Modal

yeah

it's like sort of a

content question. The project, like reflection in particular, is like a small part of a number

of different things that are all connected in, I think, not particularly obvious ways.

Yeah.

So like, where would you see the boundary of the topic?

I guess the best way to go about this is just to talk a little bit about

what the podcast is about, and then I'll let you discern what becomes relevant or not.

That's actually really good. I listened to one episode with Nico, but like,

other than that, like I don't really have any context.

Yeah, okay.

I'm really research.

So this podcast started, started as a quotation, like five years ago on Scuttlebutt.

And originally my idea was to just do a solar punk podcast.

But with Scuttlebutt imploding and becoming like offshoots into like a gazillion different

projects, this has instead become a series where the general podcast is about solar punk.

So that would be like pre-figurative practices.

So like, what are practices that people are doing right now that by existing exemplify what a

better world could look like, or the utopia that they would like to see. The series that this

podcast is starting with is on peer-for-peer networks, which is also like a new term,

but basically offline for a local first routing agnostic and open source mutual aid, there's

something else peer-to-peer networks. Not every project I interview fits into that category

like precisely like, but there's 10 gentle interests in different ways and forms.

And definitely on the technical side, I'm sure like.

What's becoming seems to become more hopefully of an ecosystem.

Welcome to today's episode of the Solarcast Podcast. And this is the third episode in the series.

And the voice you just heard belongs to none other than to Pia Spernard from the newly

founded Modal Collective. He has a long history of working with the Nome community, and he will

bring us with insight into what Modal is and the newly launched application reflections.

Beyond that, we will explore some of those everlasting questions of,

can we be without the computer? The answer is often no, but if we could, what would it look like?

With no further ado, let's dive in.

Welcome to Pia Spernard. And you have a bit of a background and you're also part of a newly

founded collective. Yes. Called Modal. Do you first want to bring us with you on how you entered

Modal and what your background is? I can. That's a kind of a long story, but maybe to make it very

short. I've been involved in the Nome and like free desktop Linux community for about a decade.

And I've been mostly doing things around design, but not only like I think in the last few years,

I've been doing a lot of organizing and kind of like making the things that we on the design side

want happen in other ways as well. And I think like one thing that we've done as part of that in

maybe, I guess since the pandemic sort of like receded a little bit is local events here in Berlin.

I think even before we had like a local community and we were trying to do this, but after COVID

is really like sort of took a more concrete form and we had more people also from other sort of

friendly communities such as the people around post market who work on the kernel and like

want phones to happen, which is also a thing that we care about a lot. Like sort of free software

phones that are not Android or iOS and so on. And also, yeah, like very relevant to this peer-to-peer

people like actually having our software sync between devices is something that on the Nome side

with one that's in forever. But there was never like really a strategic way to do it. I think like

maybe say like I don't know 2017 people have talked about, oh yeah, we're gonna like do

GNOME Cloud or something. Like I remember like there's some conversations around this, but

of course it never happened because like just operationally for a free software project that

like doesn't actually have a channel to users like it kind of doesn't make sense. And also like the

economics wouldn't work out. And like for I think a large number of reasons like there

there's never been like a sort of solution to that or even like a strategic approach that

people felt like, okay, this you could go down. And I think that's how a lot of us came to local

first because it sort of promises this idea of yeah, you can have sync but like you don't have to run

servers. And so I think between that and like yeah post market and some other friendly communities

sort of the Berlin crew, I guess, grew beyond just GNOME. And I think that's where modal comes from.

We're like we wanted a way to include more of those people. What was the name of these events

you were hosting? Something with something? Boiling the ocean yeah. Yeah. Which I think just randomly

happened because Casey like named the first event. I don't like some really long abbreviation like

GNOME plus post market plus whatever. And then it was like an acronym of like 15 letters. And

then we looked at like what the cell is having come on. Oh it's like stuffed at like whatever

it's really hard and takes forever. So that was sort of like the stupid metaphor and I guess we stuck

with it. But it also has like quite a environmental like take on it. I mean I think it's no secret

that like there's a decent overlap between like the GNOME Berlin community and like the

climate justice scene. And so they're definitely part of it. But I think it actually mostly came

from the idiom. But also yeah like I mean the ocean is literally being boiled and we maybe need to

organize for the world that that's gonna produce. Yeah. And I don't know I think like there is

definitely like a big political side to all of this for us. But I think like in the day to day

work that like is only reflected like a little bit because sort of regardless of where you want to

go in 10 years you need to be doing all the same things now anyway. But yeah I think that's sort of

like the thing with organizing work. Like it's it's a long term. Yeah it definitely is. It's one of

like the key things when interviewing resilience researchers that they say that we have to have

what's needed in the future already now. So it's good that it's starting now.

