Transcript

00:00:00 - Speaker 1: This ends up becoming a question about file standards more than it does about application functionality. I can take a notion document and fairly easily translate that into a text file, a very linear document format. There’s currently not really a file format for spatial canvas. Right now there’s just not a good way for Muse to talk to another spatial canvas app.

00:00:30 - Speaker 2: Hello and welcome to Meta Muse. Muse is a tool for deep work on iPad and Mac, but this podcast isn’t about Muse the product, it’s about the small team and the big ideas behind it. I’m Adam Wiggins here with my colleague Mark McGrenigan.

00:00:45 - Speaker 2: Hey, Adam, and another colleague, Adam Wulf.

00:00:48 - Speaker 1: Hey everyone.

00:00:49 - Speaker 2: And Wulf, I understand you recently spent some time in prison.

00:00:53 - Speaker 1: That’s true, and it’s so good to be out with everyone again. I, of course, was not technically in prison, well, I guess I was.

I volunteer with a program called the Prison Entrepreneurship Program. They actually help felons who are near release go through and kind of 3 month entrepreneurship school. They do character development and then learn about starting their own business, and really help them get their feet under their ground again, find their sea legs when they get released.

So they do a lot with employment. And housing and support, and a lot of education on the inside of the walls, and a lot of support on the outside of the walls and family reunification, and it’s really just a wonderful program that helps. Inmates helps their families, and ends up helping society.

And so the, the big number that matters is national recidivism, which is the number of people who get released from prison and go back into prison, is extremely high. It’s somewhere around like 40 or 50% end up going back into prison.

And graduates of this program, it’s as low as like 5 or 7%, and so it just has a dramatic effect for these men and for their families. And so it’s been really fun to volunteer with the past. Gosh, probably 5 years, something like that.

So, yeah, if anyone ever wants to be locked in prison with me, then give me a shout out. We’ll make it happen.

00:02:18 - Speaker 2: Sounds like a really worthy program. Well, we’ll link them in the show notes. I’m definitely a believer that how a society treats the people that need to be removed for sort of justice reasons and what you do when they’ve, you know, fulfilled their debt to society, as the saying goes, and how you make a transition back to normal life says a lot about it. It seems like a really great program you’re involved in there.

So we can jump straight into our topic today, which is listener questions. This is our second mailbag episode. Mark, you and I did one year and change back, and I think it was quite a lot of fun to go through all the questions people submitted. And now feels like a good time to me just because we’re 2 months or so out from the launch of our 2.0 product and the dust is still settling, but in many ways we spent a lot of the last two months just answering questions through all channels Twitter. Hacker news, but most especially through our support channel, that’s hello@museapp.com and the in-app thing. So we’re in a question answering mood and we have a lot of common questions that we thought would be good to kind of address on air as well as we put out a call on Twitter for folks to submit questions. We’ve got lots of really interesting ones, more than we’ll have time to answer, so we’ll do our best to get to as many of them as we can. You fellows ready? Let’s do it.

Yeah. So I guess we’ll start with roadmap just because that tends to be the biggest or most numerous questions are in that category, what features we building and when, but we can go from there and how people use Muse or how we use Muse, things about the broader ecosystem, tools for thought, as well as more even broader than that, some things about how our team works and even some things about Ink & Switch. So the nuts and bolts of roadmap doesn’t work for you. You can jump forward a little bit and things will get more far ranging. But yeah, starting at the beginning here, so I think a broad question many people ask, but here I’ll quote from Penny Chase who basically just said, I’d like a glimpse into the Muse roadmap, and we answered that question mark, I think a year, year and a half ago, and I think it included, you know, going local for sync and going multi-device and having desktop apps. So check, has that roadmap changed other than what we’ve accomplished since then.

00:04:30 - Speaker 3: Well, the good news is that with this launch, I think we’ve got a lot of validation on the direction that we’ve been going since day one, really, which is this idea of a tool for helping you have better ideas that spans your iPad, Mac, iPhone, eventually the web, and is very rich.

And what we’ve heard from our users, I think, is, yes, and let’s see the rest of it. And just to give a few buckets there, I think one building on the local first sync, you have the phone, that’s a pretty obvious gap for us right now, a more complete phone app, both in terms of making the phone a better tool for getting things into and out of your corpus, and also being able to look up stuff on the go.

You can also see something with the web, sharing on the web or even a full blown web client. and also building on sync. We talked during our tech episode how that’s really the foundation for collaboration, both synchronous and asynchronous, so I wouldn’t be surprised if we see something there. And then back in the what we call like app features, there’s some pretty standard stuff that I think we’re missing. Better search, linking, these are things that people ask for very often.

Another one I would say is more rich tech support. That’s a really foundational content site, and I could see other content types like videos and better support for free ride web pages and so on.

And then there’s a few more things I think we need in terms of like organizing and managing your content, stuff like non-spatial collections we’ve had on our list for a long time, a better inbox, which might be a variant of that. But those are, I think the main buckets on the horizon and then I still have on my medium to long term was this idea of end user programming or like scriptability, programmability more generally. I think that’s incredibly powerful and we have a really compelling foundation for that. I think we need a little bit more on the core app first.

00:06:11 - Speaker 2: Yeah, one way I’ve summarized it to myself and others is Muse 1 was a multimedia thinking canvas for iPad only. Muse 2 goes multi-device.

We’ve got the desktop and iPad, the local first syncing between them, phone, I think is going to be part of the 2. X series, at least that’s my hope, kind of it’s part of that vision.

It’s everywhere you need it, and then Muse 3 is where you get into the sharing collaboration, making it more than.

It will always be first and foremost a personal tool as something that is better connected to the outside world, both other humans, but also other tools through integrations and things like that.

And then yeah, maybe the end user programming feels like the, I don’t know if that’s Muse 4 or just kind of the finishing move that ties it all together.

00:06:58 - Speaker 1: In some ways I see use one as use the teenager and Muse 2 as used the college graduate, and now we have all of the education and skills and life foundation to be able to bring Muse to this great career beyond in collaboration and teams and end user programming and text formatting and we finally entered adulthood, I think, ready to go out into the workforce and make a difference in the world?

00:07:23 - Speaker 2: Absolutely. Now on some specific features folks have asked about, a big one here is Zoom, being able to zoom in closer on some things, usually images or PDFs, and being able to zoom out further, especially boards as your boards get bigger and more complex, and a representative tweet on this is from our friend Marsen Igna, who says, where can I read more about why Muse has no zoom out for bird’s eye view of my board? Yeah, it clashes with the navigation gesture, but I’d love to work at 50 to 75% zoom.

00:08:01 - Speaker 1: When I think of Zoom, I think of solving at least two different problems.

The first problem is on a very large board, I want to zoom out and still be able to work and move selections around to kind of reorganize a very large desk or very large workspace.

The second problem is I’m working on a very large board, and I want to quickly jump to a different location. Scrolling is currently just kind of wandering around in the wild, and I’d really like to be able to see a map and quickly go from the bottom left to the top right or the top middle.

So I think whenever we do build Zoom, we need to think about it in that kind of context of which problem are we trying to solve? Do we find a solution that maybe can solve both problems? It’s not obvious to me which of those is the most important to start with, or the right lens to look at the feature through.

00:08:55 - Speaker 3: Yeah, Zoom is a very subtle challenge in Muse, just mechanically speaking, say you want to zoom out, OK, you pinch to zoom out on a board. How do we know that that’s different from wanting to zoom out to the next level up? That’s a problem we can solve, right? And you can have a quasi mode to toggle between the two, or you can have a detent in there somewhere or something. It’s just it’s quite subtle.

Another challenge with Zoom that we’ve known about since our research in the lab is with these freeform canvases, especially if you have Zoom and especially if you have infiniteness, you have this real risk of becoming lost. Like you’re looking at this solid off-white thing and it looks the same regardless of where you are and how far you are zoomed in, and people just get totally discombobulated. And so we’re trying to push back against that a little bit.

