Skip to playerSkip to main contentSkip to footer
  • 4 days ago
The Verge’s Victoria Song joins the show to talk about a new genre of gadget, which both she and David have been testing a lot: the AI-powered, always-on voice recorder. Vee shares what she’s learned from devices like Bee, and why it’s going to be so hard for AI to figure out what really matters in our lives. After that, The Verge’s Nathan Edwards and keyboard maker Ryan Norbauer tell the story of the Seneca, a $3,600 keyboard that Norbauer built to his own incredibly exacting specifications. They tell David about what it really takes to make a great keyboard, and why making one is worth the effort. Finally, in the lightning round (call 866-VERGE11 or email vergecast@theverge.com!), we answer a couple of questions about the future of Chrome.

Category

🤖
Tech
Transcript
00:00:00Welcome to The Verge Cast, the flagship podcast of outrageously expensive keyboards.
00:00:07I'm your friend David Pierce, and I am on my couch in my basement, and I think I live
00:00:12here for the next several weeks, hopefully.
00:00:16So I'm not a particularly superstitious person in general.
00:00:20I don't have a lot of, like, rituals.
00:00:22I don't have a pair of lucky socks.
00:00:24But here's the thing I know for sure is that I sat in this spot on this couch for every
00:00:30single Warriors playoff game in 2022, and the Golden State Warriors won the NBA title.
00:00:36Did they do it because I sat on this couch?
00:00:39Who's to say?
00:00:41But these are the rules.
00:00:42I have to do this now.
00:00:43So ironically, it's possible that by the time you're watching or listening to this, you'll
00:00:47know who won the game that I'm sitting here about to watch.
00:00:51But I don't.
00:00:52And these are the rules.
00:00:54And maybe I can leave the couch, and maybe I live on the couch.
00:00:56Maybe this is where we Verge Cast from, from now on.
00:00:59I don't know.
00:01:00Go Warriors.
00:01:01We'll see what happens.
00:01:02Anyway, today on the show, we have a bunch of fun stuff to talk about.
00:01:06V-Song is going to come on, and she and I are going to compare notes on a device category
00:01:09that we've both been testing.
00:01:11We have these recorders that you're designed to wear and kind of have on and with you all
00:01:16the time that say they can keep track of all the important stuff in your life without you
00:01:21having to do any extra work.
00:01:22I find it fascinating and weird.
00:01:24We're going to talk about it.
00:01:25Then Ryan Norbauer, who made maybe the most expensive keyboard I've ever seen, is going
00:01:31to come on the show with Nathan Edwards from our team.
00:01:33And we're going to talk about fancy high end keyboards and why they might matter more than
00:01:39you think.
00:01:40I think there is a there is a cool spiritual story to tell about keyboards, and we're going
00:01:44to tell it.
00:01:45Also have some hotline questions about what's going on at the Google trial.
00:01:49I have been at the Google trial a bunch of the last two weeks, and I have some answers
00:01:52for you.
00:01:53All that is coming up in just a second.
00:01:55But first, I have to figure out how to drag myself off of this couch and not feel bad
00:01:59about what happens next.
00:02:01Wish me luck.
00:02:02This is the Verge cast.
00:02:04All right, we're back.
00:02:09V-Song is here.
00:02:10Hi, V.
00:02:11Hi.
00:02:12V, we have to talk about the weirdest gadget category that you and I think are both like
00:02:16needlessly obsessed with.
00:02:17Are you ready for this?
00:02:18Yes.
00:02:19I'm so ready for this.
00:02:20OK, I also feel like I'm going to wear I'm going to wear this while we do this because
00:02:25I'm I'm recording us currently in two ways, and now I'm going to record us in a third
00:02:29way with this weird necklace around my neck.
00:02:32How do we describe these gadgets we want to talk about?
00:02:34They're like I've come to talk about them as like a voice recorders.
00:02:39Is that reasonable?
00:02:40I think I think I voice recorders is reasonable and or the way that B described it.
00:02:48And we'll talk about B is your A.I. memory, and that's like ultimately where they want
00:02:53to go with it.
00:02:55So am I like A.I. memory wearables is like I think the high level of what they want to
00:03:02be, but what they are essentially now is A.I. voice recorders.
00:03:06Yeah, they're all sort of different spins on the same thing, which is basically a dedicated
00:03:11device that is mostly some kind of wearable microphone.
00:03:16And the idea is that it records things, how many things and when and what I think is interesting
00:03:22and we should talk about it.
00:03:23But it's designed to be sort of an easily accessible recorder that then uses A.I. models
00:03:29to summarize the stuff that happened in your day to day life or let you search through
00:03:34it or pull out action items.
00:03:36They all have like slightly different versions of the same idea, but it does all feel like
00:03:41roughly the same idea.
00:03:43Is that a reasonable description?
00:03:45Yeah, it's like I guess if you had like an A.I. stenographer in your pocket that was
00:03:51just also telling you what you said, giving you a transcript of your life and then sometimes
00:03:58depending on the thing, giving you like action items based on the crap that you just randomly
00:04:05say and will like nine out of 10 times forget to write down.
00:04:10Yeah.
00:04:11Which like I actually really love that description, because like if you just think about it, that
00:04:15way, it sounds so compelling, right?
00:04:18The like.
00:04:19Yeah.
00:04:20Keep keep track of everything you need to keep track of, including all the things you
00:04:23forget to write down.
00:04:24My brain immediately goes, yes, I do want that.
00:04:27That is a thing that would be useful to me.
00:04:29And yet my experience with and yet very different.
00:04:32So you you wore the B in particular for a long time.
00:04:35So let's let's start there.
00:04:36Tell me a little bit about just like day to day life wearing the B.
00:04:42So the B is they kind of market it as you're like outsourcing your memory to A.I.
00:04:49You can wear it one of two ways.
00:04:50You can wear it as a little like fitness trackery looking bracelet or you can wear it as a pin.
00:04:55I have limited real estate on my wrist, so I chose to wear it as a pin for most of the
00:04:59time.
00:05:01It just records all of your conversations.
00:05:04And it's one of the ones you're supposed to have on like 24 hours a day, right?
00:05:08Like it.
00:05:10It has seven day battery life, so you don't even have to charge it at night per se, just
00:05:17like wear it all the time.
00:05:19And it's supposed to like as it currently functions, they have some bigger ideas about
00:05:24how it's going to go, but you know that we're not there yet.
00:05:27But as it is right now, you basically get a transcript of your life, summaries of your
00:05:34day, like little A.I. diary entries of how your day went, notable moments, a suggested
00:05:41A.I. to do list based on the things that were said.
00:05:45And it wrote very interesting fan fictions of my life.
00:05:49There's a little section called they call it, I think, like fact review.
00:05:54I called it A.I.
00:05:55Fact Tinder because you're basically swiping left and right about assumptions that it learns
00:06:00about you throughout your day.
00:06:02And that's supposed to help it learn extra context as to whatnot.
00:06:06So it could be something as mundane as Victoria has a living room and a kitchen, an actual
00:06:10fact that it gave me.
00:06:13And then another one was Victoria has a friend named Kendra Montesha who likes mustard and
00:06:18turning TVs off.
00:06:19And that's because it cannot interpret the lyrics of Kendrick Lamar's TV Off.
00:06:24So yeah.
00:06:27And then you get ones that are mundane that you're like, this never happened, which was
00:06:30Victoria had like, I forget exactly what it was, but it was something about my dietary
00:06:35restrictions and lollipops.
00:06:37And I was like, I did not talk about lollipops, though.
00:06:40Now thinking back on it, was it thinking about me talking about lollipops?
00:06:45I don't know the drink.
00:06:46So, you know, I you just review these fact tenders on like a revolving door basis and
00:06:54you just look at them and you're like, I never fucking talked about lollipops.
00:06:58What's it talking about?
00:06:59But yeah.
00:07:01I've had a really weird experience wearing this thing, even around people who don't know
00:07:06I'm wearing it and thus don't live their lives any differently.
00:07:09I feel weird in the sense that I feel like if you were if you were like working for the
00:07:13FBI wearing a wire, you would always be aware of the conversation that you were having.
00:07:18I became that just with this thing and I've been using it for a while and that has not
00:07:22gone away at all.
00:07:24Did it feel that way for you, too?
00:07:25I started talking less, one, in my actual day to day life.
00:07:29I stopped.
00:07:30I'm not like a big talker to myself, but every once in a while I'll be like, shit, or something
00:07:35like that.
00:07:36I just stopped doing that because there were there was an incident, a gastrointestinal
00:07:41incident in a bathroom and I said something and it summarized it.
00:07:44And I was like, this is fricking rude.
00:07:46And then it told me to start taking lactate again.
00:07:49And I was like, I don't need this.
00:07:51But as regards to how the people in my life feel about it, I'm just going to quote my
00:07:56spouse.
00:07:57And they said, this is not powerful or useful enough to violate my privacy consistently
00:08:04in this way.
00:08:06Because are you a human using an A.I. memory tool going to remember to mute in appropriate
00:08:15conversations?
00:08:18Like if you're in the middle of a fight and the fight just like pops up spontaneously
00:08:22because I don't know, maybe your cousin called you and wanted to hash out an argument that
00:08:27you've been having for three years randomly out of the blue one day.
00:08:29A thing that happened to me.
00:08:30I was going to say purely hypothetical.
00:08:31Purely hypothetical.
00:08:32Are you going to remember while you're screaming about like lifelong grudges you've held against
00:08:38each other to mute?
