Unfortunately, being a big company, they can’t do anything very radical or work from first principles. Just like every other big company today, all of their key

Unfortunately, being a big company, they can’t do anything very radical or work from first principles. Just like every other big company today, all of their keyboard designs end up being mediocre spinoffs of a basic keyboard designed in the 1870s which had evolutionaruy changes (not all ergonomically positive) in the 1920s–30s, and then again in the 60s, then again in the 90s. The IBM keyboards produced from 1960–1990 are better than anything Microsoft puts out today: faster to type on, more ergonomic, much more reliable, sturdier, ... and of course much more expensive to produce. For particularly nice examples, take a look at these, https://deskthority.net/wiki/IBM_Beam_Spring_Keyboards https://deskthority.net/wiki/IBM_Beam_Spring_Keyboards Or pick your favorite other vendor. Pretty much all of keyboards sold in the 70s–80s were better than anything available at mass scale today. Back then computers were competing head to head with typewriters, and the best electronic typewriters were really nice to type on. I’m partial to the Canon typewriters of the early 1980s with nice Alps switches, but e.g. some Olympia typewriters of the early 80s were deliciously clicky. For at least the 2 decades from 1980–2000, pretty much every change in mass-market keyboards was driven by cost cutting. It’s not too surprising that quality degraded. Even if we assume we can’t change the basic keyboard shape, one of the important features that Microsoft (and most other) keyboards have dropped since the 90s is that further-away rows on keyboards used to be elevated above the home row in a sort of step-like pattern. This makes it much easier to reach the tops of those keys, and therefore speeds up typing them. The original designs from the typewriters of the 60s and 70s were based on research done by Honeywell and IBM and imitated widely, but later keytop profiles (including for modern Microsoft keyboards) were designed by people who didn’t understand the reason for the design, and just imitated a progressively watered down form. (Indeed an even more aggressive step than the ones used on those old keyboards is ergonomically preferable for most typists.) * * * All of Microsoft’s fancy lab studies seem to me like A-B testing all of the possible choices of features for a penny-farthing bicycle without ever considering adding a chain. https://en.wikipedia.org/wiki/Penny-farthing https://en.wikipedia.org/wiki/Penny-farthing If what we care about is ergonomics (which is to say, reducing static strain on muscles in the hand/arm while using the strong and efficient part of the main finger flexors’ range of motion to type, and reducing the finger and arm motion required to reach the keys), then the Maltron from the 1970s is much better than any of the dome-shaped Microsoft keyboards https://en.wikipedia.org/wiki/Maltron https://en.wikipedia.org/wiki/Maltron Better still, my favorite concept is a never-mass-produced ergonomic keyboard design called the DataStealth from the 1990s, http://web.archive.org/web/20000601172323/http://www.protomic.com/products/products_01.htm http://web.archive.org/web/20000601172323/http://www.protomi... which is in my opinion the most anatomically informed design ever seriously undertaken, developed by an expert from first principles. But also check out this awesome IBM patent from 1964, https://geekhack.org/index.php?topic=63415.0 https://geekhack.org/index.php?topic=63415.0 Or see the Japanese TRON project keyboards, http://xahlee.info/kbd/TRON_keyboard.html http://xahlee.info/kbd/TRON_keyboard.html Or more recently, Keyboardio, https://shop.keyboard.io/ https://shop.keyboard.io/