I think that's that's how we've always seen it. As the crises sort of get worse we'll maybe have

more of the solutions. Maybe not. Maybe it turns out all this computer stuff we won't need it. Yeah.

But I guess our our sort of like bet is a little bit that we will at least to some degree.

Do you think the name itself brought in other kinds of people than you would have expected?

I'm not sure. I think it's mostly I mean it's not like we advertise it very widely. Like we

posted on a sedan. I think until like whatever a month ago it was only on like someone's individual

account. So but also we don't really like have the capacity to host like that many more people.

Like if if a hundred people showed up I don't know what we would do. No. So don't show up. Yeah

I mean do do show up if you're like in one of the like very few specific communities. We're always

really happy to have new people. Like sometimes actually people come from kind of far away like.

So now you're mentioning something interesting. One of these very specific communities.

What are these very specific communities? You mentioned like I mean so there is like people

who I guess still believe in like native apps. That's one group because like nowadays everyone's

doing electron bullshit like everyone's whatever shipping chromium with all the stuff like that's

that's one community. Then I guess there's like people who both are interested in like I guess

we call it emancipatory technology but like we haven't super like put out a definition for it.

Our overall take on this is that it's technology built for people who don't give a fuck about

computers. Why? I mean because that's most people like why should you give a fuck about computers?

Like it's it's not a thing that like everyone wants. But why would you care about technology if

you don't give a fuck about computers? Well because the world is built in such a way that you have to.

I mean there's a whole like argument you can make that like actually we really what we should

spend all our time is like campaigning to make it optional to interact with computers. Just

practically the dependency that was our way to widespread in society. I mean even setting

aside all the new bullshit you need to have a phone to enter like an event or like book tickets

or something or interact with the government have a bank account all that stuff. Even setting all

that aside just the absolute basic functions of society like required that you are able to

print a document or. Engaging with technology in multiple ways. Yeah so like I think there are

like multiple different approaches to this and I think ours is maybe a little bit more intervention

is than it is like just fully what are the English word like autonomous? That could work as well.

Yeah like I think like that is a little bit too far out for us. But I empathize like.

But I don't know I mean I think there's also like on a personal level like I always think about the

one of the hundred rabbits talks like has one slide in it that's like sort of this.

Okay I know I'm the rabbit's but not everyone does so. So it's like this two people art collective

that lives on a sailboat in Canada I think and they sometimes give these talks about

permacomputing and like sort of just generally the idea of like constraining computing and like

sort of how they are live on this boat and they don't have like electricity or like it's really

constrained. And so like they think about technology differently and so then they started rewriting

like all of their software in like their own assembly language for a special like whatever

virtual machine because they care about preservation or whatever but like I think in one of their

talks they also get to this point where like oh fuck do we need computers at all like this

viable and I think there's like the slide that's just like says like oh the ever enduring allure

of the computer. And I think we all do feel that. Like there is there is an allure to medium that

you can do whatever with and it's like very little resource constraints to like doing a lot of stuff

which isn't really the case anywhere else. But I think that's like more of a personal take on it.

Yeah I don't know like a personal motivation why a lot of people are in this space.

So I had a we met at D-Web workshop day just like two days ago here in Berlin and I was part of

the longer day and one of the topics that kept coming up was this aspect of like how do we get

rid of the phones from a physical meetup where we're all going to be there in person. And there

was so many different ideas of like how to do that and what kind of like technologies we could

use instead from mesh networking and creating our own meshes to just having hand-held microphones

instead of having a phone recording everything. So the thread that I was interested in like exploring

a little bit here is if you think of like emancipatory technology to use the term you just used

what is the dream state like where do we envision going?

I think like as a collective we've only defined this to the most minimal

then but I would say like step one it's fully optional to use any computers

sort of we were talking to a utopian world like it is possible to do anything as well as whatever

anyone else with no computers involved there's no like I don't know requirements to use specific

software especially a specific format all this shit like you are able to conveniently do everything

with paper or whatever. Does that also mean choosing what software you use like having alternatives?

Um I think that I would be more bivalent on because as long as it's free software it's not as big a

deal. I think the important thing is that you don't have to like use some proprietary budget

tool that like is surveilling you or whatever has to connect to the network needs to run on

our specific kind of hardware that like you can only get from Apple or like I think those are the

bigger problems and that ties into the emancipatory right um and then I think beyond that like you

want people to be able to yeah just like use computers for the things that they want to use

them for but like in a way where they're not sort of being constrained by the business model of the

people making both the hardware and the software which tends to usually be

B.O.S.F. works. Yeah. Uh well like yeah in order to like whatever make this poster you need to like

have your Adobe subscription and uh if I came in account and uh it needs to whatever be updated

to the latest version of iOS or yeah and I think like all of that budget is in the way of the

things that are actually interesting about computers uh or in a lot of ways just like makes it not

possible that's I guess the the most basic goal um but ideally the way we would do this is or at

least the way we think it should be done is by having like a fully free software um ecosystem of

native apps that like run your computer um work with no internet work with uh whatever without

like everything being a gigantic browser like actually small and like efficient nice apps

that are like designed not like just made by someone for themselves but like there's someone

there who knows how to design stuff and like makes them a nice experience um and I don't know yeah