00:08:39No, you're not.
00:08:40You're just going to be in the moment having that fight.
00:08:43And then you're going to look at a transcript later and you're like, cool, I get to relive
00:08:46this tense moment because it's analyzing your your emotions through the tone of your voice
00:08:53and what's being said.
00:08:55I think one time I was like, Victoria had a tense conversation with her cousin who,
00:08:59and because my cousin has a Korean name, they got their name wrong.
00:09:02So it sounded like I was fighting with the Bee Gees, which is funny.
00:09:06So it was just like, do I want to relive that?
00:09:09No.
00:09:10Is having this transcript here helping me like be the bigger person and move on from
00:09:14this argument?
00:09:15Also no.
00:09:16So it's very strange.
00:09:19I often forgot to turn it off while I was in the office or commuting.
00:09:24That gave me useless to dos, told me to check on my patient in Louisiana who is at danger
00:09:29of harming someone.
00:09:30And I was like, well, that's not my patient.
00:09:32That's someone on this New Jersey transit bus, either watching a medical drama or discussing
00:09:40private patient things out in public, which is wild.
00:09:45Can't differentiate between TV shows and actual conversations.
00:09:49So, you know, forgetting to mute the Bee while watching Severance this season was funny.
00:09:56Can I read you one that I had?
00:09:57So I, out of curiosity, testing the thing, sat down and turned up the volume most of
00:10:04the way and just watched TikTok for like 15 minutes, was just like curious what the thing
00:10:08would do.
00:10:09I just want to read you what it came up with for TikTok.
00:10:14It goes, recounting a sleepless night with children, which was a TikTok that I watched,
00:10:18a confusing exchange, another confusing exchange, discussing headphones and a music lesson,
00:10:23random thoughts and observations, Spanish conversation and cooking, golf commentary,
00:10:27musical interlude, waiting for someone who isn't coming, playing a game and stopping
00:10:31a drug dealer.
00:10:32That's TikTok.
00:10:33Yeah.
00:10:34And so this is the thing that that was so crystallizing for me listening to this, because
00:10:40it's like the actual individual things it's pulling out, it actually does a reasonably
00:10:47good job with.
00:10:48Like if I just sit here and I play some audio for it or I have a conversation, at least
00:10:53the, I've used mostly the limitless pendant and the plod stuff.
00:10:57There was the thing you attach to the phone and then they had a pin that was kind of like
00:11:00another thing you wear on a lanyard.
00:11:02Both do a pretty good job of like summarizing information had in some kind of conversation.
00:11:08They have absolutely no idea what real life is like.
00:11:14And I like, this has made me realize how high the bar is for all of this.
00:11:18Like to your point about the, what, what is important?
00:11:22What is important that I want to remember?
00:11:24What is important that I would prefer never to think about again?
00:11:26What is not important at all?
00:11:28And you should just throw it away and understand that actually saving this at all is the wrong
00:11:32thing to do.
00:11:33Like the bar for this stuff, understanding how a person moves through their life is so
00:11:39high.
00:11:40And with all this AI stuff, none of it works in any kind of meaningful way until you do
00:11:44that.
00:11:45And that is tough.
00:11:47It's tough.
00:11:48And then as you, the person who has to review it, you feel emotionally and mentally gaslit
00:11:55a lot of times.
00:11:56Like you become kind of a fact checker of your own life and you look through these transcripts
00:12:01and like these summaries, like, cause B was very big on summaries of what happened to
00:12:05your day.
00:12:06And I'd be like, listen, I never told my boss Todd about my bowel movements.
00:12:09I would die before doing that HR complaint of a thing.
00:12:13But it said that I told, it said that my boss found on a public social media platform that
00:12:18I had vocalized my bowel movements and then we laughed about it in our one-on-one.
00:12:24And I was like, no, that is not what happened.
00:12:27That is not what happened whatsoever.
00:12:29I, you know, human memory is fallible, but I would, I would never do that.
00:12:36It doesn't understand humor.
00:12:37It doesn't understand riffing or things like that.
00:12:40It doesn't understand when you're listening to music.
00:12:43It doesn't.
00:12:45I watched several episodes of season four of Abbott Elementary and it gave me to do
00:12:48saying like, keep an eye out for the septa strike because it'll affect how your students
00:12:53get to class.
00:12:54Right.
00:12:55And I'm like, I'm not a public school teacher in Philadelphia.
00:12:58It can be useful.
00:12:59I thought I recorded some briefings as journalists do, and it was able to distill some facts
00:13:06like pricing, launch dates, the general idea and the concepts behind things.
00:13:11Got the product name totally wrong, but that's besides the point.
00:13:15So that's an interesting one for me though, because I think the, the pro case, I think
00:13:20the most generous case you can make for these devices right now is to use them as a kind
00:13:25of on purpose recorders, right?
00:13:27Like the, the limitless one that I'm wearing right now there, you can use it all day if
00:13:31you want to, but they mean it much more specifically as like, I'm sitting down to have a moment.
00:13:36I'm meeting with somebody and I turn it on and record it and I get summaries and action
00:13:39items.
00:13:41That's fine.
00:13:42And actually like AI is reasonably good at that kind of thing.
00:13:45If you're just like, here's 20 minutes of audio, pull out the most important stuff from
00:13:48it.
00:13:49A lot of models can do that fairly well.
00:13:50That stuff is so unbelievably commoditized already though.
00:13:54Like if I just gave you 20 minutes of audio and told you to pull out action items from
00:13:58it, using AI, there are a hundred places that could do that all to the same, like B minus
00:14:04level of quality.
00:14:06And the idea of this as a like dedicated device to me only makes sense if you're going to
00:14:13do it in the like always on sort of wearable kind of way.
00:14:17Like Abbott elementary is an interesting one where like one thing I would like for this
00:14:20to do for me is keep a running list of all the shows that I watch, right?
00:14:25Like every year I read Steven Soderbergh's blog post where he's just like, he just keeps
00:14:29a daily list of like, here's how many episodes of the show that I watched and here's the
00:14:32movie I watched tonight.
00:14:33And he like publishes at the end of the year.
00:14:35And I'm always like, that's so cool.
00:14:36I should keep that record.
00:14:37This should do it for me.
00:14:38It should be able to listen to the thing.
00:14:41Like this tech exists, Shazam exists.
00:14:43It should be able to tell me all the songs I heard, all the shows I watched, like it
00:14:47should give me a compendium of all the TikToks I laughed at today, right?
00:14:50Like that's the kind of stuff that you can actually make something out of.
00:14:54But none of these are within a country mile of pulling off anything like that.
00:14:59They're all like, you should know every word you said out loud because that's valuable.
00:15:03And like, maybe I'm just a boring person, but it's not like it just it's super isn't
00:15:09has been my experience.
00:15:10You kind of realize just the scope of all the conversations you're having in a day because
00:15:15one of the things that it doesn't do is make a note of your silent conversations, your
00:15:21text messages, your emails, your Slack messages.
00:15:25Obviously, Bee does have some integration with email as well, and we'll summarize the
00:15:31things that you need to do from your email, which could be useful.
00:15:34But there was one instance where it's like, oh, you got notified about a Park Mobile class
00:15:39action suit.
00:15:40Here's your ID number, went to search four separate email inboxes, could not find this
00:15:45email.
00:15:46And I'm like, did you hallucinate this email for me?
00:15:49It was a real class action lawsuit with a real deadline.
00:15:53I could not find this email in any of my inboxes.
00:15:57So I I literally had an existential meltdown and crisis with that.
00:16:02But also, it can't differentiate for like the I shop on Stylevana from to stockpile
00:16:08my Korean sunscare at Suncare.
00:16:11It always has always has a sale.
00:16:12And it's just like, oh, you got to take advantage of this last minute sale from Stylevana.
00:16:16It's like, no, I don't, because it'll be a new one in two days.
00:16:19Like it can't differentiate that as well because I can't really.
00:16:24The idea is is compelling because there's just a lot of junk in our lives and trying
00:16:29to sort and memorize what is and isn't important.
00:16:33That's a thing.
00:16:34But at the same time, it's just like I have a lot of really deep and meaningful conversations
00:16:40over text with friends who live far away.
00:16:43And those are not like jotted down anywhere in the app because it wasn't spoken aloud.
00:16:48So it's that whole thing of like if a tree falls in a forest and no one's there to hear
00:16:52it, did it happen?
00:16:53Well, if a conversation didn't happen out loud, does your AI memory even know it happened
00:16:57unless you grant it like integrations with messages?
00:17:01And then that feels icky.
00:17:02That feels no bueno.
00:17:04So it's it's it's a lot of I think the most interesting thing that people could learn
00:17:10from wearing these things is what you say in a day.
00:17:14Like what do you actually say aloud?
00:17:16And what is the value of keeping your mouth shut?
00:17:21That was like one of the things I learned.
00:17:22I was like, oh, you really don't have to say anything during your day.
00:17:28But then your transcript becomes very boring.
00:17:30Yeah. It's funny you find yourself talking less.
00:17:32I found myself talking more again.
00:17:34Like I became very aware of the fact that I had a microphone and I'm like, OK, I'm doing
00:17:38something that my recorder doesn't know about.
00:17:40So literally, I caught myself last night narrating myself playing video games because
00:17:46I like I wanted to communicate to this thing that I was playing video games for some
00:17:50reason, like as a record of some kind.
00:17:52I don't know. It was just like I'm doing something.
