Keyboardio is a Hackable 'Artisanal Keyboard' That's Already Kickstarted (Video)

Video Keyboardio is a Hackable 'Artisanal Keyboard' That's Already Kickstarted (Video) 46

Posted by timothy on Wednesday July 15, 2015 @05:42PM from the it-even-has-an-'any'-key dept.

Video no longer available.

I bumped into Jesse Vincent and his keyboard project called Keyboardio at last year's Solid conference. Then, it was a developing project with a lot of literally rough edges, but since I'm a bit of a keyboard enthusiast, it grabbed my attention. In the time since, his plan to bring a truly hackable keyboard to the world has gained momentum, and the dozens of layouts and material combinations that he and partner in design Kaia Dekker have considered have been boiled down into one nearly-ready final version. The result is a compact split keyboard housed in an "heirloom quality" wooden case. It has some features you might consider overwrought -- like an RGB LED beneath each key, a precision mouse feature via WASD keys, and the ability, theoretically, to put more than a dozen feet between each half of the board. But if you're designing a keyboard from scratch, why not?

Vincent and Dekker put their project onto Kickstarter, then spent weeks on a road trip showing it at hacker and maker spaces around the U.S.; the project updates make a nice travelogue about just how widespread and varied is the world of DIY culture. I caught up with him in Colorado Springs, Colorado, on the road between some of those demo gigs, to talk about the long path from idea to (hopefully) shipping a product to backers. By the time we had this conversation, the project was well past fully funded, andI was impressed enough to order one myself; hopefully, the clicky keys will be worth the cost of a middlin' Chromebook, though Vincent admits they're not going to fool anyone looking for a buckling spring action. On the other hand, at least at the Kickstarter price, it beats some of the Maltron keyboards I've been eyeing for years. Plus, it comes with a screwdriver.

Slashdot: Jesse, what is wrong enough with keyboards right now that you see a place for a $300 plus keyboard in the world?

Jesse: So, the weird thing is that a $300 plus keyboard isn't actually that atypical of the high-ended keyboards. If you look at a high end gaming keyboard or pretty much any reasonable ergonomic keyboard out here, that's about the typical price. So typically they are made of plastic, the entire body is made of plastic, very often they use to do sub-par key switches. They're not actually all that comfortable, they're super un-hackable; they don't look pretty. A lot of them look more like medical devices or torture devices than things you'd actually want on your desk to be using every day.

Slashdot: So, by the way, I should interrupt here and point out the reason we're hearing all these torturous noises in the background is that you're actually on a multi-city tour promoting the keyboard you're working on. Where are you right now?

Jesse: So, right now I'm at a Starbucks somewhere outside of Colorado Springs, Colorado. We are now just past the halfway point by mileage in our Kickstarter roadshow. We're basically - we're going city to city stopping mostly at hacker spaces letting folks play with the Model 1, talking about how we made this thing and how if they don't like what we've done, they can make their own keyboard from scratch. I started off as a software guy and I've had to skill up on pretty much every aspect of making this whole stuff to get to the point where we are now.

So our second day, we actually stopped at Kickstarter in New York. One of our beta testers is a Kickstarter employee. And talking to folks there, as far as they can tell we are the first Kickstarter project to ever do a Kickstarter roadshow other than roadshows where the whole point of the Kickstarter campaign was funding that trip.

Slashdot: So back to the thing you're doing, so people aren’t too lost here, you are working on a what I'd call an artisanal keyboard. It’s got a little bit different materials, you mentioned the ones that look like medical devices. What by contrast are you doing with Keyboardio?

Jesse: So the Model 1 is – that enclosure is milled out of two blocks of maple, the key caps are individually sculpted to help guide your fingers into the right places. Some of the keys that usually is up on your pink keys, now within a nice arc underneath your thumbs, there's a function key that lives underneath your palm. So for example when you drop either palm under the function key, HJKL becomes the system arrow keys. WYST will move the mouse; EYLP are your brackets, RTFG will actually let you binary search the screen, it's potentially a graphics tablet and lets you hit pretty much any button on screen in about seven key strokes. No, it's not actually a proper replacement for a mouse, but it's kind of useful. Other things that we're doing that are a little different. There's a red, green, blue LED pixel underneath each and every key so you can make the keyboard light up to do whatever you want.