I mean I guess I'm sorry said local first but like yeah they they sink in a local first way

such that you don't have to have gigantic servers somewhere they can like easily surveil other

traffic and uh ideally the whole thing sort of is built in a privacy preserving where where like

sends all the traffic through tour but I mean whatever you can go as deep as you want in any of

those pieces but like another side to this is that like you know this is a really nice utopian dream

but like realistically the world we live in is an absolute nightmare um and so like really what

you're designing for is that world not not the whatever nice oh yeah everyone can make art on

their computer world it's more like oh fuck we need to be using I don't know this machine that we

happen to have that requires a computer keep it working somehow yeah and we need to do it without

the internet and without access to the manuals and we need to that is the more realistic computing

scenario I think that's also where we're not like totally sure like how anything really fits

together I think that's that's where case writing is really an important north star yeah you need

to threat model everything way way way more than we did like even a few years ago so before we

dive into this because this is a whole round of all right yeah exactly so speaking of rabbit holes

let's wind back the time and go back to the original question of like who's on your team

and who's part of model yeah um so I mean I think like the basic idea for this came from the

Berlin community and then it's the people in that community who like were interested in doing a

bit more organizing work plus some friends who happen to be here more often that's it that's

the way very very very very very very modern summaries well and the other hand I think like in

Berlin there is like kind of a unique concentration of like people with these very specific focuses

and we sort of happen to have a lot of them like come to our events often so I think it's also

like a good cross-section of various fields like for example Casey from Postmarket is like one of

the most gifted kernel developers of her generation and really involved in Postmarket and like I

think has a really good grasp of like that whole part of the sack on the gnome side of Jonas or

Sebastian no more than probably other than like a handful of people at Apple I don't think very

many people know more about input than Jonas does or like you know there are like a lot of these

really niche fields that we happen to have some like uniquely qualified people in this community

and I think that that's sort of like I don't know yeah covers a lot of bases in practice like we're

not as a collective not doing that much technical work at the moment we're basically just coordinating

work that's happening anyway like it's it's not like we have like our own project of like oh this is

whatever a repo in the modal space somewhere where we're working on we're working on all the

same gnome flat back p2panda whatever projects that we've been working on for decades but like with

sort of a more strategic focus and then like we're organizing how to basically make more of that

happen essentially yeah and so people are essentially working a lot on their own projects

and then jointly coming together and organizing but one of the more concrete results that has come

is an app that I first heard about now at CCC 39c3 and that's called reflection yes and reflection

is an app building on p2panda as far as I've understood and has had input from a lot of different

people as part of modal right for sure and I think that's sort of like the counter example of like a

new thing that did come out of this but specifically because like there wasn't anything because like

in as I said maybe I briefly sort of touched on earlier in the gnome community they're like

wasn't and really isn't like a good way for apps to just have sync as like on iOS like if you're an

iOS developer you use the whatever iCloud standard API and you get sync for free as a platform we

don't have that and we want it but like there wasn't really a project that was doing the networking

parts of this and also not really like the higher level whatever data type CRT and any of that was

just like fully missing and so the idea with reflection was that we're like actively so I think maybe

three years ago we had a hack first where we talked through like all the different parts of the stack

that we would need and actually had like someone from incand switch Adam Wiggins came by and like

gave a little talk on like their muse app and like said like oh this is how we do it like maybe

you could do this too and we talked about all the parts of the stack and we're like fuck yeah we're

missing this and this and this and this and then I think like a month later I met Andreas at offline

and he was like oh we have all these pieces and so our idea then like was sort of we need a project

that actually does this and that's where reflection comes from like the idea is like an actual app

that puts all the pieces together and therefore is like an example for this whole developer ecosystem

to actually get sync for free so that's that's sort of like why we started our own or a new thing I

guess would you say it's like the prototype for future yeah for sure how did it go well it's still

ongoing we we actually tomorrow uh March 3rd is the first developer lab session so we we

as part of so maybe three more steps back uh we we we we stuck basically we had a little

like sprint over a weekend with the p2panda people and we started like building this this

text editor app because the idea was like oh we need an app that people will actually dog food

because otherwise it's not gonna get any testing and then we're not gonna get any feedback that sucks

so it's dog fooding uh the idea of like actually using your own software every day

honestly I think like strategically it's the only trick we have like it's like using the

software that yeah yeah I think I think at every level like whenever there's a problem it's like

oh yeah the problem is because we're not actually using it every day uh and that's why I don't know