00:17:54And this thing that I'm trusting to know what's happening with me has absolutely no
00:17:58idea. I think humans in general just have this or at least people in media and people
00:18:02who've become journalists and writers have this like impulse to document what is
00:18:06happening. And it's not like I wasn't doing some degree of that.
00:18:09I'm an avid junk journaler or just a journaler in general.
00:18:13So I write every day like three or four bullet points of things that I find memorable
00:18:19that happened to me. And then comparing that with what Bea said was memorable and what
00:18:23happened to me was very interesting at times because when one thing that I noted that
00:18:29it just didn't pick up on was there was like a so we work we share offices with Eater
00:18:35and they were having a Paddington day and they had marmalade sandwiches and tea
00:18:40cookies in the office. And I really, really liked one of the cookies.
00:18:45It was like a transformational thought process.
00:18:48But I ate the cookie in silence. I had this thought in silence.
00:18:51There was no one around with me to like share that moment.
00:18:53So I just wrote it down in my little journal and I was like this strawberry shortcake
00:18:57cookie was phenomenal.
00:18:59And that's nowhere in that record.
00:19:01And that's just like kind of one small example of a thing.
00:19:04I still think about this cookie from time to time when I see other cookies.
00:19:08So that's how transformational this strawberry cookie was to me.
00:19:12But that was there's no record of that in my A.I.
00:19:15transcript of life. And it sometimes is it's a weird thing because you're navigating
00:19:21consent at all times if you're wearing this device.
00:19:25I have a bunch of friends who are like, oh, she's on her bullshit again.
00:19:28Yeah, whatever. Tell me what the A.I.
00:19:29says about me.
00:19:31But, you know, there are people who don't want that.
00:19:33Like my spouse is a very intentional tech Luddite to a degree.
00:19:37It keeps me sane and healthy.
00:19:39But, you know, there was that caused friction in my household, too, because he's just
00:19:44like, is it listening?
00:19:46Could you like turn it off?
00:19:47And so that's another thing that you have to learn to navigate, especially when you
00:19:52have tense conversations about like household finances or you both dogging on that one
00:19:59friend in your friend group who is making some questionable life decisions.
00:20:03Do you want to have a record of that?
00:20:05Like if you're a doctor, how does this relate to HIPAA?
00:20:08How does that is that even ethical for you to have, even though it might be a thing
00:20:13that you as the doctor want to have to think about certain things?
00:20:19I think these people were telling me like, oh, yeah, lawyers, professors, a lawyer
00:20:23should not be wearing this at all just because you are working on sensitive things.
00:20:28So even when we talk about it in a professional capacity is like, is it even ethical to do
00:20:34that?
00:20:35Do do we need things recorded all the time?
00:20:38The coolest thing about an analog recorder is that it's analog.
00:20:43The data is not going to be uploaded to a cloud server, per se.
00:20:47So it's just you start to have existential crises when you wear these for a long period
00:20:53of time and how people move.
00:20:56Because if you are using it ethically and responsibly, starting out a friend dinner
00:21:00going like, hey, guys, I'm actually testing this AI thing.
00:21:03Do you guys all consent?
00:21:05Are you guys all OK?
00:21:06I won't use it in this way.
00:21:07Ethics aside, it's also illegal in a lot of states like if you're in a two party consent
00:21:11state, you are technically committing crimes by doing this to people like it's this stuff
00:21:15is just messy.
00:21:16And I think to me, the big challenge here is this kind of on by default status that
00:21:22a lot of these want you to have.
00:21:23And I think, again, I'm actually pro the idea of AI voice recorders.
00:21:28I think like I've started using some of this stuff to like journal just by sort of spouting
00:21:34thoughts at the end of the day and have it sort of pull out an outline of my day for
00:21:37me.
00:21:38That's really great.
00:21:39I really like that.
00:21:40But the idea that I should just be able to walk around and capture my life all the time
00:21:45and that it will be useful, I think like I kind of agree with your spouse that like maybe
00:21:49there's something there, but it's it's nowhere near good enough to be worth all the other
00:21:55strangeness and tradeoffs.
00:21:57And I will say to to your point, the I had a very private, personal, long phone call
00:22:03last night and I was wearing this and forgot about it just while I had it.
00:22:07It had been on for hours.
00:22:08Right.
00:22:09Like I've just been testing this thing for a long time.
00:22:10It had been on.
00:22:11I wasn't even thinking about it.
00:22:12I go to open up the app this morning and the whole damn transcript to that phone call,
00:22:16or at least my half of it is just sitting there in this app, which means it got uploaded
00:22:20to somebody's servers.
00:22:21All these companies have very serious privacy policies.
00:22:24I believe them when they say they care about it.
00:22:25But like that private conversation is is by definition now no longer private.
00:22:31And I would be doing that to me.
00:22:32I'm doing that to the person I talked to on the phone who didn't know about this.
00:22:35I'm doing it to all of the strangers whose conversations I hear while I'm standing next
00:22:40to them on the subway.
00:22:41Like it's it changes the nature of what it is to like be a person in the world.
00:22:47And I think we've reckoned with this with cameras in a really interesting way.
00:22:50This is like that on a completely other level, because I can at least sort of see that you're
00:22:55pointing a camera at me like the Ray-Ban glasses and all that stuff makes it all really messy.
00:23:00But like I can probably see your camera.
00:23:03If it can see me, I can probably see it.
00:23:05These are different.
00:23:06And I like in ways that freak me out.
00:23:08These are way different.
00:23:09Like the conversations that I heard in the that I heard, not me, but this pin heard and
00:23:17then memorialized and decided were yours was just and were decided like I'm hearing snippets
00:23:24of conversations of our colleagues talking about videos on other verticals that I have
00:23:28no business knowing but have been recorded.
00:23:31It really just kind of it made me feel paranoid in some ways of just like you never know who's
00:23:38listening.
00:23:39And to a degree, we know that because like in the early days of the Internet, I don't
00:23:42know if you knew this site, but it was like called overheard in New York.
00:23:45And it's just like little conversations that people overhear and upload online.
00:23:51And that's fine in a way because it's anonymous.
00:23:53You're not describing who these people are, but it captures names that captures projects
00:23:58that like capture that two people somewhere in the Vox Media office.
00:24:02We're talking about big cats and sanctuaries on a day in February and that I had no business
00:24:06knowing this, but I overheard it and they don't know it.
00:24:09I know it.
00:24:10And you just sit there with that knowledge and how like immense it feels and how icky
00:24:16you feel having perpetrated it on someone that no amount of disclosure can really fully
00:24:22make you feel right again.
00:24:24Having done it, you feel like a spy, like it's a weird transition to these AI wearables
00:24:30like the smart glasses that can record the AI recorders that listen to everything.
00:24:35I'm getting a pair of the nuanced audio glasses that like enhance your hearing.
00:24:40And the lens crafters lady sat there and she's like, oh, so you're signing up for our latest
00:24:45spyware tool.
00:24:47You're going to hear everything.
00:24:49And I was like, what?
00:24:51Oh, oh God.
00:24:53Am I just on like the spy beat now?
00:24:58It's frightening on some degree because I consider myself a person who thinks about
00:25:04the ethics of it quite deeply, but I just know there are people who are not.
00:25:09Do you think there's a version of this thing that works for you?
00:25:11And I think like take the sort of we are journalists who record our interactions with people a
00:25:16lot out of it because that is like true and valuable, but kind of a niche use case, right?
00:25:22Like most people do not need to record that kind of thing all day, every day.
00:25:26But lots of people have meetings.
00:25:28Lots of people need to remember things from class, things like that.
00:25:32Is there a turn in kind of either the ethical approach or the sort of way that you use this
00:25:38thing that would make you feel better about them?
00:25:41I think if it could just listen to me, my voice only, and differentiate my voice only.
00:25:49And the thing that I was most attracted to for using this was the to-dos.
00:25:55I am constantly like, yep, got this, got that, going to get this back to you.
00:25:59And I write maybe 70% of them, which means 30% someone has to come back and go, did you
00:26:05do the thing?
00:26:06I'm such a forgetful person.
00:26:08But then it really just has to negotiate the knowledge and the human wisdom of knowing
00:26:13what's important to remember and what is very important to forget.
00:26:17Because forgetting is a survival mechanism as well, emotionally and for a lot of different
00:26:22things.
00:26:23And the AI is like, I will remember everything.
00:26:25And it's like, no, no, no, no.
00:26:27And I think before I started these tests, I would have guessed it's like 50-50 throughout
00:26:31the day, right?
00:26:32Stuff that's, even if it's just tuned to my own voice, stuff that's interesting and important,
00:26:36lots of that.
00:26:37Also stuff that is uninteresting and unimportant and it can throw it away.
00:26:40And actually I think it's like 5% important, 95% throw away.
00:26:46And I think the actual value of these things will come from proactively throwing away the
00:26:50correct 95%.
00:26:53It's just that that's a really hard technical problem.
00:26:55And so none of these things want to try making that assumption because if they get it wrong,
00:27:00it kills the whole value of the thing, right?
00:27:02So they seem very happy to just overshare everything all the time in service of having
00:27:08the one thing that you're looking for.
00:27:09And that just creates a lot of labor for you.
00:27:13Right.
00:27:14To get a little philosophical here, part of the human mission is trying to decide what
00:27:17is important in your own life.
00:27:19And we have dedicated millions of human hours to like, what is the right answer to that
00:27:25question?
00:27:26I think it's a little silly to expect that AI would be able to do that when emotionally
00:27:31it's not very intelligent at this point in time.