One of the coolest things that somebody suggested that we hadn't thought of is when you're in a root shell in a production server, your keyboard should glow red, so you know that if you do something wrong, you'll get yourself in trouble.

What else is in – oh, the entire thing is super programmable. So we made the decision fairly late in the process to switch to Arduino. So most hardware products switch away from Ardunio fairly quickly because folks consider it to be expensive, folks consider it to open your in production; using something like an Arduino or the chips inside an Arduino is not the cheapest way to make this, but it makes it super hackable. So the way we do – explain this to people is that your keyboard is becoming the source code and the screwdriver. Open hardware typically ships as a kit and gorgeous finished products typically ship as these sorts of things that you're never supposed to open. And we believe that it's a false dichotomy. You should be able to – the thing you buy from us you own it. You should be able to open it, you should be able to flash the firmware, you should be able to flash the bootloader, you should be able to make it do things we never imagined.

Slashdot: Two things I want to ask you out there, one you actually provide accessible external physical access to the Arduino for reprogramming?

Jesse: So it is not actually - so what we did is we took the chip that’s on an Arduino Leonardo or an Arduino Micro and we actually built it into our own board but we then chose the Arduino bootloader so that you can flash it with Arduino sketches and we actually wrote our default keyboard firmware in Arduino C, so that it is easily accessible for folks who might be nerdy like us but aren't necessarily hardware hackers or embedded programmers. Arduino is embedded programming that's designed to seem easy, so essentially what we're doing is we're tricking people into embedded programming by letting them hack at their keyboards. But also if you want to do bit further and reflash the bootloader, it's just a couple of screws on the bottom of the enclosure and we provide the pins and we provide the bootloader source.

Slashdot: And the second thing I want to ask, since you mentioned it, many projects go as we know from Arduino to a smaller factor chip instead. What are you moving away from and why?

Jesse: So we're actually moving away from other things based on the same chip, but they didn't have open bootloaders. And so it was a thing where we could – if we stuck with what they had, we wouldn't have been able to offer quite as much hackability to our users. The reason that we stuck with the AVR chip, the ATmega32U4 that we did is because it's one of the options in the Arduino at Heart program, where we commit to doing a bunch of things to Arduino and Arduino commits to us in a way. So we’re actually listed on Arduino’s at Heart program or at Heart product pages and I think they're putting out a press release about us.

Slashdot: Can you talk about the hackability that you mentioned. What are some examples that somebody might want to do. You’ve mentioned the red lights to indicate you’re on a server. Are there any other sort of booster ideas that you run into?

Jesse: So there are people who talk about wanting to add footpedals and we're not planning on adding foot pedals, but there are a couple of different ways that you might do it at home on your own that won't void the warranty. That's kind of cool, and one of those ways is, literally opening the keyboard up and attaching wires to pins inside and that doesn’t void the warranty in and of itself. Other things that we’ve – I mean sometimes it’s just simple as just remapping the keys so that they are where you want them to be. Sometimes it's doing cool things with the LEDs. One of the things that I really want somebody to do or I want to do is to add a counter so when your typo rate gets too hot, typo rate could be measured by doing something like looking at keystroke backspaces or other keystroke.

Slashdot: You mentioned also the VI keys, whether it be a special set for those of us who are not big VI fans or let's say those in the world who aren't big VI fans.

Jesse: Well, so to be clear that is just the arrow keys are on home row and it’s just that we picked the positions that are most familiar to UNIX, C people and they are labeled. And what we're doing for keycaps, what we have committed to for keycaps, it is the ability to QWERTY caps that have the VI keys labeled – the four arrows labeled along with some other stuff. You will be able to get complete blank caps, and you will be able to get caps that have little circle in the middle of each one, so the nice bright LED underneath shines through. We think we're going to be able to offer aftermarket custom laser engraved keysets at a pretty attractive price, but we are not promising it because we don't have quotes present to us that we can do it at a price that’s not obscene.

Slashdot: One of the things before we startedthis interview, you mentioned to me that one of the things you might not be able to open source immediately are some of the CAD designs and that's interesting to me, it seems like CAD designs are something that should be easiest. Explain what makes it easy to release or not release parts of the project for public viewing?