And this isn’t just a challenge for boards. We’ve also got a lot of requests for zooming into PDFs and images. So there’s a lot of stuff going on. You got zoom in, you got zoom out, you got navigation versus the document, you got different from document types, got temporary versus permanent. There’s a lot of stuff to figure out. So the answer is we just gotta sit down and do it. I think it’s very doable. It’s gonna take some time and some design work, and maybe some cutting of the Gordian not as Adam would say, you know, just get the 80% in there. I think we’ll get it done eventually.

00:10:05 - Speaker 2: We actually even discussed that as a potential thing to work on post launch. We were looking for things that would be more smaller projects, quick wins, crowd pleasers, just things to refresh our palettes after working so long on this big massive release with big data migration and so on.

And actually the conclusion we came to is it was too big of a project because we do want to think about all that stuff holistically and even if we do just carve off a small piece of it to do first, just doing a kind of boring and obvious way to do it like there’s a zoom level drop down or something like that we think will quickly create the problems that I think make a lot of other software not that enjoyable to use, which is disorientation and so on, and it’s particularly bad in the infinite canvas setting, but yeah, we need it, we badly need it as people’s boards get more sophisticated, as people are bringing in more different types of PDFs and images as they just want to do more things.

Yeah, you need the ability to zoom and we’ll solve that hopefully much sooner than later.

00:11:03 - Speaker 1: I think Zoom is also related to accessibility and potentially text size and text formatting, which we’ll talk about in a little bit. But tech sizes for some users can feel a bit small in use, and being able to zoom into a board or into an image in PDF mark, as you mentioned, is also related in some ways to allowing for custom text sizes or larger headers or text formatting, just generally being able to. See content no matter what your screen size happens to be, if you’re on a small iPad or a large iPad or a big screen.

00:11:37 - Speaker 3: Yeah, then we’re getting into the real Pandora’s box on the implementation side of when do you rasterize this content. If you only have one zoom level, you have a lot of flexibility and kind of do it whenever you want. But if you’re zooming all over the place, you either need to rasterize later or essentially suffer the effects of rescaling, so it’s quite gnarly.

00:11:55 - Speaker 2: A related one that’s challenging technically is dark mode. Again, another one we’ve gotten many, many requests for over the.

Well, years now, and one of my favorite stories actually, actually one of the best ways to ask for features is to tell a story, as we would say in our podcast episode about storytelling, attaching something to a story makes it much more memorable. So one story I remember well was someone writing that they were on an international flight, you know, overnight flight, 10 hours, 12 hours or something, and they spent the whole session basically with Muse and they had a bunch of PDFs loaded up and deep reading and deep thinking, perfect opportunity for that, right? But they’re in this darkened cabin. They turn their screen brightness down, but it’s just still too bright and they’re kind of afraid of waking the other passengers and so on.

00:12:44 - Speaker 2: Oh, that’s such a vivid image just blasting the cabins, but many folks have asked for it for similar reasons or just because they like dark mode.

But because we have the zooming interface that uses what we call internally snapshots, which are basically those thumbnails you see of the boards that give you this, you have this scaling transition, this sense of seamless zoom and traveling around in this kind of open world where you’re never loading a document or what have you, and that’s very nice, but the snapshot rendering. actually pretty CPU intensive.

You see this also, if you log in to a new device like an iPad with an account that has a lot of data in it and it downloads it all, the downloading it may take a while, but then generating all the snapshots actually will also cause your device to be pretty busy for a while if you have a lot of boards and deeply nested boards and so on. And so dark mode has the problem that, OK, now we basically need to generate new snapshots for every single board and do we keep both of them all, you know, do we make two all the time and slow down every regeneration every time you change something for a feature that maybe most people won’t ever use. Or is it when you switch modes, do we turn on the regeneration and suddenly your iPad is heating up and you grinding the CPU for 5 minutes while it tries to re-render everything and maybe you just want to check it quickly and then you switch it back and now it’s grinding again for another 5 minutes, so totally solvable, but I guess it’s, as with many things because we have this unique zooming interface and nested boards, that’s something that sets me apart, makes it unique and pretty special to my mind. But it also can make what seem like basic features much harder.

00:14:17 - Speaker 1: I use Zoho mail for my male client and their dark mode. It was an interesting choice because they need to be able to support dark mode for attachments that anyone can send and so it’s not only their own interface, but they’re trying to create a dark mode for the content of the email itself and what they settled on was for those attachments to literally invert the colors, which was functional but a bit jarring.

00:14:51 - Speaker 2: When greens turn into purples and everything else, and so there’s er colors can be really funky, especially like a picture of a face, for example.

00:14:56 - Speaker 1: Yes, exactly, yeah, photos or anything else, and so I’m not sure what their heuristic was for when to invert the image and when to lead the image in plain color. There were certainly some surprises when I first turned that on.

00:15:12 - Speaker 2: Another frequent request is search. A representative question here was asked by Josh Job, who asks, Are there plans for search inside the app, i.e., text? I worry I won’t be able to find anything if I can’t search the contents of blocks if I go big on muse.

00:15:28 - Speaker 1: I look at search As a navigation feature, almost more than anything else.

Back in the very olden days of the internet, there was the Yahoo directory, and you just manually search down into homeownership, lawn care, to find the nearest nursery to get some plants, whereas now you just go to Google and you say find me cheap plants.

I think the same thing for M where a lot of navigation is pinch in, pitch out, pinch in, pinch out, trying to find. Where you want to go and muse and search. Has this ability to jump very quickly from one side of your tree to the other side of your tree and back, and Supporting the content, both PDF content, text block content, and really making as rich of a search as possible. I think it is really gonna open up new workflows inside of Muse, and something I’m particularly excited about.

00:16:26 - Speaker 3: Yeah, I would emphasize that I think search is really important for navigation as well as the more obvious use case of finding stuff, especially on the desktop where you have a keyboard.

So if I want to go to my to do list and muse, it should be like command PTO enter all in one flow, 100 milliseconds or so and I’m there. And by the way, search should be. Much faster and better in use because of local first data.

A huge issue that you have with search on most apps, traditional SAS apps that they got to go to whatever Virginia and, you know, query the database and then come back. And then when you load the actual page, you gotta go all the way back again and get the data. This should all be able to happen locally within a few tens of milliseconds. So there’s really no excuse for searching out to be awesome and you just need to spend a little bit of time on it.

00:17:10 - Speaker 2: Yeah, definitely slop that one as it’s important, we definitely want it.

Everyone asks for it. Now it’s a matter of when does it, you know, when does it bubble up on the priority list of all the stuff we want to do.

I will go ahead and reference something you and I have talked about before, Mark, just because it was, I think pretty foundational.

We started Us, which is a book called The Science of Managing Our Digital Stuff where they do a pretty thorough academic survey of basically how people use computers over the last 20 to 30 years.

And looking at files and search through things like Google and searching on your local computer and so forth, and there’s a lot of predictions of either tagging or search or other things kind of killing the file system and well if you kind of maybe hinted in that direction, which is we don’t need a Yahoo style directory of the internet anymore because we have Google.

But I think the total sum of human knowledge of Google or Wikipedia or something is a little bit in a different category from my own stuff and spatial navigation and particularly the sense that things live in a place.

This was one of the takeaways from the research in that book as well, which was things like symbolic links, for example, outside of file references where you basically can have a pointer.

To file in multiple places, the vast majority of even relatively power users didn’t use these or didn’t use them much because it actually is very powerful for our brains, spatial memory to say I know where this lives. It’s 3 folders deep. I get to it from here.

And certainly I think that can be heavily supplemented by things like search or tagging or other ways to get at your data and providing more approaches is good, but we felt like the foundation of the spatial navigation is actually going back to something that has proven to just work incredibly well for personal data sets.

00:19:02 - Speaker 3: Yeah, I do think the main use case of search is going to be, I basically know this thing exists and probably even know where it is. I just don’t want to traverse the whole mind palace. I want to teleport with my search command right to where I know what it is. There will be some cases where you are actually doing a legit keyword search and just I’m not sure what, if anything, you have in your corpus about that. Yeah, the navigation is really important.

00:19:26 - Speaker 1: Using search to jump into a document, feels very natural to me, and I’m switching context and starting a new task, and then once I’m in that task, I think it’s more natural, uh, Wiggins, as you mentioned, to walk around the mine palace and just pinch in, pinch out and stay in that context because A search result in jumping through search feels very jarring sometimes to switch context so suddenly inside of the app, and I don’t know, we’ll leave it there because it’s not fully baked in my mind.

