ThunderCast #4: Will The Real moz://a Please Stand Up?

Ryan: Welcome to the fourth exciting episode of Thunder Cast. Mixed: Woo hoo! Ryan: We're fresh back from all hands. The Mozilla all hands. And we've got Jason, Alex and yours truly on the mic today. Ryan Sipes And this is going to be a fun one. Jason: So all hands for anyone who doesn't know it is a global gathering of the entire Mozilla organization gathered in Montreal or I think it was a five day yeah, five day event. And it it, it touched so many aspects of who like what Mozilla is.

And I don't even I'm still a little bit I've been telling people I am overwhelmed. In the most positive way by by all hands, because you go from being part of a 26 person team. To seeing a thousand of your, you know, a thousand of your Mozillians out there that, you know, just I don't even know what I'm saying. I'm not doing very well so far, but this is great. Alex: You're doing great. Actually.

Maybe we should give a little disclaimer to everybody listening here that we're still a little bit foggy from Montreal. Like we came back two weeks ago. Some of us, like mostly me, didn't recover from the time zone and the talking a lot every day and meeting hundreds of people every day. So this is going to be an awesome episode. We're so like perfectly in tune. We know what to talk about. Everything's going great. Jason: I do. I do have to call out.

One of the funniest things I heard on on day one or a day Zero. I think it was Ryan. I think it was Alex had had Ryan and Alex had gotten together with some people before I got into into town. And I just remember Ryan saying something like, That's what you guys look like from the shoulder down. Yeah. You know.

Ryan: Most of the Thunderbird team has never seen each other because, you know, the first time we even thought about planning something, just the Thunderbird team was right before Covid and Covid struck. And. And we never got we never got back to it, you know, after kind of not. Is there an after? I don't know. But like as the world kind of came out of that no travel period. And so yeah, this is the first time I saw pretty much everyone with maybe a couple of exceptions outside of a zoom window.

Jason: And what a difference that is.

I um I have a bit of a confession. I, I kind of swore to myself I will never work in an office again. Um, because I've grown so comfortable with the remote working lifestyle, you know, the flexibility and the hours and the location and. But honestly, now, if if we actually had an office where, you know, where the Thunderbird team could report to, I would want to do that 2 or 3 days a week because. It complete.

It really took me by surprise, but I enjoyed so much the camaraderie and the energy that was felt by by having the same conversations in person. And there's there's so much more productive. Than they are in a video conference. Alex: Yeah, it changes a little bit. It changes a lot, actually. Um, yeah. I think like in general, the hybrid model is the best model because you have the freedom and flexibility to work from home.

But meeting in person, especially when defining things from the ground up, when doing brainstorming. And that's what we did a lot during the all hands in our sessions.

We brainstorm, we prototype, we thought about the future implementations of Thunderbird and what we're planning to do, like being face to face and having also the amazing thing of being everybody in the same room from people that technically is very, very difficult to have a conversation with, like for time zone reasons, for lifestyle

reasons. There are some colleagues that don't get the chance to sit in the same meeting and having them in the same room and sharing expertise and sharing knowledge. It was like it was beautiful. There's nothing else. Jason: It was. It was beautiful. Beautiful. Beautiful is a good word. I mean, we're collectively from Canada, from the US, Switzerland, Germany, New Zealand, UK, Australia, UK. Ryan: Yeah, we're from all over the place.

And what I've what I said going into it and during it, it just proved to be true, is that super high bandwidth setting you can just cover so much ground. It seems like Zoom would be the same, but it's, it's just not something about being with somebody. And I don't know, maybe it's body language, maybe it's just everything. Something about it makes it so that so much more comes through.

And I feel like I, I it was like, I mean, time wise, it certainly wasn't thousands of Zoom meetings, but that's what I feel like it was the equivalent to is like, um, it was so just high bandwidth, you know, and you, you get to know people, I guess, you know, you never know what people are like in those moments when the zoom meeting. Is it happening? You know what I mean? Like before and even after a conversation, like we never really, truly greet each other. Um, and so, like, we just say hi.

And when you first meet someone on Zoom, you're like, Yeah, I'm from wherever and blah blah, blah. Um, but to see people's greetings, is it a hug? Is it a handshake? Is it a, you know, whatever it. Jason: Was like I think lots of hugs in our. Alex: Case. Oh, I forced everybody to hug. Like, I don't. I don't care. Some people's like, no, don't, don't hug me. I'm a bit sweaty. I'm like, I don't care. Come here. I miss you and things. So we were all.