like nowadays a lot of us use normalize because that way we like just get every single commit that

lands in any of the no modules we get it tomorrow yeah and then you actually can test stuff

months before it's out rather than months after it's out and notice stuff immediately or something

breaks it doesn't solve everything but it fixes a lot of things and I think like with with the

local first stuff that's just a classic playbook that we applied and that's why we chose text editing

even though like from a data perspective it's like not the easiest thing to do um because like

text is quite complicated and like you can be editing over each other quite easily in a way

that like if it's a database and you just like replace fields it's either there or it's not

it's not like oh and and I deleted the whole paragraph that you edited something inside over

whatever yeah um but so yeah we had the sprint and we like put together a prototype and then

after that we applied for a prototype fund and then um that's been going it's also that was

luckily accepted and that's been going since last June and is ending at the end of this month

so that's like sort of about a nine month period and one of the last really important things that

we're doing now is like sort of this developer outreach because um so maybe slightly technical but

like um in order to use a rust uh piece of rust software from a GNOME GDK like GDK is the GNOME

toolkit uh the thing that actually renders uh the user interface and so on in order to use

that kind of library in a GNOME app you need bindings and for that there were a bunch of

other things that were needed to do that and so now we're like at the point where we're actually

pretty close to like having a good nice API for like auto GNOME apps other than reflection

which is written in rust so it's easier uh to use that so like now we're like at the stage where

we really want to involve a lot more developers so tomorrow first session I don't know when this

comes out probably after maybe actually even already on Thursday okay but then then at that

point uh you can come to the section session which is probably gonna be later this month okay so

that's gonna be a series of these developer uh sessions and what's the purpose of them um

essentially to so exactly our original idea was we're gonna just like do a conference here like

in Berlin at local event but then we asked around and like actually a lot of the developers that

were most interested in this were not around here and so it felt more like inclusive to do

something online so it's online sessions yes it's just like a big blue button call like it's not

not that fancy but like the idea is that uh we we create a space where like people can come with

their app ideas and they're like okay I want to build this how would we do this and then we can

talk them through um like how to do it or like ideally like also get feedback for the APIs like

oh I think they want to do it it doesn't actually work without the APIs designed and so that's sort

of a process that's starting now so you're kick starting the ecosystem that's the idea yes

exciting and ideally we would like to go a lot further like so um the the first step is like we

just have apps use this but ideally a lot of this networking stuff especially would just live in

the system yeah and so you wouldn't have to do it in every single app like the app just calls

just like with the iCloud stuff the app calls the system whatever peer-to-peer thing and it gives

them the address of another device and then like maybe in the app itself we also have a library

that gives you a lot of those pieces for free so but it's a long-term process that we don't

like think it's gonna be done anytime soon but maybe like I don't know in the next years we'll see a

more complete story for this okay so one of the key questions that often happens is how to fund

this kind of development right and but it sounds like within NOM y'all are already dedicated to do

this is it like we're dedicated to do it no matter what and we're gonna put resources towards it or

is it like we're gonna do it but we definitely need funding and otherwise it's kind of called

off what's this that is on that i mean nome as a maybe for those not too familiar with the free

duster world like gnome is not like a company it's like there is a foundation but it doesn't

really like it's not involved in development and the funding as it is tends to like just be via

companies paying for individual people to work on module upstream and so like there isn't really

like sort of whatever gnome decided to do this and therefore funding goes there that i mean there

are people in the foundation trying to kickstart this which maybe they will

grow into something in the future but right now that isn't really the case

i would think like instinctively that this kind of development is so needed in so many companies

because it's like relatively groundbreaking like also replacing like the i cloud capacities

what do you think there's an awareness that this is possible in the broader scene

i think well we've been trying to build this awareness for the first few years so i think

that's another reason why i think some of us felt the urgency to like have something other than just

like all these separate organization is for someone to actually tell this full story and like advocate

for it in the various projects and i think that's where we see model like sort of as an additional

value just like by existing because it puts forward some of these i think slightly more radical

notions in like a coherent package whereas like on the gnome side there's like 100 different

opinions for why like they're interested in gnome and like there is i would say like gnome has a

very strong superstructure especially around like things like design and like sort of maintainability

good api is sort of stuffed to be long lived and so on i think there's a lot of really good

ideas there that are shared by everyone but for example like peer-to-peer stuff i don't think

things share by everyone and so in a way like model is maybe a bit of an intervention into

internet space to like try and push that more and i think we've seen some success but the local

first aspect it's just more functional in a lot of ways for like well i think like so think about

this side like if you work at redhead and like you're working on an operating system for some

big company clients that are fully bought into microsoft whatever the fuck but can i swear on

this podcast yes okay maybe a bit later ask but so um so i think like you know if you come from

the corporate side then you're so bought into these cloud ecosystems that you don't really see