00:20:03 - Speaker 3: I do think there’s a very interesting question around what is the flavor of search.

The default way to do it is that search is a feature. It’s like this little box up in the corner and you type things and then a medium sized box appears with links and you click on a link and then you go back and tap.

We’ve long had this idea, again, going all the way back to the lab. of search that’s more integrated into the flow and capabilities of the app. So just to give you an idea, you can imagine, I’m not promising to do this, but you can imagine you have a search view, which is like a card and can be saved and persisted and moved and shared and resized within your corpus. It’s like a magic card. You type in the top of the card, and as you’re typing, the view changes to present different subsets of your corpus, which you can then teleport through to go to those rooms, if you will. But then you can do stuff like save this search, you know, I want to save all my long chair references on a card so I can go back to them later. That’s pretty cool. And you can also start to imagine. All the things you’re building around search, like the results for you as being a capability that’s general to the rest of the app. Things like seeing a view over your data is something you want elsewhere. So for example, we’ve talked about the idea of an archive. That should just be like a different view. It’s like search is the lead equals true and you might have some sugar in the app around that to make it easier and more standard to access that. But I like the idea of searches as building powerful generalized views versus a very specialized feature and wanting to get that right is I think why we’ve Delete a little bit on implementing it.

00:21:36 - Speaker 2: Daniel Rivera writes, uses my new go to for brainstorming, love how simple it is for quick ideas and sketches on the iPad, but I do wish it had inking for desktop. Is that coming?

00:21:48 - Speaker 1: I would love for it too. There’s no technical reason we can’t have ink on the desktop.

All the code is obviously shared between the iPad and the Mac.

I think the big question is, what does that input mechanism look like? Is there a way to Use the trackpad for ink input. Is that something where the the mouse is generally very bad at organic ink lettering if it’s anything beyond just a simple line or a simple arrow. Another thing we’ve talked about is using Wacom tablets and stylus tablets that I think could be very promising, but is obviously fairly niche use case for a subset of users, but I think it would be a very powerful way to bring ink into the Mac.

00:22:36 - Speaker 2: You know, one interesting thing is we already support that, not kind of on purpose, but we’ve had a number of folks write in to say, hey, I see that I can ink on Mac with my Wacom tablet, but I can’t, I don’t know what, get out the inkwer or you know, it’s sort of like partial because it just so happens that the input events that it sends is the same as the Apple pencil and we respond to that.

So we’ve done nothing explicit to support it and it’s kind of doesn’t work great. You could imagine if we had a lot of demand there, we would put the effort in to make that work.

Yeah I agree drawing with a mouse is just the worst freeform drawing even with a trackpad.

What I do think we might want is first of all, just highlighting, which could be highlighting of text, highlighting of cards. Craft does a really great job at this. Just a very simple way to add colors. Colors can give you context or they just brighten up your document, which is nice. But I think you can also look at something like a diagramming tool where you’re drawing, say, arrows between elements or you’re making a connection between them like an OO. So I think those things would make sense on the desktop, the idea of just being able to kind of circle something or try to handwrite something really hard for me to imagine there’s anything we can do there that would have a feel that would be acceptable from our perspective.

00:23:50 - Speaker 1: I’m glad you brought up the mind mapping because that’s, I think the most common scenario for me on the Mac. is wanting to draw just a very quick line between two cards, or a very quick box around a couple of cards.

It’s a feature Leonard has been experimenting with some various designs to start connecting cards and create a bit more of a mind map and I think there is a way that we can create some of these features in a mouse friendly and Mac friendly way.

That’s also balanced with not trying to create too many tiny little features or expand Muse into a grab bag of a billion features. I think the simplicity of Muse is extremely important, and so balancing what each platform is able to do and what each platform is designed to do, is a very delicate balance between supporting literally every use case on every platform. And optimizing for specific use cases on specific platforms.

00:24:46 - Speaker 2: If it starts to feel like a technical drawing tool with a huge toolbar and lots of options, and let’s see, I just want to draw an arrow between two things. Do I use the loose, you know, wire-like connection? Do I use the structured arrow? Do I use ink and draw an actual arrow? And if you find yourself making choices like that in the moment, I think that’s really gets in the way of your thinking process. So yeah, it has to be balanced against our kind of overall mission and design values.

00:25:15 - Speaker 3: Also part of the thinking here is about inking on the desktop as in Mac and inking on the desktop as in your office desk.

And part of the original idea with Muse was that you have these different complimentary devices, and they’re not only used at disjoint times, you might actually use them at the same time.

I was actually just rummaging in my archives from like 5 years ago yesterday, and I was looking at one of the original presentations I did about what would eventually become used, and we had this idea of you have A desktop like a Mac, and then right in front of you, basically where your keyboard is, you have an iPad and you’re using your Mac for the big heads up display and seeing all these PDFs and so on.

And then when you want to do a little drawing that you insert into your presentation, for example, you just draw that on your iPad because that’s the device that’s better for that.

So, that’s not to say we shouldn’t have some sort of inking capability in the map, but I would also just point out that there’s this possibility of running two devices at the same time.

We might need to add a little bit more kind of smoothness to make that even better, but that’s a possibility as well.

00:26:17 - Speaker 2: I do that with some frequency, actually, I’ll kind of convert my desk chair from the sitting upright mode, good posture, you know, 90 degree knees, feet flat on the ground, which I’ll use when I’m typing and using the mouse, and I’ll kind of convert it to the leaning way back mode, stick my feet up on the desk and have kind of my iPad and my pencil in my lap, but I still have muse up on the big monitor. I can see it live updating as I’m going and often see a bigger view, just have a much bigger monitor of the board I’m moving around.

But yeah, if we wanted to put more effort into supporting that simultaneous use case, one for me that would take a lot of friction out would be having the iPad be on the same, you know, when I open it, it’s on the same board that my desktop is on, but maybe not everyone wants that feature. Maybe you actually want to keep them on different ones, so I don’t know that would need some thought.

00:27:05 - Speaker 3: Well, maybe you should be able to make a search card that’s recently seen boards and put that in your inbox.

00:27:12 - Speaker 2: Smart. So another category of feature request is web sharing, and we’ve got a few examples here, including InterPlato ask or board shareable and navigable on the web with a shared link, and then Nikita, who asks, I wish to have notion like level of control on which canvas and which depth I want to share a link. My biggest barrier to using Muse is that I want to share my research as I do it.

00:27:37 - Speaker 1: This is something we’ve experimented with pre-sync that was very useful for a number of users, and I think post-sync we’re finally in a place where we can start thinking about this again.

One of the limits in the presync era was the web share was fully static, and there was just no ability for anyone reading it to add comments or certainly to add content, whereas now post-sync.

We can imagine a web viewer that is a full client of Muse that is able to interact with and sync content back up into your corpus, and so you can share feedback, get comments and content from your team, and have that go straight back into your muse. And so I think that’s a powerful team and collaboration feature that we’re certainly thinking about that sync has finally given us the foundation to support.

00:28:32 - Speaker 2: I’ll note from that earlier very brief beta test we did where we just gave it to a few folks, the ability to kind of share a pretty static, basically they kind of dumped your muse board to Kind of like static image, static HTML page was not the lack of interactivity or not even being able to add comments, but actually just that it wasn’t live.

So I think one of the things we get with the modern web conception of real-time collaboration, Google Docs, figMotion, etc. is that a URL becomes a place where you know you’re always seeing the latest version of something. So we do quite a bit of news board sharing internally. And actually a lot of times folks send us news boards either as, you know, PNG exports or full board bundles with feedback on the app or other comments, which we love. So it’s a great format for sharing work in progress and thinking with your colleagues, but it never fails, right? You go to share it, you know, I’ll just post it in a slack as an upload, which is a little clunky, but whatever, and then I realized I need to change three things that I left out or that are wrong. And I’m going back and deleting it from Slack and hoping that, you know, no one’s downloaded it yet and uploading it again, and this of course the whole world of email file attachments and finalfin2. doc and so on. So I think to me that’s why we kind of sunsetted that little experiment, the early shared web experiment, but now we do have the technology foundation for a live, even if it’s completely static, all you can do is navigate through it and maybe copy paste stuff in and out. If it’s live updating and you know you’re looking at the latest version, that is a game changer.