Jason: Sweaty because Montreal is so humid all the time. I just okay, so I just wanted to remark that I appreciated that we all discovered more layers about each other. Yeah. You know, I realized in that room we were going around at the very beginning and we were kind of doing like an introduction, like we were meeting each other for the first time. And like, half of our team are multi-instrumentalists. Didn't know that.

Um, didn't know, for example, that someone does like classical ballroom dancing as a hobby, which is extraordinary. And you know, and other people have these, these, these fantastic hobbies or pursuits or passions. Um, and yeah, just those little things like seeing, seeing the humanity behind your coworker. Yeah. Ryan: And what our, what our users should know or our users and fans of the Thunderbird Project should know is everybody works on Thunderbird.

I came away and you guys heard this the last day I came away thinking how great it was that everybody on the team is an authentically good person. Um, I've, you know, we've all worked places where you can't say that about your entire team, but. And I guess that's why you end up working on you know, open source software that is not about profit or about, you know, insert all these different things here. But like, like we don't mind data all the, all the things we say every time.

But I like to remind folks listening but like that attracts a certain type of person who just wants to produce things that are good for the world and wants to and wants to leave a positive mark on the world. And so that's what I kept being blown away by, is thinking how good everybody was. Alex: Yeah, we're praising ourselves a lot during this. This is going to be like a wholesome podcast episode. Everybody like love each other, hug each other. No real information here. It was awesome.

And this is great. We love everything. Jason: Well, just the celebration. Like for me, it was it was it became a celebration of Mozilla's culture. And I wasn't I wasn't keyed in to just how amazing Villa as an organization is. Yes. Because I have met a total of basically people from Mozilla. And those three people are all on the Thunderbird team. Um, no, but like being able to just. Gosh, I don't even know where to. I don't even know where to start with it.

It's. It's being, being aware of being made aware of all of the, the projects that are underway for the good of the world, for the good of the Internet that are just ethically awesome, that are trying to, you know, remove bias and push boundaries and increase inclusion. And so many things that I, I didn't think of. Yeah. You know, I'm not I'm not a person who thinks that Mozilla is just Firefox, but.

I didn't see Mozilla beyond the products and now I see it as the like hundreds of humans who kind of like Ryan said, like are just authentically good people trying to make the world better. Ryan: Yeah, The what fascinated me was, you know, everybody knows kind of or a lot of folks because I hear it know the social aspect of Mozilla, the social good that it's trying to do in the world around privacy.

And and what I was surprised about, though, is the technical prowess that was displayed that goes into creating the things that promote the social good, the AI black box work, and trying to make it not a black box, trying to hack, trying to build tools that make it so that you can see and and test and I don't know, I guess just sanity test. Yeah. The different parts of an AI model are actually like truly important.

Yeah. And what one of the examples was like trying to figure out what layer the answer came from, you know, that you got from the LM. We all know ChatGPT sometimes you get untruthful answers, sometimes you get, you know, hallucination, true, real big hallucinations. Sometimes you get, um, not so much. Chatgpt is pretty good about this, but sometimes you get bias that it picked up from whatever source, Reddit or whatever.

And having seeing that Mozilla was building tools to let people inspect the layers of those models and then fix them is like a truly technical innovation and something that, you know, all these startups and even these bigger companies are just trying to build the most robust models, but not so worried about like all of the stuff that's hiding in that model that could be harmful.

And I was also struck by kind of the pan Mozilla and I say pan Mozilla meaning all the Mozilla organizations, how much they were actually in sync around values.

We're we're we work on Thunderbird exclusively you know we're different than MoCo that works on Firefox and some of those other products and the foundation Mozilla Corporation sorry Moco shorthand and the Mozilla Foundation that is working on kind of like the the stuff that that that you can't that isn't a product so you know policymakers you know they want to know if they want someone's opinion about a piece of

legislation that isn't Google or Microsoft who have usually an inherent they have a bias towards like protecting their monopoly over their respective markets that, you know, it's good to have Mozilla in the room to say well but you know this doesn't really benefit the end user for X, Y and Z reasons because, you know, frankly, we live in a world of monopolies right now in tech.