Jesse: So it's not so much that it is not possible to open source those as we're still trying to figure out where to draw the line on what is open hardware and what's not. It’s not so much that we can't really solve the CAD designs, but that we're still trying to figure out exactly what we're comfortable with. You're definitely going to get schematics. You're definitely going to get source code. You’re definitely going to get CAD for the enclosure. You're definitely going to get CAD for the interconnect mechanism between the two halves, if you wanted to build a custom mount that changes the angles that the two halves of the keyboard are connected at.

We're still trying to figure out for example if we're going to release the CAD for the keycaps initially, part of that is us being dumb and scared about it. It’s the thing that we’ve put the most effort into, it is the thing that is most costly for us to do. And I'm not sure we actually have a good reason not to release it, but it's a thing that is very hard to step back from once we do it and so we want to be very careful and consider exactly what we're doing as we do it. We have committed that if at any point we go under or stop being able to make keyboards, we will open source 100% of everything.

Slashdot: Sort of an escrow?

Jesse: Sort of an escrow. Not a formal escrow, more of an escrow based on ruining my reputation if we don't do it.

Slashdot: Jesse, I like the way that you have connected the halves of your keyboard, which it's not obvious from a single static shot, but you can actually separate the halves by quite a bit. Can you talk about the interconnection there?

Jesse: Sure. So, I'm going to hold up the keyboard that we’ve actually got here. So, this is the one I use every day and so it’s a little more beat up than usual. But there is this interconnect mechanism in the middle here that is part that will hold the two together. But the cable that we’re actually using to connect the two halves of the keyboard is ordinary phone cable. And we use that ordinary phone cable for a bunch of reasons. One, it has the six-pins we need for all of the wires of that interconnect, and two, it’s really cheap, and three, it's really easy for you to find and you to make your own cables, or to buy your own cables.

And so, you’re not stuck with whatever two or three lengths we end up shipping you. But you can take the two halves of the keyboard and you can actually get them – the furthest we've tested them is about 15-feet apart connected by cable; 15-feet works great; at 25-feet, it does not work great. If you can tell me why you need more than a 15-foot separation between your halves of the keyboard, we will make it go, but nobody has been able to come up with anything.

Slashdot: So, when your trip concludes and your Kickstarter is over, do you already have a vision in mind, are there things that you've already discovered and going from prototype to a first-level production. What’s on your wish list for five years from now?

Jesse: So, we're still working on our wish list for six months from now. So what we've been showing off on this trip are still essentially prototypes or beta units depending on how you want to describe them. And so we've actually learned a whole lot. We had to lock in this design about six months before Kickstarter, so that our beta testers could use it, so we could get twenty of them built out and we've learned a whole lot about things we’ve done wrong and so there is a whole lot of small improvements we're going to make before production.

So the interconnect mechanism, the physical, that plate that holds the two together, it's not as sturdy as we'd lie, so we are redesigning it to be more flexible and sturdier. The positions of a couple of the keys, we’re going to tweak just a little bit. The shapes of the key tops we’re going to tweak just a little bit. And right now, the widest part of the key caps is the narrowest part of the enclosure. So right now, this right there; with a little bit of redesign, we can make the whole keyboard about an inch narrower which might fit comfortably in a 13-inch MacBook bag, which solves a lot of transport problems.

We're not thinking about the next model until after we get this one into production. We are a small company and as we put our focus to another high end product before we got the first done, we wouldn’t do as well. We got some little hobby projects that we may put out in the interim as little kits or something. But for the most part, all of our attention is going into making the Model 1 good.

Slashdot: So, Jesse, I want to ask you about the aesthetics and the materials, two things that really intrigue me because I'm somewhat of a keyboard fanatic is, one you're offering a click-key version, which I'm very happy about that, and the other is that since you’re using wood, what are some of the complications of working with wood? Where do you get your wood and how do these pieces come together and is the production done in five different places around the world or what?

Jesse: Sure. So, I’ll start with the click-key. So, the key switches we’re using are made by a company called Matias. They’re outside of Toronto. They got their start making replicas of Apple's really good keyboards that Apple discontinued a number of years ago. And they were using switches made by a company called ALPS. And then ALPS stopped making their switches. And initially Matias got the last order of those switches by committing to a gigantic production run. And then once they were no longer available, they ended up reengineering and designing their own switches.

