. There was no natural convergence toward this shape: it's so bad it often injures the user! The only driver is the value of familiarity to the act of selling a keyboard.
With the advent of cheap 3D printers, we have seen a surge in hobby keyboard design. These people care so much about getting a better keyboard shape that they are willing to spend hundreds or thousands of dollars to get there. What they want they cannot buy; this was especially the case 5-10 years ago. The very existence of this group proves that traditional keyboards are the result of conformity, not convergence.