outside of them also because like it's not like there is an easy to switch to local first alternative

to microsoft 365 calendar or whatever invite system i don't know i've recently seen some of

the complexity of that because someone's working on implementing more of those features for just

like calendar invites and you realize like oh yeah you can share like your whole calendar but like

not the details just like when you're not free and so on and you're like okay wow yeah there's a lot

of existing sort of interfaces to that stuff that i don't think are easy to sever so even if people

were fully convinced it would be like a multi-year effort to like pour away from it yeah and hopefully

that's gonna happen but yeah i think that's that's a much more uh it's a slow transformation process

for an entire ecosystem for sure yeah um i'm kind of curious so you've gone through a little bit of

the story i also read the whole intro to modal and in the text there's a specific section that

suddenly it's not super sudden but it goes through why modal exists it talks about like the big tech

and mass surveillance systems and etc etc all these issues that we're all aware of and and then it goes

into okay what is it we want to do and one of the things that was mentioned was building a

Linux iPhone for software and i thought that was interesting because that's such a specific

goal and we haven't even mentioned it in our conversation so far but i'm curious like what

what what do y'all mean with that and how does that compare with other like Linux phone initiatives

that are ongoing right now um i think for us like the idea was that we have a very concrete

way of framing all the things that we are actually doing like it's that's not like you know a goal

in addition to all those other goals that's literally just the the summary of like if you put all the

things that we've described together that is what you get ideally like because i think sort of our

our criticism i suppose of the the broader free software like movement is that it is largely

built by experts for experts and to us like that is not a mandatory that is like sort of great for

your little club of like ccc hackers but like for everyone else this is not actually providing any

value and at that point like at least personally i'm not interested in that and i think one of the

things that gnome like has really like in its DNA is this idea of like building for the commons

building for the public and i think if you take that to its logical conclusion and try to find

a PC cashphrase that is sort of what you end up with the idea of like all the people in your life

who have iPhones and who are like i understand why you have an iPhone like i don't have a good

solution for you really other than like whatever coming to your house every other month and like

helping you whatever maintain your your like linear to us which i think a lot of us still do like

it's not like we understand the precarity of it and like the reason why a lot of it is the way it is

it's not like necessarily people's fault like main lining an android device is incredibly hard

work things are the way they are for reasons but also we don't want that and i think a lot of people

like in in the free software world like that become a little gatekeeper what's up and it's like oh yeah

whatever if you don't know the terminal like you shouldn't be using this stuff and it's like that's

that bullshit like we we don't want that yeah and to us that is i mean of course i understand like

every metaphor it's leaky and you can be like oh it's like the iPhone in the sense that it's like

extractive uh whatever and like trying to sell your new phone every year and like spying on you

and so on that's obviously not what we mean but uh i don't know it it it seemed like a fun

slightly progressive tagline and i thought it was great i mean it caught my attention

yeah i as recently as just two days ago at the d web thing i heard someone say

yeah what we really need is just to build the capacity of regular users so that they can see

the security risks themselves um uh yeah i would just be really curious to see whatever infrastructure

would be in place to teach the entire population of smartphone users what security risks are

and how to avoid them especially when we're already using like Microsoft and Apple and all these

MAGA allied and i mean so i think that there is there is like a sort of realistic argument to be

made that like you know education still matters and like people still need to think about what

you're doing and so i'm like none of that goes away but like it goes away to the level of the iPhone

like maybe you're gonna have like the iPhone with advanced protection or whatever enabled

where like you have to know why the image previews are not turning up but i think you're not going

to go very much further than that i'm sorry and like your alternative is no computer that's fine

i'm also happy with that but if like you know that is the space for operating in and i don't

think people are like oh yeah if everyone just learns a nixo s config language then like wherever

they can manage their own sovereign computer like no that's not gonna work like this is also like

tied into like your team as well as as you said with gnome y'all have a lot of

effort and emphasis made on the UI and interaction and how that is facilitated

but also within modal y'all have a team of people who are deeply technical and some of the most

skilled people in within the field for their specialties but y'all also have a lot of a lot of people who

are focusing on the UI and security and threat modeling and we talked about rabbit holes before

should we dive into what's this like what's this changed approach what does it mean in practice

to have a higher awareness of threat modeling when building these kind of networks or applications

so maybe like whatever for the listener like to take another four steps back

so you may or may not be familiar with our friend kdm and he's been researching sort of these

social technical issues around computing for a long time and that often includes like just sort of

novel ways of threat modeling things so threat modeling in general is just like the idea of

thinking very thoroughly about like all the different ways stuff can go wrong and how people can attack

a system or also like people using the system and so on using various properties of it and I think

kade has this sort of unique way of looking at the non-textual aspects of this stuff that then

often do turn out to be quite prescient about exactly how stuff is abused and often it's like