And so they were a little bit more boutique about making switches that a lot of folks are familiar with. The Matias quiet click switches, which are default switches, they are damped on both the down and up stroke, so you can use them in an open plan office which generally we consider to be a win because I don't want somebody shanking me for using my keyboard. We were actually a little bit surprised by how many people really, really wanted the WOWed version of the Model 1.

And we hadn’t seriously considered making a loud variant as part of the Kickstarter until the third day of the campaign when we were in Philadelphia and Eric Raymond showed up for the Keyboardio meetup at the Hacktory, which was the hacker space we were in there. And his comment was basically, I love the layout, it's super comfortable. Once you get mechanical pieces which – once you get loud switches – we hadn't seriously considered making a loud version of the Model 1 until the third day of the Kickstarter campaign, when we were to meetup at the Hacktory in Philadelphia and Eric Raymond showed up. And he looked at and said, it feels great, it looks great, I really like it. Once you get loud switches, I'm all in. And he was not the only person to say that, and talking to our switch manufacturer, they confirmed that it's very doable to do some small percentage of the keyboards with loud switches and so we went with it. They will be a little bit louder than the cheery blue that a lot of folks are familiar with. They won't be as loud as an IBM buckling spring, they’re not like machine guns. There were a few more really loud switches, you could probably drop a solenoid inside each half of the keyboard and have it fire each time you had a key-press and then you could really have variable sounds.

Slashdot: Now how about working with wood, talk about the complications of that.

Jesse: Yeah, so early on we got told we couldn't do wood, we got told it can't stay on wood, and for the most part we believed the folks who were telling us this because they've been doing this for decades and we haven’t. And then they recanted which was kind of cool, and it's true that if we were going to be selling 500,000 units of this keyboards, there's no way we could get away with wood, because the way that you make this wood enclosures is that you put them on what's called a CNC mill, which is basically – if you think about taping a Dremel to a robot, that’s what a CNC mill is. And so each one of these has to be done one at a time, really high end CNC mills can do either the time by taping essentially a really high end Dremel to a robot, and then they have to get hand finished, they need to get sanded, treated or polyurethaned or stained or however we finish them and there are a lot of complications to this, making sure that it's done in the right way, in the right place. If for example we make them in China and a lot of the places we do in China are very humid, there is a pretty good chance it will crack when they showed up in the very dry parts of the U.S. And so it’s something you need to be cognizant of, we need to be careful to source the wood basically from sustainable sources. And so there are places that farm the maple we are using, that can certify that they're not cutting forests, that can certify that they're not doing all sorts of horrible things, and it's very doable, it's just that we got to be really careful.

Slashdot: Let me ask one more thing. You've been using one of these for how long now? And what has been your experience as inventor, what sort of glitches have you run into and had to face?

Jesse: So, I've been using some model of what we’ve been doing for over two years now, and we've made over 30 different prototypes and I've used each one of them as my daily driver for a while, as we learned from it, things I've done wrong include incorrect key placement firmware plugs, soldering issues, the wrong shape of the enclosure, but I mean in general anything you can do wrong in development I've done, but that's iterative or evolutionary development; that's how we get the right thing by trying lots of things and paying attention to what things work well, not just for me, but for all of our clusters, and when something doesn't work well, changing it.

Slashdot: Will it need any kind of a special driver and will it come with any software to tweak it specifically?

Jesse: Sure. So to use it as a keyboard, it needs no special drivers. It should work out of the box on Linux and Mac OS and Windows and iOS and Android and really anything else that talks USB HID. We've only built it for one platform so far, but there will be a little script that you can run on your computers to tell the keyboard things like what the current foreground application is, so that the keyboard can bring up application specific macros or key maps. But all that is, is a program on your computer that opens the USB serial port of a keyboard and supports down a little bit of instruction. There will be nice web GUI for change in the keyboard layout and doing basic LED configuration. That's something where we’ll host a version, but it will also be a 100% open source so you can host your own copy, or of course you could look at the text files to change the config.

This discussion has been archived. No new comments can be posted.