00:27:34Who knows if it will ever be, but it's like, I think we talk a lot about how AI is smart.
00:27:39We don't ever talk about whether AI is wise.
00:27:43So it's sort of like, I need AI to be a little wise if it's going to help me, because that's
00:27:49actually what I'm looking for.
00:27:50Maybe I'm not looking for an assistant.
00:27:52Maybe I'm looking for my own Mr. Miyagi to point me in the right direction.
00:27:56And an AI Mr. Miyagi is a lot harder than I think we're ready to acknowledge at this
00:28:04point in time.
00:28:05All right.
00:28:06I do.
00:28:07I just want to leave you with the best thing that has happened to me, and then we're going
00:28:08to get out of here.
00:28:09So this morning, my two-year-old son has been up since three o'clock this morning.
00:28:13So I just left the pendant on all morning.
00:28:16We're hanging out.
00:28:17And here is just a chunk of our day, as described by the Limitless app.
00:28:23Going to the playground and observing a garbage truck, suggesting a trip to the playground,
00:28:27pointing out a garbage truck, expressing awe, identifying a garbage truck again, discussing
00:28:31going to the playground, identifying a garbage truck, which is an unbelievably good description
00:28:37of my morning.
00:28:38And it's like, there's a little bit of that that I really enjoyed.
00:28:42It can play some of the audio of him pointing at a garbage truck and yelling garbage truck.
00:28:46And that's nice, but just give me that.
00:28:49Forget all the rest of this stuff.
00:28:50Just give me my son saying garbage truck over and over, and I will pay for this thing, and
00:28:54I'll be happy about it.
00:28:55It's the best experience I've had with any AI product ever, is just the thing that shows
00:29:02me the recording of my son yelling at garbage truck.
00:29:05Garbage truck.
00:29:06Yeah.
00:29:07It's the dream.
00:29:08There you go.
00:29:09That's all we need AI to do.
00:29:10That's all we need.
00:29:11All right.
00:29:12V, thank you as always.
00:29:13Thanks for having me.
00:29:14All right.
00:29:15We're going to take a quick break and then we're going to come back and we're going to
00:29:17talk about keyboards.
00:29:18We're going to talk a lot about keyboards.
00:29:24Support for the Verge cast comes from Upwork.
00:29:27Finding good freelancers used to be a real pain.
00:29:29They were out there, but they were hard to get in touch with.
00:29:32You might have to use an expensive agency, and even then the talent pool was small and
00:29:36it was hard to know what kind of work you were going to get.
00:29:39Now the process is simple with Upwork.
00:29:42With Upwork, you can access a global marketplace filled with top talent in IT, web development,
00:29:48AI, design, admin support, marketing, and more.
00:29:52Posting a job on Upwork is easy.
00:29:54There's no cost to join.
00:29:55Once you register it, you can browse freelancer profiles, get help drafting a job post, or
00:30:00even book a consultation.
00:30:02From there, you can connect with the right freelancers for your business.
00:30:06Upwork makes the entire process easier and simpler, all with industry low fees.
00:30:11Post a job today and hire tomorrow with Upwork.
00:30:14Visit Upwork.com right now and post your job for free.
00:30:18That is Upwork.com to post your job for free and connect with top talent ready to help
00:30:23your business grow.
00:30:25That's U-P-W-O-R-K dot com, Upwork dot com.
00:30:30All right, we're back.
00:30:35Before we get into the next segment, let me just explain to you a little bit about the
00:30:39keyboard that we're about to talk about.
00:30:42It's called the Seneca.
00:30:43It's made by a guy named Ryan Norbauer, who we're about to talk to and you're going to
00:30:46hear a lot from.
00:30:47This keyboard is $3,600.
00:30:50And the only way I can explain it is it's this sort of giant, heavy monolith of a mechanical
00:30:58keyboard.
00:30:59It has gray keys around the sides.
00:31:02And for the numbers, it has a couple of these like really lovely sort of salmon colored
00:31:06accent keys.
00:31:08The thing weighs, I think, seven pounds.
00:31:10You can buy it in a bunch of different finishes.
00:31:13It has this huge sort of funky metal casing.
00:31:17I mean, it's a truly lovely keyboard.
00:31:22And it is the product of this guy, Ryan Norbauer, building himself the perfect keyboard.
00:31:28You're going to hear a lot about switches, which are essentially the actual mechanism
00:31:32that moves the key up and down.
00:31:34You're going to hear a lot about keycaps.
00:31:35You're going to hear a lot about the force that gets applied and the brass plate underneath.
00:31:39All of these are like the sort of component parts of a keyboard that most people never
00:31:44think about.
00:31:45But they're actually individual pieces that you can buy and put together and tune and
00:31:50reshape and rethink and you can invent your own.
00:31:53And they're made by different companies.
00:31:55There's different ideas about how all this stuff works.
00:31:57The keyboard world, frankly, is much bigger and much wider than I ever realized.
00:32:01And so Ryan's story is just what it looks like to go through it and find the exact right
00:32:07thing for you, no matter what, no matter how complicated, no matter how expensive.
00:32:13And I think it's really cool.
00:32:14And it was really fun to talk to him about not only what this keyboard is, but why it
00:32:18is this way and why it was worth the work to make a $3,600 keyboard.
00:32:23I don't think I'll be buying this keyboard, but I love that it exists and I really enjoyed
00:32:27talking about it.
00:32:28So you're going to hear from Nathan Edwards on our team and you're going to hear from
00:32:31Ryan Norbauer.
00:32:32Let's get into it.
00:32:33Nathan Edwards.
00:32:34Welcome back to the Birchcast.
00:32:35Hi, David.
00:32:36Hey, buddy.
00:32:37Ryan Norbauer.
00:32:38Welcome.
00:32:39Thanks so much for inviting me.
00:32:40We are here to talk about very expensive keyboards, and I'm very excited about this.
00:32:43So Ryan, a thing you should know, you've known Nathan a long time, Nathan's keyboard guy.
00:32:48I am, I would say I'm like keyboard curious.
00:32:51I know nothing.
00:32:52I'm on, I'm on a keyboard switches for beginners website right now that I was using to prepare
00:32:57for this podcast.
00:32:59I know nothing, but I am fascinated.
00:33:01And so we, this, this is, this is the energy we are all bringing to this and I'm very excited
00:33:05to talk about it.
00:33:06Welcome to the fold.
00:33:07I'm happy to report that you don't have to start with a $3,600 keyboard.
00:33:11There are many excellent options now, especially since COVID, the options available have really
00:33:15transformed and it's never been a better time to get started as a keyboard enthusiast.
00:33:20For sure.
00:33:21So Ryan, I think mostly what I want to do is, is sort of talk about your story and your,
00:33:24your descent into, I don't know, something like madness, trying to build this keyboard.
00:33:29I'll go along with that.
00:33:30But I kind of want to start at the beginning.
00:33:31You've been working on keyboards for a long time.
00:33:34Why keyboards?
00:33:35How did you get sort of into this world in the first place?
00:33:36I feel like once you're in the keyboard world, it just pulls you deeper and deeper and deeper.
00:33:41But most people seem to have a story about kind of why they got into keyboards in the
00:33:43first place.
00:33:44What's yours?
00:33:45So, I mean, there's two phases to that, I guess.
00:33:47There's why keyboards as an object, and then there's why keyboards as like a profession
00:33:52and community.
00:33:53And those were actually, you know, different for me.
00:33:55My actual enthusiasm about keyboards per se, as objects and devices, stretches back
00:34:01to like the fifth grade, when I was first learning to program on an Apple IIe, learning
00:34:07Apple Basic in my schoolroom, and was very excited about the future of technology.
00:34:15And you know, this was in sort of the late 80s, early 90s period, running into, you know,
00:34:21when I was in high school, into the beginning of the internet.
00:34:25And there was just this immense culture of optimism around the future in general, and
00:34:31specifically how computing would transform our relationship with other humans on this
00:34:36planet, and sort of interconnect us and make us, you know, as individuals more capable
00:34:40of doing more interesting things.
00:34:43And for whatever reason, that idea, like really captured my imagination as a kid.
00:34:49And so just the memory of my first experience typing on an Apple IIe or an IBM Buckling
00:34:57Spring keyboard, and just this period of extreme excitement when I was a kid about computing,
00:35:03correlated with a time when computing hardware was rarer and treated as more valuable.
00:35:09And therefore, you know, manufacturers could invest more in making them actually satisfying
00:35:13and durable.
00:35:14And so they were not commodity products.
00:35:16And so they had a certain sound and feel to them that I associate with that kind of
00:35:20techno excitement, you know.
00:35:22And so that's the original spark.
00:35:25And I think for most of us who have weird, sort of enduring obsessions as adults, they
00:35:31usually spark in childhood.
00:35:33But it wasn't until probably the, you know, maybe 2009, 2010, when I discovered that there
00:35:40was a whole online subculture of people who are really into this too, right?
00:35:46And most of us who, you know, came to it at that time, and I think to some extent still,
00:35:52had that same spark about keyboards having that vintage quality.
00:35:56And it's why we're sort of drawn to what are, you know, somewhat vaguely described as mechanical
00:36:02keyboards.
00:36:03Like it's a kind of slippery and maybe not entirely accurate term, especially in the
00:36:06case of my keyboard, because it's hybrid capacitive and mechanical.
00:36:09But nevertheless, the term came to be called mechanical keyboards, but I think a more accurate
00:36:13term would be like vintage feeling or vintage style keyboards.