very common sense stuff and some sort of stuff that like no one's are coming but I think like in

general so at the base of it is just the fact that computers are everywhere and people do have to

use them for daily tasks that maybe before they didn't and especially yeah like in the context of

whatever like government requiring digital identities and banks and all these other stuff like this

is why he always criticizes the declaration of independence of cyberspace people where it's like

this is not a separate world this is not a people the world people live in and that means actually

taking that responsibility seriously and often stuff is designed in a way where it's like well

the network is just kind of like neutral everyone can just always access the network you know what

what could go wrong it is it's sort of downstream of like this this way of building technology that

doesn't really sort of take a lot of that stuff seriously and I think obviously it's it's hard

to take all that into consideration there's a reason why like famous with cloud service is

right it's way easier to just store everything in one server and then that's a canonical copy

cool we're gonna go back up and that's that's it but of course also yeah like all these other

problems and I think similarly here a concrete example of that way of thinking is the way that

we think about networking which is that ideally it should be like very transparent what is actually

happening by the network and there should always be the ability to make decisions about the power

imbalances that that are at play what's what's novel about it like how so I think that ties in

to like the walk away stack work that the dandrias and co-op been doing where the idea is that like

sort of as a concrete example with networking in a network application there are various points

where like even assuming that like all your data is encrypted there are various points where

metadata can be leaked in different ways for example if you're sending the data to another

person on the other side of town or on the other side of the world like your device is going to

first connect to a wifi network and the router is like then connected to an ISP and the ISP somehow

goes through I don't know various cables somewhere maybe some undersea cables that the NSA is tapping

or whatever and then on the other side it's the same thing and so at various points on this path

people like know your IP address if they know your IP address they kind of know your location

and there's all these like things connected to that if the whole thing goes via Tor

some things are gone right like then they no longer know your IP address but they do know that

someone here used Tor which is also metadata that also might be a problem you could say okay but I

only really want to sync with these 15 people who are like right nearby via bluetooth but then what

if there is a cop in the next room who's also using bluetooth and then like now they know all of your

whatever device identifiers and so there's always a power imbalance like sort of as the

person being surveilled or attacked you're usually less powerful than I don't know a state or large

corporate or whatever attacker but like the idea is that you can always choose those trade-offs

you're like okay I'm in a context where I think no one is nearby who is like a cop so I'm gonna

use bluetooth or I think like okay someone nearby is like I don't know I mean the domestic abuse

situation or whatever and like nearby is really dangerous but far away it's not a problem so like

then it's maybe fine to use Tor and obviously like as Kato is like to come back to it's like

really the only real safe thing is like people's physical threat model which is like just walking

around with a USB stick that's basically the same as like walking around and being just physically

apprehended and often that is like a really easy to think about model for like transferring data

and if you have like sort of a fancy database system that you can just like take a full version

of it no no you're speaking walk to the other person you want to sync with and plug it in and

it all works that is also a really valuable way of syncing and so I think that is that is sort of

this maybe full stack of possibilities that maybe all of them have trade-offs but at least you can

choose them and it's not like oh yeah you accidentally like whatever opened your computer and it like

did 50 sync connections to things and like now you've been fully docked in in every way but that's such a

tricky topic to explore like absolutely every single one of these things has 16 asterisks on it

I would like you to imagine those asterisks yeah exactly there's way too many and I was having

this conversation with Nico from Briar relatively recently and because Briar as far as I know is

one of the most secure and also lets you choose which route and you want to do and and what options

you want to have and what your own assessment of your environment is it's a peer for peer app

he was saying that it was such a challenge because on one hand he wants to enable usability but he

also wants to build something for as high of a threat model as possible so that the people who

really need it can use the application which is Briar Check so what I'm wondering here is like

I fully see what you're saying and I think it's it's whatever I mean I said with so many of these

things there's a fully unsalted research yeah a lot of this ultimately like needs testing you

build a version of it you test it I think what information we do have from the real world is that

often you come up with these funky solutions that then turns out in practice no one is like really

willing to use one example that I heard is that like there's this the city of sharding right we're

like oh yeah whatever as your backup thing you sharded among five friends or something like

dark crystal yeah yeah so dark crystal is a project that was initiated quite a few years ago

also in connection to scuttlebutt but it was trying to solve a wider problem for identity

protection in peer-to-peer networks so a lot of peer-to-peer networks have this challenge of how to

approach things when you don't have a centralized point of authority that can just send you your

password back and instead of passwords in peer-to-peer networks you often have key pairs

and and these key pairs represent a public key which is your identity and the private key which

you unlock your identity with so the question then is when you don't have a centralized authority

that can send you your private key back how do you then recover your identity in case of

lost identity and this is the question that dark crystal was tackling and the solution that they

came up with which we're then discussing a bit here in the episode is to break down your private

key and with mathematical algorithmic magic you can then send out these shards so-called