I'm struck by one conversation I had with one of their AI fellows. She is working on all sorts of interesting things in in AI. But I was I was struck by. How nuanced her thinking was about like, okay, you know, most policymakers, most. Consumers. Most people using the technology they don't actually know, like all the layers involved in this. So my job is to help explain when where these models are useful, where they fall short, why they fall short. What is the you know?

How are they built? What is what? An intrinsic like problems do they have and how does the world need to think about accounting for those and fixing them? Or doing so that we don't have like major problems? And I and this is a person, you know, I don't actually know if she has a master's or PhD, but she's she knows like this stuff. Through and through and through and thinking about not about how to raise another round of money or, you know, mind people's data to build a model.

But trying to think about how am I going to explain to the world. Both the good which he was really intent on saying like there is good both the good and the bad of this technology and how how can we educate people so that we can maximize the good and minimize the bad and someone with that kind of background can go somewhere and make a lot of money, a lot, a lot more than working for the non-profit Mozilla Foundation and utilize those skills to maximize personal gain.

But you have talented folks like that in Mozilla who are there to promote the mission. I'll say one more because I have a really good examples talking to the web compatibility team, working on Gecko and trying to make sure that it's everything just works. Those are people too, who could go work on the blink. Alex: Yeah, extremely talented and smart people that are beyond normal engineers. Ryan: Yeah, exactly. And facing incredible problems.

You know, not everybody is familiar with this, but I've seen a number of articles about now, years after the fact, about certain browsers who are also owned by companies, own popular websites and engineers, breaking those websites on other browsers in order to convince users to use their browser instead. So, you know, you go to your favorite search engine, for instance, and something doesn't work, right? And it says, Oh, this works great in insert browser here.

And, and so that's, that's reporting that I've seen you know is it engineers cop to doing that and that's really terrible but they're like yeah but it's the best way to get somebody to convert. Even if the website does work fine in the browser, if you make it work crappy on another browser, then at that moment the pain is high enough that they'll make the switch. The Web compatibility team spends a lot of their time fixing all sorts of web compatibility issues.

In a world where there's an extreme monopoly around browser engines. Google's Chrome engine being the the largest and by an incredible degree. Even browsers that aren't Chrome, many of them are using the Chrome engine.

And so Google has a high degree of say over the future of Internet standards, which sometimes is really not great, especially when it comes to privacy, as folks might imagine, to be fighting an uphill battle like that every day and to be trying to retain competitiveness, but doing it the right way, like how Google does.

It might not be the right way or Apple, you know, but to to meet those people who have an incredibly difficult job in the environment that they that they have but are doing it and they all I'm like of course I asked them, I was like, why are you guys doing this? Because you're talented. You could go work at one of these places. And of course, it was for the same reason. You know, they don't they think a web ruled by Google is or insert another monopoly.

Any other Internet monopoly here would be bad. And so they want to ensure that it's not just one company that decides what the Internet is and isn't. The great thing about Mozilla is I never heard have heard anything like we're doing this for, you know, because we think that it's worth a lot of money or, you know, like there's talk of sustainability, like, you know, you want to generate some revenue so that you can continue to work on the things that are good.

But it's always about like, how is this benefiting folks? How is this improving the Internet, How is this improving people's lives? Like that's what people talk about even all the way up to, you know, the exec, some of the members of the executive team that I met, like it's of a the foundation and the corporation, they're always talking about how is this good for everyone? How does this improve lives? How does this make the Internet better?

Alex: It's interesting. One thing that I heard a lot in the past week in Montreal was Mozilla is not a tech company. It is a people company that happens to work on technological issues. But at the core is a people company and we care about the people and we care about the ethical aspect of what we do and what we produce because it affects people.

I wasn't part of Mozilla in the past and I've been affected by the perception of what Mozilla is and that perception has been destroyed in the past week because I was exposed to all the things that Mozilla is doing and all the projects and all the teams that are moving forward, like the most common criticism is like Mozilla, just just do a browser, just focus on a browser, just fix Firefox, make it better and things like

that. The problem is that Mozilla and at the core of the Mozilla Foundation and Mozilla manifesto, those are the core rules that we abide in order to guarantee an open ethical internet. And for Mozilla to say let's just make a browser, let's just do a browser, just focus on that it opens up or it leaves monopoly space to everything else. It leaves monopoly for the ad system, leaves monopoly for the social media. It leaves monopoly for the AI.