00:36:16And so, you know, I discovered this community of intense enthusiasm on the internet.
00:36:20There was a website, still is, called Geek Hack, which still looks like it was programmed
00:36:24in 1998, if you check it out.
00:36:27But it's a very, like, positive and interesting community of people, especially in those early
00:36:32days.
00:36:33And it was very small too, it was like, I don't know, maybe 1000 people on Geek Hack
00:36:37or something.
00:36:39Maybe Nathan has researched this and could give us actual figures, but, you know, it
00:36:42was very, very small, not that long ago.
00:36:45And now, you know, I go to keyboard meetups, like the Seattle Keyboard Meetup is over 1000
00:36:49people, I think, the last time I went to it.
00:36:51And there's just like, the one in San Diego, I think also is in the four figures for sure.
00:36:57And so you can just regularly find people all around the world who know about this.
00:37:01But like, even 10 or 15 years ago, this was an extremely obscure, weird thing that nobody
00:37:06knew anything about.
00:37:07And it was mostly people, with the exception of people building their own keyboards out
00:37:11of Cherryamic switches and a few vendors, it was mostly people restoring old keyboards
00:37:16like the IBM Buckling Spring is this sort of legendary keyboard in this space and whatever.
00:37:21So anyway, I found these people, I realized that these people are my people in a very
00:37:26weird way.
00:37:27And I got really kind of deeply involved in that community.
00:37:30So where does the Seneca story start in all of that?
00:37:32Because you were you were making housings and sort of building around building your
00:37:37own full keyboard here for a while.
00:37:39But then, like, at what point are you like, it is time to make my own creation?
00:37:45I didn't intend for this to even be a business in the first place, I kind of stumbled into
00:37:49it because I was making my own keyboard housing for myself.
00:37:52You know, I took a machining class at the Artisans Asylum in Boston, and I learned how
00:37:56to do milling of metal on an old Bridgeport mill from World War Two, right.
00:38:01And so this is not a scalable process.
00:38:04I just literally, you know, I had been learning this, because another one of my hobbies is
00:38:07making replicas of props from science fiction, particularly the Star Trek.
00:38:11And I want I like I took these classes in order to learn how to like, I want to make
00:38:14Dr. Crusher's laser scalpel, you know, and then, and I did all those things.
00:38:18And I realized, hey, actually, maybe I could make one of these keyboard housings that I
00:38:22like so much now that I know a little bit of metalwork.
00:38:25So the very first one, I prototyped on that mill, totally manual, no CNC.
00:38:31And I'm like, oh, okay, I guess this actually works.
00:38:35But the surface finishes I was able to get out of the machine were pretty poor, mostly
00:38:38because of my inexperience as a machinist.
00:38:40And so I wanted to get the same design made, which I had proved by prototyping at a factory
00:38:48that knows how to actually machine things and then bead blast them to a uniform surface
00:38:51finish and then anodize them.
00:38:53And so I did what people did and still do on Geek Hack all the time, which is I started
00:38:59a group buy, which is, you know, in order to buy one thing for yourself, you order 10
00:39:05so that you can kind of split the costs among other people.
00:39:08And so the first one was just one literally made for myself.
00:39:11And then the second one was a very small batch that I did on Geek Hack.
00:39:16And for whatever reason, that just kind of like caught on and people kept asking me to
00:39:20do it again and again and again.
00:39:23And then I realized that there was much more of an appetite for this kind of thing than
00:39:27I thought.
00:39:29And so I started using the resources from each subsequent run to do more creatively
00:39:34interesting things.
00:39:35And so an obvious conclusion from that is to move from just making aftermarket housings,
00:39:41which is pretty limited in scope, because it's a very sort of DIY project.
00:39:45It's a big ask of people to buy a keyboard from Japan and wait six months for it.
00:39:49Sometimes they're not even available, void the warranty, take it apart, put it inside.
00:39:54Only so many people on earth are going to be willing to do that.
00:39:57So the obvious idea for if you're interested in creative possibilities is like make your
00:40:02own keyboard, the whole thing, because then you can control every aspect of the experience.
00:40:08And so I knew, I think, pretty early on, maybe the first few years, that that would be the
00:40:11fun and interesting thing to do.
00:40:13But I knew that there were a lot of skills and abilities that I had to acquire in order
00:40:17to get to the point of being able to do that.
00:40:18And so I had various intermediate projects of increasing technical ambition that sort
00:40:22of got me to that place.
00:40:24So sometimes I vacillate between whether this was a project that took me 10 years to do
00:40:29or five years.
00:40:30So 10 years to learn everything and to get to this point, but I think I only started
00:40:33actively working on the Seneca as a named project, what became this keyboard about five
00:40:40or six years ago.
00:40:41Nathan, you've mentioned Ryan's name to me years ago.
00:40:45When did you first become aware of this man and his many goings-on?
00:40:50Were you a heavy grail user?
00:40:53I still, to this day, actually do not own any Norbauer housings.
00:40:57I knew about, I think he first came onto my radar pretty early when you did these Galaxy
00:41:04Class keycaps, which ended up being officially licensed eventually.
00:41:08But I think I heard about them back when they were a little bit bootleg.
00:41:11Well, they weren't bootleg, so this was a Star Trek-inspired key set.
00:41:14They had no IP on them.
00:41:16Yes.
00:41:17That's correct, yeah.
00:41:18They were just like vaguely inspired by the graphic design user interfaces of Star Trek
00:41:21the next generation.
00:41:23They were Star Trek-ish.
00:41:24Exactly.
00:41:25Yeah, I got you.
00:41:26Exactly.
00:41:27I was actually a columnist for Tested for a couple of years back in the day.
00:41:29Oh, cool.
00:41:30And then I worked with Ryan at Wirecutter.
00:41:31It doesn't matter.
00:41:32Anyway, I found out about that.
00:41:33Then you made a housing called the Heavy 6 for the Leopold FC-660C, which is to this
00:41:40day the best keyboard known to man, in my humble opinion.
00:41:44This is mine.
00:41:47And so Ryan made a housing, he put it up for a group buy.
00:41:53I saw it on Geek Hack during the interest check, and then I was like, okay.
00:41:57There was a week where in a fit of mania, I ordered a Heavy 6, I ordered a keycap set
00:42:03called GMK Space Cadet for different kind of keycaps, and I ordered a minivan, which
00:42:07is a 40% keyboard.
00:42:09Much too small for me.
00:42:10And then like three days later, I had to go cancel all three of those things because I'm
00:42:14like, I don't need the keycaps and I don't want my keyboard to be five pounds.
00:42:19But in the process of emailing Ryan to be like, hey, can I actually, it was very embarrassing,
00:42:25can I actually cancel my order for this beautiful thing that you've made that I realized I don't
00:42:29actually want?
00:42:31Because I specifically, I used to throw this thing in my back, it was a keycap, I used
00:42:36to throw it in my backpack to take it on planes.
00:42:37Like if I was traveling for more than a couple days for work, I'd take my keyboard with me.
00:42:41And I can't do that if it weighs five pounds and it's going to like break your laptop.
00:42:48So I emailed him to cancel.
00:42:49It was very nice about it.
00:42:50We exchanged like five or six emails back then, and then talked about Star Trek, talked
00:42:55about keyboards.
00:42:57He mentioned actually starting this project to make a ready-to-type keyboard.
00:43:02And then like six years later, when I heard about the Seneca, I just shamelessly neck
00:43:08rode.
00:43:09I just replied to that thread from my now work email address.
00:43:14I'm like, hey, remember me?
00:43:16Can I check out your keyboard?
00:43:17Let's talk about those six years.
00:43:19Part of this process that I'm fascinated by is like, it's very clear that you went in
00:43:22and were like, I'm going to build the best keyboard, like end of sentence, right?
00:43:27Is that a fair characterization?
00:43:29Like you weren't thinking the best keyboard for whatever price, the best keyboard with
00:43:32whatever specs.
00:43:33You're like, I am going to build myself the best keyboard.
00:43:35Is that where it started?
00:43:36Yeah, absolutely.
00:43:37But there's an important caveat there or an acknowledgement that we must make, which is
00:43:40that the definition of the best keyboard is enormously subjective, right?
00:43:44And so it's my definition of the best keyboard.
00:43:48And I've rigorously held to that, but with the acknowledgement that other people will
00:43:51probably not agree about some of those choices, but that other people certainly, and the sort
00:43:58of people who I think were inclined to like my old products would also agree, okay, that's
00:44:04pretty much what I would have done if I had to go from scratch and engineer everything.
00:44:09So it's subjective.
00:44:11That thing is the sort of subjectivity of it is what I'm really interested in, because
00:44:16How much of that could you have sort of sat down and made a list of on that first day
00:44:21where you're like, here is what I know I want out of this keyboard that is still true today
00:44:26of the Seneca that you've finished?
00:44:28Like, did you know what the perfect keyboard for you was when you set out to build this
00:44:33thing?
00:44:34That's an interesting question.
00:44:35Nobody's asked me that before.
00:44:37And I think it's interesting because I think my answer is yes, I did know exactly where
00:44:41I wanted to end up.
00:44:42The only thing I didn't know is how I would get there and how incredibly circuitous and
00:44:47challenging the path to get to that place would be.
00:44:51I wanted something that was electrocapacitive and had this like very vintage sound to it,
00:44:56but was an acoustic improvement on anything that was out there.
00:44:58I wanted something that had a more tighter on-axis key feel relative to existing capacitive
00:45:05keyboards out there and that really like hyperengineered the acoustic and tactile nuances
00:45:12of the board and also just had this kind of retrofuturist visual aesthetic that viscerally
00:45:17appeals to me.