Alright, so one last roadmap question here is about linking hyperlink or wiki or Rome style linking. So one example of someone who asked us is Robert Haysfield, who says any plans on enabling users to place the same board in multiple boards? I struggled before trying to use Muse because I don’t know, hierarchical organization without an ability to bridge trees makes it difficult for me to find things and pick up where I left off. So I think for sure linking you know Ted Nelson’s excellent branding with hyperlinking in the web is obviously hugely empowering. We saw it in wikis. I think Notion is the modern version of a team wiki and then Rome with its kind of invented the category of, personal knowledge graphs, back linking. Now there’s a whole profusion of apps that do this really, really well, often with the double bracket kind of linking format. And yeah, I’m happy to say on that one, we’re actually working on it right now. So for those of you who are prom members, keep an eye on your backstage pass, some goodies coming up there soon.

00:31:04 - Speaker 1: The Backstage pass on this one is gonna be a really helpful piece of feedback for us to hear from customers what they think of this linking feature, cause there’s a few different ways we’ve looked at building it.

I know Yuli and Leonard are taking the lead on this, and It’s a lot more complicated than just, oh, put a link down, you click it and you go.

There’s a lot of nuance in how those links are presented and How the content appears in a link and in a backlink, particularly because, at least for me, a lot of my boards don’t have titles.

That’s probably a personal failure of myself, but linking to something that does not have a title, there’s not much metadata there, and so, showing a piece of the board you’re linking into is important, but then it gets a little confusing because it looks like a normal card, and so how do you tell the difference between a linked card and a normal unlinked card that’s not a mirror.

So there’s a lot of nuance and kind of how these things are displayed and so the Feedback we get from customers in the Backstage Pass is really gonna help us to iron out all of those little details to make sure that this is not only as powerful as the linking features we all expect, but really fits within the music universe and fits within all of the rest of our content in a tidy way.

00:32:25 - Speaker 2: So another big category of questions is using Muse, and this is a place I feel we’ve underinvested, which is that you can’t get a lot of information beyond the mechanics of how to use the app, but in terms of how people use it, and we have some projects in the works on that to try to better showcase what people’s boards look like and what the different kinds of uses you might use muse for and what things Muse is not a good fit for, but a few questions here that folks raised. The first one is from Alex Antozek. Who says, how often do you guys use Muse for non-muse related work?

00:33:02 - Speaker 1: Most recently, I started using Muse for custom keyboard organization.

So I’ve entered the rabbit hole of mechanical keyboards and building your own keyboards. I’ve recently come up for air, and I’m just using my regular laptop keyboard, but I think I’m gonna fall back down and fight the dragon again soon.

But Muse has been very, very helpful because there’s 1000 parts and there’s 1000 versions of each part, and Linking to a random Amazon link just does not give enough context for what it is that I’m actually looking at. And so Muse has been very helpful. I can put it in a photo, a short description, a link, and some context about each of the different parts and each of the different steps of construction. So using Muse to organize all of those different parts and be able to physically see photos of them next to the link next to the description has been a really helpful way to map out that rabbit trail that I have found myself getting lost in.

00:34:02 - Speaker 2: You’re just looking over my current news home board, I think it’s about a 50/50 split, kind of depends on what big personal projects I have going on.

You know, on the work side I have things like a little editorial calendar for the podcast, upcoming guests and things. We’ve got the chapter plan board that basically Leonard led the planning session and produced the shared board.

I’d like to be able to go back and reference that things about Twitter, I draft all our shareholder updates there.

I have a whole board that’s not news related, but it’s kind of career related, which is I which held a lovely un conference recently, and I have a bunch of notes from all the talks there and so on.

Then on the personal side, I have things like, I don’t know, the Kita, that is to say the kindergarten daycare that my daughter attends, you know, they sent out a PDF that’s, here’s the days we’re off this summer, and you know, in some cases I excerpt out the like coming up month, especially when it’s a critical, OK, we’ve got the summer holiday, and how does that match up to our summer holiday and this sort of thing.

But also holiday planning for sure. I mean, travel planning. It is, you know, in some ways a fairly basic use case, but it’s really helpful though you have pictures and maps and PDFs of tickets and it’s all in one place.

I do personal journaling, you know, just classic daily pages, sit down and write out what’s on your mind in the morning.

And then bigger projects, for example, when I was searching for a home a couple of years ago and eventually made a home purchase, but that is a place where the sharing and collaboration capabilities would be a big help because there I was doing it with someone else, my partner, and so I kind of have my personal world that’s inside Muse, but at some point we need to share and I can’t really share it directly with her, so then we kind of have some stuff stuck in an ocean or a Google doc somewhere, but that’s really awkward. So, yeah, when I have a bigger project, something like that, home improvement, yeah, personal hobbies, that sort of thing, that tends to occupy a big portion of my home board.

00:35:57 - Speaker 3: Yeah, I also really like Muse for personal projects, and one of my favorite pieces of it is that you can use it from the very beginning, where you’re just literally sketching out, you’re jotting down ideas, you’re taking little scribbling notes, and that board can evolve over time as you add text and PDFs and web links and images, and that rich multimedia is so important for these projects I find I can’t imagine going back to a world with just one content type. I also use Muse just for my to do list, that’s probably my heaviest use case.

I’m in there every day, and that’s mostly just text, but it’s not all texts, you know, every once in a while there’s like, whatever, you gotta call the plumber, so you put the link for the plumber’s page on your to do list or what have you.

And also just like that it’s two dimensional so that you don’t just have one. Huge multi-page linear list, which is, I don’t know demoralizing for me. So I have on the top left, it’s like today’s most important stuff, and then way off on the right side is, you know, longer term stuff, and you can just kind of move stuff in two dimensions. So, I like it for that. I also use Muse as my reading buffer. So for links and PDFs, I have a big per month reading board, and I kind of queue up stuff there and then maybe once on the weekend, I go and read it all and then archive that board and go back next month and so on.

00:37:08 - Speaker 2: Mm. One I’ve seen from a number of customers is making a daily reading board where it’s actually a link or a PDF and some notes or thoughts that was generated by that. It could even just be like a half a sentence scribble, nothing too complex or this made me think of this other thing, or here’s a tweet that’s connected to this, and they just make one per day and it just scrolls vertically or horizontally, and then they just like kind of accumulate these over time. And it’s a way to synthesize ideas and get more out of what you read and pull together the sparks of inspiration that you’re getting from consuming the information hot fire hose.

So when we’ve heard frequently and I think we even addressed this in the last listener questions episode, but maybe it’s worth touching on here is basically about the home board and how you structure things or organize things there without explicit work workspaces. So a lot of apps, notion, craft, even Apple Notes or something like that will have kind of top level workspaces where you say, here’s my daily notes, and here’s my to do’s and here’s my stuff for work, and here’s my home.

Improvement projects and you kind of bucket things that way and they find it surprising maybe that you have the top level home board, it’s just like any other board, it’s a free form thing you can put stuff however you want. You can make neatly organized, you can scroll vertically, stroll horizontally, scroll both ways, you can have a lot of stuff, a little stuff, it’s like completely up to you. What’s our answer on that one?

00:38:39 - Speaker 3: So in terms of home organization, I tend to have two dimensions going on. The first dimension is more obvious, which is, it tends to be organized by projects, ish or domain.

But there’s also an organization along the dimension of like recency or proximity to the cash. It’s like the things that I’m currently and actively working on.

So what I end up having is I have a few boards, both work and personal, that I’m very actively working on. And then I have a few more like archival boards that are more defined in terms of the domain. So I might have a board for my today’s to do list and a board for today’s reading, and a board for whatever my personal project is for this week. And then I have my muse board and my personal board, and I might even have, if I’m just a little bit discombobulate that day, I might even have just a few things blasted on the top level, like here’s a PDF that I want to read, and here’s a, you know, repro case from a youth book I’m working on.