So my reply now to the criticism is why not Mozilla, why Mozilla shouldn't be in those spaces and as they're doing for the browser where they offering an ethical and fair alternative that drives web accessibility and web standards and open internet for

everybody. Why the same core values shouldn't be applied to all these other monopolies that, as we are seeing in the past year, it was like more obvious than not that they're completely destroying people's lives and they're affecting your ability to work. They're affecting your ability to produce anything valuable. You're affecting your livelihood. Like all these companies, large corporations, they just don't care. They're just care about the bottom line.

And to ship the most groundbreaking super cool product, it doesn't matter if it actually doesn't work. And the thing that excited me a lot.

So first of all, I should be honest and say that at the beginning, before going to Montreal, I looked at the plenary and looked at all the speed, like all the talks and everything and the core. There was a lot about AI and I'm very skeptical about AI. I think it's not ready for the masses, I think is a hyped word that is just used to get more investors and venture capitalists. It's just it right now for me, AI is the same level of nfts and it shouldn't be

around. It should be just used as like, I don't know, treat it as a radioactive material. So I was very skeptical in seeing all this conversation about AI happening at the Montreal all ends. But then as you said, Ryan, all the conversations were actually about how do we destroy the black box aspect, How do we actually show the users and empower the users to see what's going on, reach the conclusion if this is a good thing or is a bad thing, remove all the barriers of Oh, these things just works.

Don't worry about it. The result is legit. It's fine. Like there's no validation, there's no moderation, there's nothing. Just create solutions for all these problems that are those problems that are affecting people. Plus, how do we actually use the good that comes from large language models and machine learning to create tools that can help the people, actually? And one of the few projects that really were exciting to me are like, how do we make these LMS as small as possible?

Like very, very small. And. Lean very focused on one task so they can run in client, they can run locally, they don't need to run on someone else's server.

Jason: I was really it's maybe a small example, but I was really impressed by better representation of underrepresented countries like Nigeria in large language models. You know, like, like if someone were to ask like, Well, what's my history as a Nigerian? Like there's no results because they're not represented by the corporation who is doing the training, you know, who's building the models.

Like little, little things like that are are so important and you don't really think about them in the conversation. Ryan: Yeah. And the crazy thing about that one was, you know, what was the core problem like the data sets that are out there, even though the language that was used, it's used by some insane amount of the world population, even though as a percentage of the population, it's like I don't remember how much a lot.

The representation on Wikipedia for articles, for instance, was like exceptionally low, like very, very low versus languages that have much less speakers in other parts of the world, especially Europe, that have much, much higher representation on Wikipedia and other forums and stuff where these eyes crawl, the or the researchers crawl for data and and feed that into the LMS.

It's just a missing piece of humanity that does that's representation in the models, doesn't match its actual representation in the world and as a world population. And so a good point, Jason. I had forgotten about that one, but. Jason: It's yeah, it's wild, right? Like, it's not I mean, I don't think people look at Mozilla and go, Oh yeah, that's something Mozilla does. Alex: Yeah, I have some data actually.

So the language is Kiswahili and the Kiswahili is an African language that is actually used like in Tanzania, Rwanda, Kenya, Uganda, like almost everywhere. But also there's a large population in the UK and it's a total like 100 million speakers worldwide that speaks Kiswahili. But the actual article and online resources are something around like the 0.01% compared to a language like Italian that has much, much fewer speakers in the rest of the world.

Jason: Because they because they don't have the influence that they need at the at the at the corporate table or whatever. Alex: Yeah, yeah. And that's the problem with these large language models. They get trained on publicly available data, so they get trained only on the data that the Western society just considers the standard. They only get trained on English, they only get trained on white people pictures. They only get trained on all these bias.

And that's how you get like AI bias that don't reflect the actual population and make these tools that might be extremely useful, not accessible to 75% of the world, which is insane. Ryan: Yeah, it is insane.

And and they also have the Common Voice project, which my background touches this a little bit because I actually spoke to the folks who who started that project way back when, which is just another layer to this, which is when you talk to Alexa, you know, or any other of these AI assistants, I, I use air quotes for that. In these, they, they speak back to you in your language and it sounds pretty good.