00:45:19That's probably what I'm sure that's what I would have told you at the time.
00:45:22And that's I think that's what we actually are shipping right now.
00:45:25Okay.
00:45:26And this is kind of a question for both of you.
00:45:27And this is where I start to reveal how little I know about keyboards.
00:45:31How much of that stuff is subjective versus objective?
00:45:34Talking about the sort of tactility and the sound and the way it feels to use, are those
00:45:40numbers you can write down and dial in or is that just something you're going to sit
00:45:45there and touch a hundred keys and one of them feels right and you go, it's that one?
00:45:50Like Nathan, you spend all your time with keyboards.
00:45:52How objective can you be about this stuff?
00:45:55There are people who will publish force curve graphs and they'll be like, oh, well this
00:46:01switch has this amount of, you know, this actuation force, this amount of return.
00:46:06People try to quantify it.
00:46:09I think for me and for a lot of people, it's just about how it feels and how it looks and
00:46:13how it sounds.
00:46:14Like, does it feel good?
00:46:16Does it sound good?
00:46:17That is subjective.
00:46:18Objective is, is the spacebar rattly?
00:46:21Is there a crunch?
00:46:23Does the key cap hit the housing of the switch before it finishes its travel?
00:46:30Is there, like, do you, when you press it down, do you hear, like, do you hear it when
00:46:39the key smacks against the PCB or the stabilizer smacks against the PCB?
00:46:43Do you hear upstroke clack when you let go as the slider slams into the housing?
00:46:49Whether or not you like those things or care about those things is subjective.
00:46:54Whether they happen is objective.
00:46:56I think, you know, it's like any human thing.
00:46:59It falls sort of along a normal distribution.
00:47:01And there is a kind of objectivity, especially in, like Nathan said, the things that sound
00:47:07bad, right?
00:47:08And people tend to agree more on things that they don't like rather than the things that
00:47:11they proactively do.
00:47:13And so there's, as with so many things in life, most of the gains in improving something
00:47:19come not from, like, adding something great, but removing bad things, right?
00:47:23And so that was sort of the obvious initial low-hanging fruit of this project, is find
00:47:27all of the things that people generally agree sound bad and make sure there are no possible
00:47:32sources of that in this keyboard, right?
00:47:35And so you actually, it's not that objectivity doesn't exist, in a sense, in the keyboard
00:47:42world.
00:47:43And I certainly notice that, especially if you talk to people who aren't, like, who don't
00:47:46go to meetups all the time and who aren't on the Mechanical Keyboard subreddit all day.
00:47:50If I just lay a Seneca in front of a random person and I, like, I have one where the switch
00:47:55plate is aluminum and one where the switch plate is brass, everyone, literally everyone
00:47:59will say they prefer the brass plate, which is why ours has a brass plate.
00:48:02Because I actually personally sort of, in some ways, like the more noisy aluminum one
00:48:06because it's more, like, retro nostalgic for me.
00:48:08But everybody's like, oh, that's obviously better.
00:48:11That is a kind of objectivity.
00:48:12But the, you know, a lot of that is just removing something that people perceive as unpleasant,
00:48:20like this sort of pinging-ness.
00:48:21I think one general trend you could describe that's pretty common or uniform across individuals
00:48:27is a preference for deeper pitches.
00:48:30People don't like high-pitched, pingy, medley-sounding things, or, like, rattly, like, thin-walled
00:48:37plastic sounds, right?
00:48:41That's probably a good way to reduce the quest of the keyboard community for the past two
00:48:46decades is to move away from those things, which were sort of the default of, like, a
00:48:50Dell keyboard you might buy in 2005, right?
00:48:55And so there is a kind of objectivity, but then once you get into the nuance of, like,
00:48:58what switch type, what switch curve, things then become immensely subjective.
00:49:03And people do try to characterize them with things like force curves, but, you know, it's
00:49:08like, you can make a waveform of the Beatles and Bach and look at it and be like, yeah,
00:49:13they're objectively different, but what do you do with that graph, right?
00:49:16Like, you'd have to hear it and be like, oh, I like that or I don't like it.
00:49:19And I just make keyboards for people who like electrocapacitive, which has a very distinctive
00:49:24sort of thocky, clacky, vintage sound signature, just because that's what I like.
00:49:29Did you have a sort of specific reference thing in mind?
00:49:34You mentioned some of those vintage keyboards.
00:49:36Did you have one in mind that you're like, this is the gold standard of one part of this
00:49:40or another?
00:49:41Or was it purely sort of going by what you like and what feels right to you as you're
00:49:45building through it?
00:49:46I think in many ways, I always come back to my vague impressions of what the old Apple
00:49:53Alps keyboards were, because that's what I really, the keyboard that I had the first
00:49:59emotional or computer that I had the first emotional relationship with, you might say.
00:50:04But those are really not, they're not as good as a keyboard switch can be.
00:50:10When I first discovered that there was this community called Geek Hack around keyboard
00:50:16customization around 2010 or whatever, I got very excited because most of it was around
00:50:21Cherry MX type switches, which are certainly the most prevalent and well known.
00:50:25Probably if your listeners are sort of casually familiar with keyboards, almost all of them
00:50:30would have been Cherry MX, right?
00:50:32And so I got very excited about that.
00:50:33I got into them.
00:50:34I tried every single switch that was available at the time, including sourcing them from
00:50:37Asia and getting people to swap the springs and the sliders and do all sorts of crazy
00:50:42stuff.
00:50:43But somehow I just stumbled into keyboards from this Japanese company called Topra, which
00:50:48were in no way made for enthusiasts.
00:50:49They were made for industrial applications to be highly durable and to sort of survive
00:50:55mini actuations without changing their performance, right?
00:51:01And they just happen, I think, to be accidentally really good.
00:51:04If you like that kind of feeling that I associate with those vintage Apple keyboards, even though
00:51:09they're not the same.
00:51:11And one thing I've seen you say a few times is that part of the process here was about
00:51:16constantly like forced to choose between the thing that is more like, you know, efficient
00:51:23or economical or whatever.
00:51:25And the thing that feels right, you just kept choosing the thing that felt right.
00:51:30And part of me wonders if you just intentionally kept choosing the thing that was more expensive
00:51:35And I want to talk about that because there is a thing here and like I was reading your
00:51:39blog and stuff and I think you seem to have this real appreciation for sort of things
00:51:42that are like deliberately luxurious.
00:51:45And so I think, like, I guess this is a long way of asking, would you like this keyboard
00:51:49as much if it cost you a third of the price to make?
00:51:52But also, like, what did those tradeoffs look like?
00:51:55Like, give me an example of something where you're faced with, OK, I could do this the
00:51:59easy way or and the economical way or the straightforward way, or I could do this the
00:52:04way that's going to take longer and cost more and be harder.
00:52:06And you just have to, like, bite the bullet and go that way.
00:52:08Yeah.
00:52:09So that's a very deep and nuanced question.
00:52:11I have lots of thoughts about it.
00:52:12So I'm happy to blabber about it for a while.
00:52:13It is very commonly the case that people, when they pay a lot for something, will rationalize
00:52:19that it's great, right?
00:52:22And that's there's obviously that danger in luxury goods.
00:52:26And I'm a very uncomfortable relationship even with calling my product a luxury product
00:52:33in a way, because much of what advertises itself as the luxury industry these days is
00:52:39very much not.
00:52:40It's just not my good definition of what luxury should be.
00:52:45It's capitalizing on that phenomenon of simply, like, let's start with the price and work
00:52:50backwards towards the perception of value, right?
00:52:53And let's sort of sell it as a status object, and then figure out how to make it as cheaply
00:52:59as conceivably possible.
00:53:00And you know, there's this French business theorist named Jean-Noël Kafka, and I really
00:53:06like his book called The Luxury Strategy, and he kind of analyzes the industry as, like,
00:53:12luxury as a way of facilitating the low-volume manufacturing of creatively interesting goods,
00:53:16right?
00:53:17It's a business model for low volume, for doing things that most people wouldn't want
00:53:22to buy.
00:53:24And that I find actually very compelling, as I said, in some ways this business grew
00:53:30up accidentally and it was just sort of for fun.
00:53:33So I think it's a viable business model and business strategy, and it's a nice – therefore
00:53:41I can, like, plausibly rationalize doing what I wanted to do anyway, which was to make the
00:53:46cool an interesting thing, which oftentimes correlates with being more expensive, but
00:53:51it's certainly not so for its own sake.
00:53:54There are advantages to selling things that are expensive, I will say.
00:53:58So one thing, for example, is it just, like, it sort of filters out people who just don't
00:54:03care, right?
00:54:04And who are going to just, like, whine about dumb shit, or about, you know – it sort
00:54:08of filters out a lot of the internet entitlement, like, because people were just not going to
00:54:13buy that thing.
00:54:14Like, it's – the people who are seriously engaged with us about buying our stuff, like,
00:54:18they actually really think keyboards are cool and are very excited about what we're doing.
00:54:22Because otherwise, why would you even be considering spending so much money on a keyboard?
00:54:26So Nathan, I'm going to make you do a very uncomfortable thing, which is explain a product
00:54:30to the person who made the product.
00:54:32But I think this will be fun.
00:54:33So I, like, what is it about the Seneca that makes it interesting, other than the fact
00:54:38that it is just a keyboard that costs $3,600?