And what happens is things filter down through this cache hierarchy over time. And this is a good example of how we didn’t want to have too prescriptive a setup in terms of workspaces. Like if we just said you have to divide your corpus into workspaces, it wouldn’t work for having the second dimension of the cash hierarchy.

Also just kind of go into the philosophy of why we don’t have a separate construct for workspaces. It’s kind of echoes the thing I was saying earlier about search cards, like we wanted you to have the full richness and power of muse to be able to organize your top level in the same way that you would be able to organize your individual boards. Like you have all these capabilities around, you know, freeform boards and inking and multimedia and different sizes. Like why should you have to throw that out the window just because it’s the top level? That seems exactly backwards.

00:40:16 - Speaker 2: I’m reminded of we talked about a very similar philosophical point with Pallo a couple episodes back where basically on sketch they originally had kind of an art board template editor that was a standalone thing but you quickly realized you wanted to do all the things to the templates that you wanted to do to a regular art board, so eventually they made it so you could just name an art board as a template and then whatever you do to that art board.

Goes into the template and then you don’t need to make a duplicate editor but worse.

And I think there’s something similar where if we’d make a top-level workspace, OK, you need to be able to move stuff, delete stuff, you probably want to duplicate things, and at some point you go, wait a minute, I want to just do all the same things I can do in my subboards.

00:41:00 - Speaker 3: Yeah, now in fairness to them, many, many users who keep requesting this, I don’t think it’s just that people don’t understand this possibility of generalized boards being used at the top level.

I think there is some sense in which the current freeform spatial boards aren’t the best fit for some top level use cases.

If logically, you just want to split your space into A, B, and C, and you want that to be Coherent regardless of what device you’re on or what size you’re looking at or whatever, or as you add and remove top level spaces that having to manually manage that on a spatial board isn’t exactly right.

And this comes to the topic of spatial collections, which we’ve had on the radar for a long time. And you can imagine a spatial collection working as follows. You put end things into the non-spatial collection, it’s like a folder basically, and then Muse automatically sizes and arranges these objects as appropriate, sort of like math finder.

You can imagine it it sizes and arranges them and lines them up or whatever in a nice way. And if we had that primitive, I think a lot of users would opt to use that at the top level where they’re currently asking for the separate workspaces feature.

But importantly, such a non-spatial collection could also be used all over the place. For example, my reading board should probably be a non-spatial collection, it’s a little goofy to have to manually manage literally 100 PDFs on a board.

Um, there’s kind of like weirdly overlapping in places and whatever.

And also, by the way, coming back to our search discussion, that should probably just be. A view that dynamically produces a non-spatial collection. So this is another example of how you get the right primitives and they can apply in a lot of different places, including this top of workspace idea.

00:42:32 - Speaker 1: I think one thing that’s interesting about the way I’ve organized my home board is.

I have all of the projects that I care about visible when I’m fully scrolled to the top left corner, and then the priority is based on kind of the size of the card, and so my personal card is fairly large for family and that sort of thing, work, I often go into that, that’s a larger board. Then I have lots of much smaller boards.

But then I’m able to actually hide less important boards outside of the scroll visible area, which is nice because some of those boards are important, but they kind of stress me out because I just don’t want to think about that project right now. And when that project shows up, I think, oh my gosh, I haven’t finished that project, and my heart starts racing.

00:43:14 - Speaker 2: And so if I just hide it out of view, tuck it off camera, exactly.

00:43:16 - Speaker 1: It’s, you know, out of sight, out of mind. I don’t have very much object permanence for some of these projects that are a bit more discouraging that they’re still around. And so, that’s been another nice thing for me, it’s just a trick to be able to focus on certain projects that I can see, and then when I don’t want to focus on it anymore, I can put it physically out of my sight, but still very readily available, and so that’s a nice thing.

00:43:44 - Speaker 2: Yeah, you point to some interesting column folk behaviors there that now I realize I do, but up until you just described how you do it, I didn’t realize it’s something I do almost unconsciously, which is, yeah, the size and the position of the board reflects how roughly important it is or how large it’s looming in my consciousness and that as something is still kind of current, but maybe it’s in a monitoring state, I’ll tend to shrink it down a little bit and move it kind of down lower, it’s down more towards the bottom.

And then eventually, once I decide it’s really not relevant to me anymore, then I’ll move it to a board called archive. And I think Mark you said you have her kind of project and category boards, I kind of just go to the Gmail route, which is archive everything. If I need to find it later, I’ll just scroll through all of it.

Which I think brings us to another question on use use and also connects a bit to the roadmap, which is this one actually comes from our support channels. So since it was a private correspondence, I won’t name the person. I hope they’re all right with me quoting their words here since I think it’s representative of the common question. So they say, what is your recommendation for archival of old news boards? I find that currently my main board is getting clustered and I have a number of ideas that can be archived. I like to export the new board and add to Devonthink database for storage. Yeah, I sometimes do the same or I’ll periodically take my archive board when it gets kind of big and I think, OK, all this stuff is pretty old news now, just kind of historical interest and maybe I’ll just export it and save it on the Dropbox or iCloud or something. But I thought this one was interesting because a feature I’ve wanted from you is an archive button similar to what GitHub has this for repositories. Gmail obviously has it for email. There’s other examples, and I asked Leonard about this before we recorded just because I know it’s part of his vision for the user flow, and he basically says that he doesn’t think it’s useful to keep everything you ever have worked in in one giant space and sort of it’s nice to put something away when you’re done with it. Lets you focus on what’s currently important and you can think of it as kind of the muse home board is your desk is the stuff that’s active and current and important right now, but then there’s a longer term personal library, things that represent your knowledge work and thinking over the longer term, and so we may eventually have just, you know, a context menu option for archive, but then you can pull it up in some kind of a search or. A non-spatial collection or something like that, or maybe just automatically adds it to a board named archive, but it’s a sort of a nice way to decide you’re done with something and put it away, but no, you can still get to it when you want it, do the much more drastic act of deleting, which always just feels really wrong, especially for something you spent a lot of time thinking about and working on even if it’s not current anymore. What do you both think? How do you do your personal archival of boards and what do you think about potential features in that direction?

00:46:36 - Speaker 3: Well, there’s actually a lot going on with archiving, so there’s archiving in the sense of removing it from your desk, which I do through this cascading hierarchy ending in what are basically archive boards.

There’s archiving in a sense of persisting the data in a file format that’s likely to be around and call it 1020, 30 years, which for me the only thing that I trust for that is like playing files on a Unix directory, so TXT, PDF, XML and JSON and so on. There’s archiving in the sense of like freeing up disk space, and there’s archiving in the sense of moving something into a third party knowledge management system. So the last one I don’t do, so I can’t really comment on it.

In terms of persisting to stable file formats, I try to do that for all of my work. I basically cascade everything that I ever do into One of a half dozen file formats into a big directory called data, and I try really hard to never lose that directory, you know, I back it up in all kinds of different places and so on. That’s a nice forcing function. I do that occasionally with Muse, basically export to a muse bundle, call it, you know, Corpus-2022, whatever, month, day, and just save it and forget about it.

00:47:39 - Speaker 2: A quick note for our listeners here actually, because you’ve used the Corpus terminology, we use that internally to talk about a one person’s muse database, the collection of everything that’s in it. I think more publicly speaking we’ll typically say your muse, put something in your muse, but yeah, corpus is, I think it’s a Latin word for body, so it’s just the body of your knowledge, data, what have you. So we use that to differentiate from board or set of boards or a bundle export or one subset of data, but it represents one user’s complete. Yeah, use database. Now do muse bundles fulfill that kind of flat file format desire for you because they are zip files containing flat files?

00:48:21 - Speaker 3: Yeah, not completely ideal, but certainly if I ever needed to go back and get something, it’d be easy enough.

00:48:28 - Speaker 2: Certainly not as browsable as, for example, exporting everything to PDF or even images if you wanted to take the effort to work through the hierarchy and do an image of each board.