two different connections and friends that you trust and by reassembling these shards you

when you lose your key and by getting those shards back from friends you can then get your

identity back so that's the idea behind dark crystal to kind of find a way of identity recovery

in peer-to-peer architectures

yeah yeah and someone said that like in actual user interviews no one was willing to like do

this the problem is not like sharding your thing the problem was keeping other people's data safe

no one trusted themselves to do that yeah i mean technically as far as i've understood like

putting those private key shards back together you don't actually need all of them you only need

a few of them but you need to find at least a certain number of people who are willing to

participate but i mean i don't know right like often i think that's that's one point that

we came back to a bunch of times is that the iphone doesn't now have that advanced data whatever

thing uh i don't actually know what it's called to you it's like advanced protection or something

i don't keep track of those realms but anyway like there's this mode on the iphone where it like

doesn't download like email attachments or whatever automatically so you they can't like exploit

the png library as easily okay uh and a bunch of other stuff i don't know but i get it makes

some functional sacrifices in favor of better security while still running over apple

whilst exactly but while still allowing you to use your expensive iphone right because that's

realistic what is this for like a world leader uh who like is using the iphone world leader with

very specific faction i mean like i don't know like did you see this post at some point by like

the Dutch prime minister is like it is your duty as a public servants to reboot your phone every

two days hi i'm doing my part are you well we gotta raise the awareness of the people

and i mean like i i think by now they probably do do this automatically or at least on graphine

they do it automatically i assume some other like higher security whatever systems do this

automatically but i think like there is like some more tolerance than there used to be a few years

ago for like slightly more friction in favor of more security but at the end of the day what you

see in the field is still people like with the highest possible security needs uh using the lowest

version of security you could have in signal but like not verifying anyone and just like

randomly sending messages like whatever having groups of hundreds of people this is just how

people do it in real world but i think like i don't know at the end of the day like the only way you

really get to a better state is to actually try some things and yeah see what works in whatever

user interviews or then the actual field another approach i see that is the whole ecosystem one

especially when you have agnostic routing and you can route things in multiple different ways

you can build applications and plugins and extensions where you assemble the things that you need

in the certain case scenario and that's something that i think is inspiring but what delta chat is

doing with like the zip file application ecosystem but to circle back again this becomes like

it's a tour of a lovey conversation all of it uh you mentioned the walk away stack

i did i mean i feel like i'm not the person to to present that in detail because it like i mean i

come out all of this from the ux side i think andreas could speak to this much more all right well i

can speak to it a little bit at least yeah i i i can like maybe briefly outline it yeah um i think

essentially the idea is that there's a bunch of parts of like the the stack that you need

to actually get local first sync our sort of interchangeable hmm um can i before you dive into

what it is can i just give you where it came from oh yeah sure oh you maybe no more than i

potentially mostly because it was kind of a term that floated around a lot on scuttlebutt

and andreas and pyrtupanda and willow and all these protocols are offshoots from scuttlebutt right

and something that was roaming around in the culture uh was in part like this whole like

wordings that were used frequently from like community gardener uh to singleton to a bunch of

these terminologies and walk away stack i would assume stems from a book that's called walk away

by kori doctorow it's the solar punkish a little bit more dystopian book that describes this um

challenge of when a community becomes co-opted how to deal with that which is a very common

problem that often occurs in open source communities where a company says they're going to start

building something and have a really nice m.i.t license on it specifically m.i.t because agpl would

not allow it so then they um build this beautiful flourishing ecosystem and i know this very recently

happened for a community that i was talking to and they build this flourishing ecosystem and then

they just add this one part of the ecosystem that's not open source and proprietary enclosed and

no one can see it but it's essential for the ecosystem to function and then that's a very common

challenge that open source communities end up encountering but it's also just a common activist

issue like what happens when a community suddenly becomes governed in a different way that some

people don't agree with or alignment is broken basically and the approach and the walk away book

and how it like depicts it and very um sci-fi esque manners is to then talk about like walking

away from a project and how to do that so my guess is it stems from there

your guess is as good as mine underwording i i only know that like there was this uh seminar

last summer uh in some places somewhere where i think they like sort of conversion this and like

the idea comes from like um i'm bad with names uh some professor at the university of Basel who's

like yeah christian christian Cindy that sounds sounds about that yeah but uh i think the way

we've talked about is that specifically you have like sort of these three parts to a local first

sync stack um the bottommost one for like networking essentially like which they call the event

delivery then there's like a data type that like it needs to have the ability to convert at some

point um and then like event processing which is all the stuff above syncing and and so on and

none about that you built the actual app and i think the idea at least as far as i understood

it uh and uh i don't know again not not my expertise is that this gives you the advantage that you

can switch out that bottom layer um so like as long as um i don't know all the other stuff is there