00:54:41Which I think, to Ryan's point, is not sufficiently interesting in and of itself.
00:54:45It has to be interesting anyway, and then we can talk about the price.
00:54:47So what is it?
00:54:48The thing that I think is most interesting about the Seneca, as a keyboard nerd myself,
00:54:55is the switches and the stabilizers.
00:54:57Like, the fact that, like, Ryan just went out and developed your own Topra-style switch
00:55:07that is better than Topra in measurable ways, and developed your own stabilizers, and you're
00:55:12like, I'm just going to fire the cash bazooka at the stabilizer problem until it's solved.
00:55:20That to me is the interesting thing about the Seneca.
00:55:22It is a fantastic keyboard to type on.
00:55:24Like it is, it's a joy to use.
00:55:29It is, but the things about the Seneca, you've said, is like, this is the keyboard for you.
00:55:36It's tenkeyless, it's a tenkeyless layout, you know, it's a flat slab.
00:55:41It's not the layout for everybody, it's the layout for you.
00:55:43The finish is great.
00:55:44It looks like stone.
00:55:45It feels amazing.
00:55:48That part is, that's like a straight line from where you started.
00:55:52To just like, be like, hey, here's my keyboard.
00:55:54By the way, I invented new switches, I invented a new stabilizer, and those things are good
00:56:00is the punchline.
00:56:01Like, they're very good.
00:56:02That's what's so cool about the Seneca to me.
00:56:05Full disclosure, I cannot afford the Seneca.
00:56:07I will not be buying a Seneca.
00:56:09I think it's lovely.
00:56:11You said it weighs seven pounds, right?
00:56:12That's just, that's the one thing I want.
00:56:13Any keyboard that weighs seven pounds, I think is inherently interesting.
00:56:17It's so heavy.
00:56:18So I happen to have one here.
00:56:19Most of the weight comes from the brass switch plate, actually.
00:56:23And that is, it's actually, you know, the keyboard is maybe a little bit heavier than
00:56:28I would have chosen it to be.
00:56:30It's just that the plate is brass for acoustic reasons.
00:56:34I would, again, if we had made it out of aluminum, the keyboard overall would have been much
00:56:38less heavy.
00:56:40But you know, I think there's an interesting analogy to draw here from what Nathan was
00:56:46saying, which is, my mental model, I think, for what I do in keyboards is very much like
00:56:52mechanical watches, like, you know, not in terms of the actual mechanical details necessarily,
00:56:59but the sort of social, commercial, cultural role that they play is that, you know, everybody
00:57:08has a computer in their pocket that will tell them the time.
00:57:11A mechanical watch is completely superfluous in 2025.
00:57:15And yet there's this enormous, you know, industry and culture of people who are very excited
00:57:20about and derive enormous pleasure from mechanical watches.
00:57:23If you guys ever, you know, the website Hodinkee, it's like an entire publication dedicated
00:57:28to daily publishing, extreme deep dives into this industry and how nerdy and how excited
00:57:34people can get about it.
00:57:35And there are people who, yes, absolutely, very wealthy people who are super into it,
00:57:39but also just very nerdy people who are super into it.
00:57:43And all for a thing that is totally unnecessary and just a way of, you know, adding a little
00:57:49bit of joy and satisfaction and something to be excited about in our otherwise boring
00:57:54daily computer-based lives, right?
00:57:57And I think that actually is pretty analogous, like in no way am I pretending that this solves
00:58:02an essential or important problem other than, you know, the, I suppose, actually deeply
00:58:08essential and important problem of like having something to give a shit about in life for
00:58:12a little while, right?
00:58:13Yeah, I was going to say, I don't think any of that is this is like, it's, it's, there's
00:58:17a sort of utilitarian superfluousness to it, but there is also something like deeply essential
00:58:23about like, I have this thing on my watch and it means something to me that is like
00:58:27powerful stuff.
00:58:29Yeah, so I acknowledge being at both ends of that spectrum and I think that's just okay.
00:58:35Because you know, as Nathan pointed out, like this is an optional purchase.
00:58:40It's not compulsory.
00:58:41No one is required to buy it.
00:58:42The stuff that's happening in the like low end of premium keyboards is really interesting
00:58:46right now.
00:58:47And it's never been a better time to be getting into keyboards and having that be a more accessible.
00:58:52I'm extremely in favor of.
00:58:53I'm just doing another thing and I think it's okay for there to be a diversity of things
00:58:57in the world.
00:58:58Does this feel like the end of a project or the beginning of one?
00:59:01It's both for sure.
00:59:02I mean, if I, if I'm fortunate, uh, then it's both, uh, if it's just the end, well, that's
00:59:09fine.
00:59:10And I'm happy with what I made and that's, you know, maybe, uh, people aren't interested
00:59:14in it and that's fine.
00:59:15But so far, you know, we've already done the, uh, first edition offering, uh, and it's almost
00:59:19entirely sold out.
00:59:20So, uh, it's probably the beginning of something.
00:59:23I do have a lot of interesting ambitions for what we could do after this keyboard.
00:59:29And some of that does involve, uh, catering to tastes that aren't necessarily exactly
00:59:34my own.
00:59:35So I think, you know, in terms of the subjective sound signature and tactility of the switch
00:59:39that will remain, but as we've discussed, there are people who prefer different layouts
00:59:43of keyboards, uh, smaller, especially more compact keyboards.
00:59:46Like Nathan wants to be able to have a keyboard he could put in his backpack and take with
00:59:50him while traveling, but still have it be a huge upgrade over a laptop keyboard.
00:59:53Right.
00:59:54And I, you know, even in LA here, I see people with, who bring mechanical keyboards to cafes
01:00:00and put it on top of their laptop because they want a more satisfying experience.
01:00:03I would like to enable that.
01:00:05And so in some ways to make a keyboard that is the opposite of the Seneca, the Seneca
01:00:08is this sort of, you know, desk monolith that weighs seven pounds and is optimized for a
01:00:14certain sound signature that I really like.
01:00:17It would be interesting for me to go all the way in the other direction to make one that's
01:00:20instead of more like deep and thocky, a little bit more, what might be described as clacky,
01:00:25where you have a more resonant housing that's made up out of some type of polymer, um, like
01:00:29a CNC machine, polycarbonate, or, um, we're, I'm actually really interested in forged carbon
01:00:35fiber.
01:00:36If we can figure out some way to do that in a, uh, economically viable way where you have
01:00:40a housing that is actually really high aesthetic quality, but it's light and durable.
01:00:47And, uh, that would also involve injection molding a switch plate, like I described on
01:00:51the heavy grill so that it, it would also impart an entirely different sound signature
01:00:55to the board by giving it, uh, like a polycarbonate plate.
01:01:00Right.
01:01:01And we just know from existing keyboards in the enthusiast world that that radically changes
01:01:04the sound signature, radically changes the sound signature in a way that some people
01:01:08really like, and some people don't like.
01:01:11Um, and, but it would be a way of answering those desires that are a little bit differently
01:01:15tuned from my own desires.
01:01:17And I find that interesting because they're hiding around all of those corners.
01:01:20I know it's a bunch of weird and interesting technical challenges, which tends to be the
01:01:24thing that actually gets me excited.
01:01:26Wouldn't it be fun to go the other way and be like, what is the cheapest possible pretty
01:01:30good thing I could do?
01:01:31Like I'm going to make a $49 mechanical keyboard.
01:01:34Norb by Norbauer.
01:01:35No, no.
01:01:36The trouble is everyone has that so covered, right?
01:01:39Like this is the, there's been this, especially since the COVID era when everybody was doing
01:01:43work from home, there's just been this massive surge of people into that space.
01:01:47Like there was a time when I was thinking about doing stuff like that, but at this point,
01:01:51everybody, like everyone can do that much better than I can.
01:01:53That's just not my specialty, uh, is, you know, operating at scale and that kind of
01:01:58optimization.
01:01:59So I think I'll probably just keep on doing my thing.
01:02:01Uh, and I'm, I'm okay with that.
01:02:03So in short, every, everybody else has that well covered.
01:02:06Which puts you in a fun place.
01:02:07You get to go, you get to go do the weird stuff.
01:02:09It's great.
01:02:10Uh, all right, Ryan, Nathan, we've kept you long enough.
01:02:13Thank you both for doing this.
01:02:14This was really, really fun.
01:02:15I really appreciate it.
01:02:16Thanks for asking.
01:02:17All right.
01:02:18We got to take one more break and then we're going to come back and take a question from
01:02:20the Vergecast hotline.
01:02:21Be right back.
01:02:28All right, we're back.
01:02:29Let's do a question from the Vergecast hotline.
01:02:31As always, the number is 866-VERGE11.
01:02:33You can email Vergecast at theverge.com.
01:02:36Please keep sending us party speakers.
01:02:37We love the party speakers, but keep reaching out about anything and everything.
01:02:41There's actually a lot of really complicated news going on and even just getting questions
01:02:45from all of you has been really instructive on how we should be talking about and covering
01:02:49this stuff.
01:02:50So please know that when you call or you email, even if we don't explicitly reference it on
01:02:56the show and we try to do lots of them here on the show, it matters and it makes the show
01:03:00better all the time.
01:03:01And we are incredibly grateful to everybody who reaches out.
01:03:04This week, I want to play two questions that both have kind of the same answer.
01:03:09And I'll explain that in a second.
01:03:11Let me just play them really quick.
01:03:12Here we go.
01:03:13Hey, I'm Dave.
01:03:14I have a question.