00:48:37 - Speaker 3: Yeah, and I can actually imagine a world where both for this persistent durable archiving and for integration with third party apps, you really lean on the scriptability and programmability. You could have a little bot that says, whenever I see a new PDF, send it to my PDF manager. That sounds awesome, but I don’t do anything like that myself yet, I see.

00:48:57 - Speaker 1: Whenever I want to archive something, I think that the ark has already hinted at this.

I think the Muse bundle format is a really nice one because it’s a zip file, you get some JSON, you get all the attachments, and so you know you always have all of your data there, kind of no matter what, and very simple to open formats that are gonna survive for the next.

50 years, and I’ll generally save that right next to a PDF export, and so I can open up the PDF, look at everything. If I really want it back, I can import it into Muse. If I’m 85 years old inside of my robot body in the future, then I can just open up the zip file and look at things that way, and so I know that I’ve got the future safety of those archive formats.

00:49:43 - Speaker 3: This is a bit of an aside, but I read a fair amount of history and you always read about these historical figures, papers, and like, basically everything is in there for a lot of these folks, like their drafts and their correspondence and their bills and whatever. And I asked around a little bit like, how do we have all this? Like, how do we have all these papers? And the answer that I got was basically people had a big box and they just, whenever they wrote something, they put it in the box or a copy of it in the box. And it’s not super useful the day, month, or year after you do that.

But it over a course of a lifetime, it builds up and it’s nice to have all one’s papers. I wish I had known this when I was younger, so I could start archiving it, but as of a few years ago, I started building this set of documents. So hopefully that accrues over time.

00:50:22 - Speaker 2: Yeah, I’ve looked at the accumulated papers of Darwin, of Alexander Humboldt, Marie Curie, and other kind of famous thinkers in the past. It’s yeah, it’s a huge amount of content, right, because it’s, yeah, every random scribble and letter they wrote to someone, but of course very valuable for people that did do these breakthroughs, understanding how their minds worked and how they came to those conclusions, what their interactions with others in the field were like.

But I wonder if there is also some function of just people didn’t move that much back in those days, because at least for me, who’s done a lot of moving in my life, including across continents, you know, hauling giant boxes of papers, feels kind of infeasible.

Happily, hopefully digital archiving should be easier again if you take the steps that you describe, Mark, because there’s many ways that digital stuff is more ephemeral.

00:51:14 - Speaker 3: Well, maybe now we’re getting to the whole theory of information persistence, but I think not only are we moving more often, but I think this digital stuff is fundamentally more brittle. Certainly the very bespoke and fast moving apps, but even basic file formats on Unix directories, if you don’t quite actively maintain those, they go away after 5 or 10 years, like the disk corrupts and the media format is no longer readable and so on, whereas with a book, if you just like don’t light it on fire, it’s gonna be there in 100 years. I don’t know, it’s a very interesting property. I think we still haven’t fully confronted the consequences of trading off dynamism for durability in the computing realm.

00:51:51 - Speaker 2: And for interested listeners, I’ll point you to the Meta Muse episode on software longevity, when we take a deeper dive into the very topic.

So one example is another question from Petty Chase, who asks how you and maybe customers who share their use cases and workflows with you use Muse with other tools of thought.

And so typically we see a lot of people linking from knowledge graphs.

This is where we do seemuse as complimentary to Rome, Obsidian, Logsick, and so forth that you can link out to Muse board.

People often do that by getting the deep link.

You just basically hit copy on the board and that will give you a muse app cola slash link.

And also you can like, of course, the other way around.

We’ve also seen folks use shortcuts to put, for example, news boards directly on their home screen as kind of like a launch point then obviously something like screen share, which I guess is less of a tool for thought, but I think is important in modern work, so using use as a real-time whiteboard or a kind of a presentation live presentation tool, including teachers, they use it for their classes kind of Choctaw style.

We obviously use that for presentations and planning sessions on our team. As well. So I think those are some of the simple ones. Can you both think of other examples?

00:53:05 - Speaker 1: Yeah, for me personally, it ends up being more of an archive question than an integration question.

I find myself almost purposefully keeping apps separate, and when I work in Muse, kind of process the information and create some sort of an output, and then I’ll take that and just physically export it into Notion is the other common app that I use for storage is kind of my knowledge graph, my personal wiki. And so I end up exporting from notion into Muse, processing around, doing some thinking and then exporting from use back into notion.

But beyond that, I think links to and from are also a very nice thing, Wiggins that you mentioned, because then it’s integration agnostic, no matter what other apps people use, you can always create a link to and from things and so it becomes a very lightweight interaction point. I think there’s some risk for creating very heavyweight integrations. Maybe we’ll talk about that soon, but it can be very limiting, ironically, to create a very deep integration with some other app because it forces that workflow, as opposed to allowing a lot more flexibility with very lightweight integrations with lots of different apps. I think there’s some balance there, but I’ve found the very lightweight integration slash archiving export to be very helpful for my own use.

00:54:25 - Speaker 2: That’s a great point. Copy paste, drag and drop, shares sheets, you know, files in and out, that kind of quote unquote integration that is basically using standards rather than needing a many to many API integrations heavyweight thing I think is most always better.

That said, that is certainly a place I think we can continue to get better.

We’ve invested a lot there, but you mentioned the case of copy pasting. In and out of notion and weird things can happen with line breaks and some stuff comes across that seems like it should be left behind and other things sometimes get copied in or things get omitted, so I think continuing to improve that which partially is just the tricky challenge of trying to kind of work out what the other app is expecting in terms of line breaks and format content and You know, if you send the text in one way, you get a bunch of individual blocks, and if you send it another way, you get one giant block with a bunch of line breaks, for example. But yeah, I think that’s important and something we can improve a lot.

It’s a fascinating question we get from our friend Tim Lloyd. Tim writes, text first tool for thoughts. Is that a noun tool for thoughts? Yeah, why not. Like Rome and spatial first, like muse. So he’s comparing text first and spatial first tools. Do you think these will converge? To tools that are great for both. Are the current differences more about technical feasibility and interaction challenges, or is it actually an incompatible vision or just two different kinds of ways to aid thinking? And if it is the latter that it’s sort of incompatible, the sort of spatial first is fundamentally different from text first, does that mean there’s things Muse would never do? That’s the end of Tim’s quote, but I’ll just add on, we’ve already talked about adding things like linking and search. What you expect from the text first tool for thoughts. So you know what’s the limit on that? And yeah, is there some world where both the text-based stuff grows to be more spatial or visual and use grows to be more textual, or is there a limit on that and they’re just sort of fundamentally different kinds of tools.

00:56:31 - Speaker 3: Well, OK, I think there’s some abstract sense in which we’re on a multi-dimensional tool space and there’s different points in those spaces for all the different tools like notion and Rome and uses and so on, and theoretically you can imagine those tools traversing the space to meet up somewhere. I think in practice. You make foundational decisions pretty early on that tend to strongly suggest which region of the space you’re going to tend to move around in.

So I guess I would expect some convergence among these tools, but not 100% overlap in the future.

I expect Rome, for example, will remain quite text focused and use will remain more free form and spatial, but You will get more tech support, you will get non-spatial support, and so on. I don’t think that they’ll exactly look the same in the end.

00:57:21 - Speaker 1: For me, this ends up becoming a question about file standards more than it does about the application functionality.

And I say that because I can take a notion document and fairly easily translate that into a text file, a very linear document format.

There’s currently not really a very good file format for Spatial canvas, and so converting a spatial canvas into a linear document.

It is currently a very difficult thing and depends very much on the tool, and in some ways taking a document can translate literally into a very tall spatial document in air quotes there.

But being able to convert to and from different formats, or even just have a standard format for what a spatial canvas document is, I think will really help.

Bridging these two worlds together, because right now there’s just not a good way for Muse to talk to another spatial canvas app. There’s just not a language that we both speak to describe what a spatial canvas is.

00:58:28 - Speaker 2: It’s a great point that For example, for the top to bottom text oriented documents, I use now and basically always have used lots of different tools. At the moment I would say I use raft and Notion fairly equally and I use still quite a bit of Google Docs.