And that's another thing where we have lots of data in English, for instance, to create these text to speech models that sound correct. But for languages like that one and that you mentioned, Alex and others, you you can't have text to speech, which is sometimes very helpful, especially in in some parts of the world where illiteracy is a problem. So if they're trying to use a tech tool, you know, most of our tech tools are very what's the word like text. So we still have text interfaces.

And if you're trying to learn information but you're suffering from illiteracy or something, it's very hard to to gain any of that information. Now, if you're able to talk natural language, talk to a device and then have it talk back to you. You can gain access to that information, but you can't. If the text to speech model can't actually speak your language. And that's where we saw a demo of a of an application that that shares some farming information that's being used actively.

And I don't remember where and I feel I feel bad for that, but I can't recall exactly where the application is, is in use. But essentially you, you ask it a question about farming practices and it shares back with you in audibly in, in that language the information just out of personal experience.

When I was working on a the same type of tool that the open source Amazon Echo before Amazon Echo was around, we often got that feedback that there were there were folks who were excited about it, but they were like, It can't speak my language, you know, my my mother tongue, I guess, and asking if we could support that language. And the answer at the time was, that's very difficult because the data we have access to doesn't include examples of that language.

So we can't train a model without the data. And it was very hard to track that data down. And Mozilla has done that clearly with at least a few of these uncommon languages. And you can actually see that work in their Common Voice project where they have asked many, many people to record that data so that they have it. Jason: Did you did you both come out of It's going to sound like we were at an AI summit the whole week or something, but that's not the case.

But it did it did make an impression because I think we all came out of all hands feeling less pessimistic. Yes. About the future of AI and more, you know, more hopeful. Yeah. It's that AI is not all about just. Just crunching data to, you know, to sell advertisements or to sell products or to productize monetize. You know, there's there's so many things that can be done for the good of humanity, really. Alex: That's what I felt at the end of the week.

I always consider like the current state of technology is like the accelerated capitalism of technology where something new, new something that appears to be new, comes out and then is digested in three months and it's exposed to the users, gets to the top of the hype, and then it collapses and it gets completely annihilated because all the major corporations tried to monetize it before it was ready and before it was actually useful to anybody.

And in the past ten years, we saw the these hype cycle how potentially very good ideas and good solutions were completely destroyed by the fact that this needs to be finished and monetized and shipped to the public and consumed and now it's not ready and now it's creating problems. So it goes like it completely gets tainted, the reputation gets tainted, and now nobody is using that. As soon as you mention it, it's just, oh yuck. Like the core concept of the blockchain was a good concept.

And now, like everything is attached to crypto bros and scams and it. Jason: Has been tainted. Alex: Yeah, exactly. And we will see these happening already is happening with AI. Chatgpt is just getting worse and worse because now it's training all the absolute trashy data that no one is had any like moderation about on top of that. So I was very pessimistic about this whole thing.

But seeing how the pragmatic approach of Mozilla and how let's take things slow, let's treat these as what are the goods, what are the bad? Okay, these are the bad. Let's try to fix the bad and expose the issues and find solutions. And then the good parts. Let's actually use them for the good and the betterment of the world.

And even the simple thing that Mozilla allows you to submit your data to train these model and you retain the ownership of the data, we don't track your ID, we don't just digest or steal anything. It's all ethical and in the open things that no one, all these large corporations that they're shipping, these AI tools they even thought about or they cared about it. It just see the internet as an open bucket. They can just grab and just use anything and anything is available for everybody.

Just already that simple thing of establishing some rules to allow users to retain their data and use it if they want to train a model and help is it's great. It just gives gives a little bit of hope and confidence.

Jason: I want to bring this around the Thunderbird. I've got a question for you, Ryan. As you know, as the guy who is kind of responsible for thunderbird's long term vision and the prolonged sustainability of Thunderbird.

I'm wondering if you came out of all hands with a slightly different perspective on Thunderbird's future, what Thunderbird's future should be, if anything, any of the conversations you had or any of the meetings that you did, if that altered in any way the direction that you want Thunderbird to go in, if that makes sense, or gave you ideas for future features or anything like that. Ryan: You know, I'm different than you two in that I had a lot more visibility into the rest of

my life. I'm very fortunate that I've gotten to have a lot of conversations with a lot of different teams who presented work there. And so and I'd love to see more of that for for everyone on the team and frankly, for the world. You know, the world should see more of what of what they're doing and understand more of what they're doing. Being introduced to the work before all hands had changed me a little bit at all hands. It just reinforced. A feeling of we can do this.