you can send the data via laura you can send it via bluetooth you can send it via i don't know

like you speak email whatever whatever you need yeah and sort of that allows you essentially to

choose whatever your power imbalances to use i think the the phrasing that uh that they used

yeah and i think that that ties in nicely with the question about like whatever uh yeah like

do you accept more friction or not yeah if you are someone who like knows that you're not facing any

whatever threats or attacks or anything then maybe you don't care at all maybe just like always

you have the never con or it's all good if you are in a context where like that is the case uh but

then later uh i don't know uh you you encounter more power imbalances uh then you you can easily

like sort of without changing your tools without switching out the apps that you use and so on

easily be like okay fuck uh we're like i don't know only working via usb six um and like but it

doesn't mean that like you suddenly can't use your software anymore definitely and i think that's

also core to like a lot of peer-for-peer network designs like the whole routing agnostic as we talked

about earlier um and brier also has that uh as well um so in order to not go way over time

let's dive into the final part of this uh lovely conversation and interview um we've touched a

little bit about how y'all started and how you're doing the things you're doing but to dive into the

more specific aspect how do you organize like within model what is your structure

uh i would say still in progress so i think this is something we've talked about a lot over

the past year like how big do we want this to be uh like what actually makes sense for the goals we

have um and i think as we've sort of touched on there's like so many different goals that are all

like maybe equally important and uh people are working on them anyway and so sometimes there's

these opportunities for collaborations and so on but i think right now i mean basically we're

just deciding stuff by like rough consensus if no one's against something like you propose we do

this and if no one's against it and you're gonna do it you can do it yeah uh that's also like sort of

naturally just like these fears of influence if something is website stuff then the people who

are working on the website do it if something is like specifically visual design i tend to do it

um if something is i don't know like specifically questions around flat back like i don't know

we're we're currently uh like for example putting out together a grant application for flat back

that involves like the people who are doing grant writing and the people who are doing flat back

similar for system D similar for i don't know like the peer-repear stuff and so on

um we are working on a legal entity and so i think that's going to change some things but we're

also hoping that like once that is in place uh we can sort of have a clearer framework for how this

can grow like how other people can join because that is something we've not really like gotten

around to yet like sort of i don't know what what makes sense in terms of size because like people

already working in various communities and already doing stuff there that is very much

soft that we want to happen and we're already collaborating um but they're not like formally

part of our structure should they be do they want to be we haven't really like spent a lot of time

exploring that because again a hundred other things are happening but it's something that

we're very interested in and like explicitly uh discussing it's a tricky time also because like

so many things are happening in society yeah like yeah puts a bit of a stress stress aspect on it

that things have to kind of just start and run right like two years of designing the foundational

structures and everyone already has 15 other like whatever responsibilities right like we're

not coming at this because we all have so much free time we rather come at it because we don't

have any free time and uh we would would like to sort of ideally i don't know make all of this more

resilient going forward and collaboration helps that but setting up these structures is additional

work in the short term and um but i i we've made some good progress and uh yeah hopefully that's

gonna be ready soonish exciting well final how question uh how do people engage with you

you mentioned in the beginning that y'all have some workshops coming up yes are there other ways

of connecting with y'all how do you want people to engage with y'all and do you want people to

engage with y'all i mean yeah this is also something we haven't super discussed uh like

i don't know how we handle uh i don't know external communications and so on other than like the

master done account uh yeah you can always add us a master done we a bunch of us have the password

and whatever uh reads the the mentions in general like we've been mostly focused on the local

community here just because we organize these events every two months or so that's actually

quite a bit of work and we don't have that many other messages for other formats but we are doing

these developer labs now and actually i mean if you want to specifically engage with p dependor they

also have open office hours i'm not sure what it's called but it's like every few weeks you can just

like talk to them on on a call and for specifically reflection we have like these developer focused

sessions but that if you are a gdk developer sort of like that's maybe your focus otherwise

maybe the another format would be better but yeah we're also always happy about email or

other other ways of reaching out we're always interested in ideas to fund more of this work

because i think setting aside like to a specific group within modal like we're part of all these

communities where there's a lot of people with a lot of talent and no money and i think we also

see our role a little bit as we are some of the people who have done more grants work than others

in our various communities and we've gotten i think pretty good at it in some ways and so

something we've done in the previous years but we also want to do more of it's just like organize

more grants so more people can do the stuff they would have done anyway faster so i think if people

have ideas and pointers towards that i was always happy to hear that but yeah that's maybe a super

quick wonderful any final notes any shout out anything you want to mention beyond what's already

mentioned before we wrap up i don't know shout out to all the modal people they're really cool

and also i don't know our wider community right like we work with so many really cool people in

Berlin across the world in all these communities shout out to everyone and yeah thanks so much to

you for the conversation this was really cool and thank you for joining today's thank you bye bye

solarcast