I think there’s a lot of things for kind of multi-writer editing workflow that Google Docs is still the very best at. And those are all pretty similar. You can really copy paste between them. There’s annoying incompatibilities and little formatting things sometimes, but fundamentally they’re the same document type. And so it would be logical if this infinite canvas as a category or a document type does become a thing as it seems to have been, you know, figmented the space with fig jam. You’ve got Miro who’s doing very well. Apple is now coming into the space with free form. And is there a dot infinite canvas file format that you can move stuff around, or could I select everything on my canvas and TL draw, hit copy and go over to view and hit paste and vice versa and expect stuff to come across reasonably well, and is there value in that? Or are these tools ones where you would expect someone to use. Multiple of these things in their daily lives and want to move between them versus just a more simple ejection of like, OK, I’m tired of this one kind of canvas app and I want to bring a bunch of stuff I’ve been doing there out into another one that I at the moment like better. I think that’s something we’ll see as the space evolves. It’s just still so, so new. I think my answer to Tim’s question is that Tool for thought nerds and personal knowledge management nerds like to really focus on very specific features and cheer for their favorite team in the sense of which product they like best or whichev casting or getting things done, you know, creative process, productivity system they like to use, and that’s all good fun. But in the end, I kind of think that all personal knowledge management and even team knowledge management. are generally kind of in the business of essentially the same thing, which is letting you put things into the computer, you rearrange it in some way and you retrieve it with new insights. So in a way we are, I consider that we are not only in the same space with the realms and notions of the world as well as the more infinite canvas style murals and Digital notebooks, you know, iPad inking oriented sketchbooks like good notes, but I would also count Evernote, you know, more classic sort of notebook keeping stuff app or even Dropbox. Honestly, Dropbox for me for a long time was the core of my knowledge management because I do use so much flat files and certainly back in the days when I was spending much more time at the terminal, less these days and indeed Muse has replaced a lot of things that I would use Dropbox for. So in a way there’s some 27 dimensional space that you could somehow draw all these like knowledge management tools, including just a file system, right, or like a plain vanilla notes app that sit in this space and as apps become more successful and they add features people are asking for. In general, moving between apps, much as we like kind of multi-app workflows and small sharp tools and whatever. The reality is like people like to just, if they’re doing 80% of their work in one app, they want to do the rest of it there as well and if you just add a couple of small features, sometimes that’s good, so. Yeah, I kind of imagine that Muse does actually, you know, we’ve already done a lot more with text, for example, things like search or linking, even back linking. I don’t know, is there an outline view someday, you know, I could imagine us doing 20% of the things maybe that an outliner text-based linking back linking knowledge graph would do, but never the host of it and as you said, foundationally, that’s not the core of it, it’s not how it feels, it’s not the core DNA to use a hackney piece of jargon.

And similarly, I expect the same elsewhere, right, that you know more text-based tools might adding whiteboarding and diagramming and visual features and but it probably would never be more than 20% of what Muse does and will always be a kind of added in thing that’s maybe a little secondary. So I think that works fine because there’s people who are very squarely in the space of, you know, the text oriented stuff and they just maybe want a quick whiteboard and so for them that’s fine and muse would basically be overkill or not a fit for their needs. And similarly, there’s people and I count myself in this category that I don’t find myself particularly drawn to the big complicated knowledge graphs and assembling, I guess like really long term kind of archival of my active work is not as important to me. I’m really about like what am I thinking about right now? It’s the project that’s on deck right now, it’s the active thinking and as you said, Wulf, in some way there’s an artifact from that, you write that. Article and put it on the web, you ship the product, you finish the home improvement project, and it’s good to have that stuff as a record of what you did and your thinking and just for basically nostalgia reasons, but it’s not something I need indeed would be a distraction to be kind of popping up top level in my work all the time. Maybe I’d feel differently if I was spending 5 years writing an epically long book or something, something like that.

01:03:32 - Speaker 1: The risk of terrible metaphors, carpenters have hammers and they have saws, but they don’t have a hammer saw. I think there is value that tools define the problem that they’re solving and make sure that they solve that really, really well, and don’t try to solve too many problems within a single tool.

01:03:53 - Speaker 2: So coming a bit to the Muse team and how we work, we had a couple of great questions here.

Ben Shelford asked one about how we decide what features to add and what we don’t add, what the inputs we use to that and the thinking and conversation, and then Ruben asks, what’s your internal structure for making decisions, which I think is kind of getting at a similar thing, setting road maps and ultimately shipping the work. So briefly, I can kind of describe that we have a very unusual way, I think, and I’m not even sure if it’ll scale in the long term, but I really like it. It’s very consensus driven and in my experience it doesn’t normally work that well, but for some reason on this team it works. Maybe it’s because we’re a small team, maybe it’s because we’re just kind of all on the same page and what we want to do, but for whatever reason, we essentially can do kind of two levels of planning. One is the Team summit that we do every 2 months, we try to make in person when we can, although the last couple of years, that’s been not so often, we actually have one of those coming up here very soon in Portland, Maine, but there’s where we really zoom out, look at the big picture, kind of spend time on little workshops, we dig in on subsets of the technology or the design and then essentially make proposals for what we think we should work on, big or small, but then the chapter discuss that and then have the The making proposals and discussing it is the fun part, then we have the later meeting, which is the make the decision, which is always the agonizing part because it all looks so good and we wanna, we want to do all of it. And ultimately the consensus is not just sort of let’s all vote but is more a sense of, I think we really let individuals who are highly motivated. To drive a project where they say, I just really think we got to build this thing. I have a specific vision for it and then one of the persons says, yeah, I’m on board with that, and then those two people are gonna be the one to go do it. And so the rest of us say, yeah, I think customers will like that, let’s do it. And maybe it needs to fit into a bigger strategy if we have a longer term roadmap, you know, what’s Muse2, what’s Muse3, etc. but still. We try to be pretty opportunistic and just follow our instincts and our hunches and the inputs we get from customers. We are all on their support channels. Every week and talking to dozens of customers, each of us individually talking to dozens of customers, so in aggregate that’s probably hundreds, and we’re in close touch with what the pain points are and what people are trying to do, and what would add value to people’s lives or solve problems for them. And we go from there to a weekly planning meeting where we just basically look at that chapter plan and say, what are we going to do this week? And that’s kind of the whole thing, I think. Do you both feel like they captured that well?

01:06:32 - Speaker 3: Yeah, and I would emphasize that it’s very energy oriented, but we’re really looking for who has the energy to articulate, motivate. Design and execute on these projects. And it’s not consensus in the sense of all 5 or 6 people need to Agree, or, you know, everyone has a veto or anything like that. It’s more just like the thing that has the most doing behind it gets done. And there’s also a very large amount of trust, so by default, if Wulf here thinks we should really do something, I tend to, you know, believe that versus needing to go in and second guess everything.

01:07:12 - Speaker 2: Part of that is certainly creative trust. Hillary Maloney, when she was on the podcast, pointed out her time working with us. She was struck by how much creative trust we have, but another part of that is context because we all have been around the business for a good while.

Relatively speaking, and we have pretty broad insight into everything that’s going on and we’re all serving on support.

There isn’t a lot of siloing, which is natural on a small team generally, but I think we take that even further, so everyone has the full context to make decisions and that doesn’t work if you have someone who says, you know, I love to scale databases, you hired me to scale the database. I’m going to do that. I don’t want to think about your conversion funnel or You know what the marketing strategy is for this week or you know what the design challenges you’re struggling with and adjust your space.

I care about my domain. We specifically select for folks who are very interested in everything. All three of us, for example, have been entrepreneurs multiple times in the past and that kind of it’s fun and enjoyable to have your hands a little bit in everything, and then you have the context to make decisions within your domain.

And you make a great point there also, Mark, that it’s definitely not someone can veto it, and it very often happens that, yeah, 1 person, 2 people, 3 people say, yeah, there’s this thing we’re going to do, it’s going to be great, people are going to love it, and maybe 1 or 2 other people are even kind of scratching their heads and going, yeah, I still don’t really get it, but you 2 or you 3 seem really driven by it and I trust you, so go do it and maybe it’ll come clear later. It was that way for me with the text box, spatial text. Stuff and Julia started working on that. I really didn’t get it. I didn’t understand how it was different from the little text cards we had currently, the early versions didn’t work very well. It’s just Texas is really hard, and it took me a while to fully download the vision, but once I saw it, I realized it was awesome.