01:03:16I'm a teacher and my students use Google Chromebook and on them they use Google Docs and Google
01:03:25Drive and Google Sheets and they submit assignments in Google Classroom and there are probably
01:03:33millions of other students in the world that do this.
01:03:38So if Google gets forced to split from Chrome, what happens to students, Chromebook, Google
01:03:48Apps for Education and that whole ecosystem?
01:03:53That's it.
01:03:54That's my question.
01:03:56Okay, so we have a question about what happens to Chromebooks if Google is forced to sell
01:04:00Chrome as the result of this search antitrust trial?
01:04:03Hold on to that one.
01:04:04We also have a question from Jason.
01:04:05Hey, this is Jason from Nova Scotia.
01:04:08Forgot to leave my name last time and I thought of something else to add.
01:04:10From the recent podcast, you guys were talking about Chrome.
01:04:12I'm glad you guys mentioned Chromium, but I did have one thought and I still don't really
01:04:17know how it would go, even if they're able to pry Chromium away from Google.
01:04:23And unless they explicitly stated in the remedies that they weren't allowed to make
01:04:27another browser, I wonder what would stop Google from just making a second browser.
01:04:32Because I think you could make the argument that what people like about Chrome isn't actually
01:04:37just that it's good, it's that it's part of Google.
01:04:40And I would think that none of the passwords or bookmarks or all the stuff that people
01:04:44have with their Google account would just go with it.
01:04:48So what would stop people from making Chrome 2, this time more blue, electric boogaloo,
01:04:54or God forbid, the biggest sideswipe, forking Firefox and then making another browser.
01:04:59But that, yes, yes, I know.
01:05:01Yeah, let me know your thoughts, because unless they explicitly stated that, I would just
01:05:04think they'd make another browser and maybe they even already have.
01:05:08Who knows?
01:05:09Anyways, thanks.
01:05:10Bye.
01:05:11Okay, so there's a simple answer and a complicated answer to these two questions.
01:05:15The simple answer in both cases is that Google would have to get out of the browser business.
01:05:21I think the DOJ, what it has asked for specifically, would prevent Google from making another browser.
01:05:28It's that simple, right?
01:05:29And so what that looks like is Google just wouldn't own a browser.
01:05:34Chrome would be its own thing.
01:05:36Somebody would buy Chrome.
01:05:37A name I keep hearing floated is OpenAI as the company with the most interest and money
01:05:42to pull this off.
01:05:44But let's say OpenAI owns Chrome.
01:05:46All of that goes to Chrome.
01:05:48The Chromium project, I think, is slightly more complicated because it is technically
01:05:53an open source thing.
01:05:54So it could either be administered by a new company that cares about it in the way that
01:05:59Google has cared about it, or it could completely languish.
01:06:03Or best case scenario, it becomes a truly open source project.
01:06:07The fact that it is an open source project essentially run by Google means, for all intents
01:06:12and purposes, it's not an open source project.
01:06:14It's a Google project that anyone can download.
01:06:17And I've heard from a lot of people who make browsers that actually the idea of it being
01:06:20a more equitable thing, that everyone agrees this is what we want to build browsers on,
01:06:25but we can't let it be run by one company, that could be a good thing.
01:06:29That would be something more along the lines of what the Matter Consortium is doing, right?
01:06:33Where they say, this is a group of companies who agree that this is a thing we need to
01:06:36build together, and it'll benefit everybody.
01:06:39Could be great.
01:06:40The way that it has worked until now is that Google has done the vast majority of the work.
01:06:44We talked a little bit about this on Friday's show, but Sundar Pichai, the CEO of Google,
01:06:48said during the trial that Google, I think, is responsible for 94% of the code commits
01:06:54over the last year to Chromium.
01:06:56It does most of the work.
01:06:57It has most of the people.
01:06:58It provides most of the funding.
01:07:00It is very much a Google project.
01:07:02So I think Google would not be allowed to use it to build another browser.
01:07:06I think the question of whether Google would be allowed to be part of building Chromium
01:07:10is separate.
01:07:11But again, if Google can't build a browser, why would it continue to invest in Chromium?
01:07:15Would OpenAI or whoever else invest in Chromium, who knows?
01:07:19The Chrome OS piece of it is, in a certain way, just as simple.
01:07:23Google would have to get rid of Chrome OS.
01:07:25Chrome OS is Chrome.
01:07:27It is part of Chrome.
01:07:28There is Chrome tech inside of it.
01:07:31And if Google were to get rid of that, it would lose it.
01:07:33Whether it had to officially sell Chrome OS and the Chromebook business to somebody,
01:07:38if Google can't do Chrome, it can't do Chrome OS.
01:07:40It's pretty much just that simple.
01:07:41And so that is another thing that would essentially get thrown out with this deal.
01:07:48This is why this is complicated, right?
01:07:50Right after Sundar Pichai testified, there was an expert who basically came in and said,
01:07:54I don't even know that it is technically feasible to divest Chrome from Google.
01:08:01And it's for all the reasons that these two callers are describing.
01:08:04What Chrome holds is not just a window to look at a web page, right?
01:08:11We think about it as you open Chrome and it's an interface to Google search.
01:08:14And it is that.
01:08:15But it also holds a ton of data about your bookmarks and your passwords.
01:08:20And it has a lot of history.
01:08:22And it has an entire operating system built around it that millions of students use all
01:08:27over the place.
01:08:28Chrome is baked into the fabric of Google.
01:08:32And the fabric of Google is baked into Chrome in many more ways than I realized.
01:08:36I mean, think about Android.
01:08:38Chrome is the default browser on Android.
01:08:40And what would essentially happen is that there would be no default browser on Android.
01:08:45Like by the letter of the law here, Google would not be able to make a browser that came
01:08:50bundled with Android.
01:08:52That just doesn't make any sense.
01:08:54Like your phone should come with a web browser.
01:08:56How are you going to download a web browser?
01:08:58It just sort of boggles the mind.
01:08:59So what Google has argued is that a Chrome is not a piece of the search business and
01:09:06that fundamentally this trial is about the search business.
01:09:09I don't find that nearly as compelling as the other part of the argument, which is Google
01:09:13says we can't divest Chrome because it is all over Google.
01:09:18It would be like saying, get rid of, I don't know, HTML inside of Google.
01:09:23Like it just doesn't work.
01:09:24It's not a thing you can do.
01:09:26And increasingly, that is a thing that is coming up from a bunch of different sides
01:09:30in the trial that it's not as simple as Google makes a web browser, take that web browser
01:09:35and give it to somebody else, sell it to somebody else for $20 billion or whatever the number
01:09:38turns out to be.
01:09:40That actually Chrome and its many offshoots, again, which is Chromium, which powers lots
01:09:45of other browsers.
01:09:46It's Chrome OS, it's Chrome on Android, it's Chrome on desktop, it's Chrome everywhere.
01:09:51It would just change the nature of how Google works.
01:09:55And all of Google's products and a lot of the underlying tech inside of Chrome powers
01:10:00other things.
01:10:02Chromium is hugely important, not just to other browsers, but across the web.
01:10:05And you could make the case that a lot of things kind of fall apart as soon as Chrome
01:10:11does.
01:10:12That is obviously Google's case.
01:10:13And I think so far, just from sitting in the courtroom, I think the judge, Amit Mehta,
01:10:17finds that fairly compelling.
01:10:19But we'll see.
01:10:20It's possible that Google is going to have to sell Chrome.
01:10:22We've seen there are a lot of companies lining up to buy it for obvious reasons.
01:10:26If you got Chrome and the Chromebook line in the deal, that becomes pretty interesting.
01:10:32You buy not just the most important browser, but one of the most important operating systems.
01:10:37That becomes really meaningful.
01:10:38But then Chromebooks without the access to the rest of Google, probably less valuable.
01:10:46This is why this is all so messy and so intertwined.
01:10:49This is a thing we're seeing in all of these trials that it's not just as simple as a company
01:10:54has several parts, you cleave off one of the parts and give it to somebody else.
01:10:57All this stuff is so intermingled and intertwined that figuring out how to pull it apart is
01:11:03in many ways the real challenge of the antitrust fight.
01:11:08We'll see.
01:11:09There's lots still to come there.
01:11:10And I would not rule out the possibility that Google is going to have to figure out how
01:11:13to do that pulling apart.
01:11:15But it's going to be really complicated and it's going to have vastly bigger ramifications
01:11:21than just when you open a new tab in Chrome, it shows open AI and not Google.
01:11:27We'll see.
01:11:28Lots and lots left to do there.
01:11:29All right.
01:11:30That is it for the Vergecast today.
01:11:31Thank you to everybody who came on the show and thank you as always for listening.
01:11:34As ever, if you have questions, thoughts, feelings, if you want to buy Chrome, if you've
01:11:38seen a party speaker, send us an email, vergecasttotheverge.com or call the hotline 866-VERGE11.
01:11:44We really, truly love hearing from you.
01:11:46This show is produced by Will Poore, Eric Gomez, and Brandon Kiefer.
01:11:49The Vergecast is a Verge production, part of the Vox Media Podcast Network.
01:11:52Neal and I will be back on Friday, presumably to talk about more antitrust stuff.
01:11:56We're getting a lot of new information about what's going on with the App Store, lots of
01:11:59new stuff in the meta and Google trials, hopefully some other gadget news to talk about because
01:12:05boy have we talked a lot about courthouses the last couple of weeks.
01:12:09Hopefully more to come.
01:12:10We'll see you then.
01:12:11Rock and roll.

Recommended