01:09:08 - Speaker 1: Yeah, for me, I think the combination of the proposals that we write up every summit is also very helpful, not only for the current summit, but for all future summits, because we very often revisit that same idea over and over.

The past number of times, sync has obviously been at the top of the list, so that’s what we focus on, but the spatial collections has been in our thought process for a very long time, and so the team is ready and motivated to work on it.

It’s just about finding the right time.

And Wiggins, another thing that you mentioned is we all work on support and having That shared experience as a team, seeing the same support tickets come in, the same requests from users, the same problems, really helps us stay on the same page about what are the major things that our users are asking for and running into, and that gives us a shared priority for a number of issues that are on our to do list when we all see the exact same urgency from the users, and that shared brain space over the past months and years working together really helps that consensus-driven approach.

We’re all playing the same ball game and we’re all on the same journey together.

01:10:17 - Speaker 2: Yeah I’ll also note that even can create energy in the short term. A good example from recently was, I think Julia did a sport shift and I don’t know, again, there’s been such an unusually large volume of support requests coming in post launch, or at least I hope it’s unusually large. I don’t know if we can sustain this forever, but Julia had seen just quite a number of people getting confused when they navigated away from a board, came back and couldn’t undo.

You know, if it’s deleted card, we have the restored deleted cards thing, so there’s there’s some recovery from that, but she realized that something that had been a technical blocker before, which is the way that core data, the persistence system we were using before, interacted with the Ndu manager made it really hard to Like an undue stack per board.

Now that we have our own persistence layer written by you, she realized that that was more feasible and just got motivated to just work on it essentially one day, I think it was like when she was on the plane on the way back from a trip or something like that. And so, you know, we just ended up kind of doing that and QAing it and shipping it because yeah, she was motivated by direct contact with customers, saw an opportunity that before was more challenging or had more blockers in the way and just went for it. So coming to a question about the lab, that is the ink and Switch Research Lab, which many listeners will know is the the place where Muse was born, Ruben asks, is MUS the product already fully financially supporting Ik and Switch? So I found this question fascinating. I’m going to give three answers here, short, medium, and long. The short answer is no. The medium answer is, I probably should explain the relationship between I Switch and Muse because I think not everyone realizes it. So I and Switch is a research lab. Essentially a not for profit kind of entity whose goal is to fund research work, publish papers, and then at some point spin out things that are ready to be commercialized and Muse was the first of those, but hopefully not the last. So Muse is a separate company, a for-profit, C Corp, uh, but Inc and Switch took an equity stake in Muse for their role in helping gate it. So in that way you can see the relationship between Muse and it can switch as being similar to the relationship between a YC company and Y combinator. So really it’s not possible for any one. Spin out to make an accelerator and I think it’s going to be the same thing for I and Switch if the spin-out strategy works, there will be a portfolio of these. Also, logically you can just look and see that you know I and Switch is a bigger organization just headcount wise than you, so you know it just logically wouldn’t make sense that the smaller organization supports the bigger one. But I do love that this points to an even longer answer, which indeed perhaps at some point we’ll have to get Peter van Hardenberg back on the podcast to talk about how the lab finances work, because it’s a very interesting and challenging question and something they’re actively developing. But sort of there’s the spin-out strategy, spin out companies that can earn money and then potentially that goes back to and switch for more research. There’s also some lab for hire work if there’s an area your your company needs some the specific expertise of ink and Switch and you want to hire on kind of a consulting basis, ideally something they can get a paper out of that’s on the table. There’s other kinds of grants and things like that as well as even potentially like public money and so on. So I think like a lot of research organizations, it’s a matter of pulling together revenue from a number of sources including just straight up patronage. So I believe the automerge project is now generating a pretty good bit of patronage through different sources and that also goes to fund research work, but we very much hope that you can think of that the portion of the proceeds when you buy Muse go to support more work from Inc and Switch, and if that model proves itself to be viable both for use. For the lab that there would be more examples of that in the future, and it produces a nice flywheel of a lab that can do innovative research, think about the future, not think about commercializing, but when they do stumble over something that has good commercial potential, they can take some of that while not distracting, you know, not converting to a startup or something and distracting away from the ultimate goal of always being looking to the future and just adding to humanity’s total knowledge rather than commercializing any particular piece of its findings. Well, maybe a good question to wrap things up on comes from Ilia Wulf, who says, would be curious to hear what you want Muse to be 2 years from now.

01:14:49 - Speaker 1: Very eager for not only muse, but the spatial canvas thinking canvas category to continue to grow.

I think some of that I’m looking forward to is the file formats for interoperability and better archive storage.

I’d really also love for Muse to be able to open source some of our work, maybe with a sync engine or with some of the other things we’ve done, because there’s the spatial thinking category, and there’s the local first category, and then there’s the overlap of those two categories, and I think out of all the interesting things Muse has already done. It would be great to be able to support a lot of the other products and companies and Pieces in this growing forest that we find ourselves in.

01:15:37 - Speaker 2: Yeah, I found the question interesting because they didn’t say muse the product or the app or the thinking tool. Muse, of course, for us, has a lot of meanings. It refers to the product, of course, but it’s also the company, team, and even the brand.

And so I I find myself thinking a bit about the company itself being an example of a sustainable kind of small giant indie company, like some of the role models we like to see, for example, Panic is a kind of indie studio example or sketch, as kind of a bigger company that’s carved out a really good niche for themselves in the design tool space.

And I think being in this middle ground between we’re not bootstrappers or indie hackers, we’re a bigger team, we’re investing in more cutting edge technology and design approaches, but at the same time we’re keep staying really capital efficient, staying small, small team, lots of trusts, everyone does a bit of everything, and that is a model I would both like to see because it just reflects the kinds of workplaces I’d like to be in, but I also wonder if it fits with maybe the way the technology industry is changing that we might be a little bit.

Getting towards the end of a winner takes all rapid frontier growth and into what’s sometimes called the deployment phase, and there is a lot more room for a wide variety of tools and obscenity space.

And so, this style of company, again, there’s role models I look to and maybe we can, if we do it right, maybe we can be an example for others as well.

And I think the brand as well continuing to see that stand for more than A multimedia canvas as good as it is and as many great features as it has, and so on.

In the end, you know, I hope we stand for thoughtfulness and curiosity and serenity and in a very noisy world where it seems that technology is just always fighting against us with dark patterns and trying to like get our engagement loops and dopamine loops going with notification. and so forth and even beyond technology just I think a greater degree of thoughtfulness in the world would just sort of benefit humanity that everyone taking a little more time to slow down a bit and just think about things that are important before they do them or before they engage in the hot take or whatever, whatever it is. I think that’s, that’s something I hope we could. Not just stand for with our team and our company and our products and our podcast, but also maybe something that in a small way we could influence the world for the better.

01:18:04 - Speaker 3: Yeah, there are likewise a lot of ways that I can answer this question, but I might focus on the product side.

I had a very particular product vision from you going back, I don’t know what it is, 3 or 4 years or something, and I feel like it’s actually stayed pretty consistent.

It’s this idea of a personal information tool that takes advantage of the unique form factors, is multimedia, is resonant with how the human mind actually operates as high performance as local first. I thought we’ve gotten about 50% of the way there, we have a very good line of sight to most of the rest. And so I really want to see this thing exist in the world. A, because I want it for myself and I think it’d be cool. But also, it’s kind of this definite optimist take where I kind of feel like if we don’t do it, it might not get done and that’d be very sad. So I really want us to succeed in bringing this vision, which is now of course, grown beyond just myself to the whole team, but see that brought into the world and see it earn the success that I think it can get. It’s a lot of work to do, so let’s get back to it.

01:19:05 - Speaker 2: Well, let’s wrap it there. Thanks everyone for listening. If you have feedback, write us on Twitter at NewAHQ by email, hello at msApp.com. Wulf Mark, it’s been a great ride these last 3 years. I’m glad to hear we’re 50% of the way through. That’s actually a pretty solid progress bar, so let’s keep pushing through to 100% right on to the next 3 years.