I felt that way about us specifically. You guys all know that I have a high degree of faith in the Thunderbird team and and know that we can execute on the things we want to do and know that we all have the right values that guides how we do our work. So I don't have to worry if somebody is doing something shady because I just know that we're all values aligned on on, you know, privacy and these other core pieces.

You know, I, I need to find a positive way to say this, but I'll say it the, the reverse positive way, which is like so many applications, they talk about stickiness and I talk about it sometimes, too. But when you talk to folks who are at these startups about the stickiness of an application, they're always talking about how do we get the user to stay in the app just any way?

And usually that results in dark patterns like how do we we need notifications, we need all this stuff jumping in front of their face so that they're just like glued to it. Because if we can show that a user is on this four hours, five hours a day, we can raise more money, we can sell that to advertisers, etcetera, etcetera, etcetera, etcetera. We don't do that, you know, and we that's not even in our universe of like

considerations. So anyway, the point is, is that everyone on the team avoids those type of dark patterns and we stick to like our core values too. So there's two pillars, you know, make up the Thunderbird team. What I saw in the other teams and in my conversations with the with the members of those teams was the same kind of the same values. You start talking to them and they're they know what they're doing. They know how they're going to do it.

And like, I felt like there was a real chance of impact. Why has Mozilla fallen behind? Why is Firefox? Why? Why do we not have the 30% market share that we had in the early 2000? It's not a lack of trying. You know, there have been market forces. There have been the ability to leverage monopolies in order to take over people's experience of the Internet platforms. I was just saying today in Windows, I don't know what they call it. It's some kind of pain on the left hand side.

It has the weather. When you click it, you see like news and blah, blah, blah blah, blah. It does not respect any of my defaults. If you click something to send an email there, it opens outlook. If you click something that opens a web page there, it opens edge. Even though my defaults are of course Thunderbird and Firefox, that's something that Mozilla as a whole is fighting and will continue to fight. But what Alex is saying, there are other fights for the soul of the internet.

And talking to folks, I was convinced that there's not only like a a good that can be done there, but there's also opportunities to make a difference there and to do things for the good of Mozilla. What I came away convinced of is that there are actually opportunities for us to show the world that there can be a better Internet and those are achievable. Like we can actually make a difference there. We can. We can.

There are moments and we're living in a moment where social and I especially are two places we could we could improve the landscape and produce tools that people find useful and that are good for them. Social is probably my big takeaway is I was aware of what was going on in social. And of course, you know, I'm not sharing anything out of that that shouldn't be shared. Like The Verge has reported that Mozilla is looking into social instance on the Mastodon network on the Fediverse sorry.

So on the Fediverse, there's Mozilla social right now and it's in a closed beta seeing the work that that team was doing and seeing the competence and the professionalism they were bringing to that work. I was thinking this is actually both good for the Internet and a good product. It looks like a really great product and I'm excited to like encourage family members of mine who have grown tired of the world of meta and x x. Jason: The social platform formerly known as Twitter.

Yeah. Ryan: The that there's there's going to be an alternative that's approachable and focused on on good interactions and not be blanketed with ads for like for me it's like weird ads that I have no interest in on on x. So sorry folks, if anybody loves X who's listening to this, but like they're all like things I have never expressed interest in on the internet, it's like survival pack. The world's coming to an end. You need a knife, a gun or a crickets that you can eat.

Like how did they decide that, that this was what you needed? But anyway, I didn't answer your question at all.

Jason That's okay. Does it does it influence my idea of where Thunderbird should go other than wanting to take some of that expertise from other parts of Mozilla and incorporate it into the Thunderbird product? Because there are things they know. Other parts of Mozilla know how to do better, like especially the AI stuff. Mozilla I there's all these integrations going into Firefox protecting folks from scams and other things on the internet.

I would love to leverage that expertise and help make something good in Thunderbird. Overall, I just came away feeling like Mozilla can be healthy and have an impact on the world, which I think is. When that's the case, that's a better world that we can live in.

Alex: There are a couple of things that I want to add to that in terms of like the same question that you asked Jason, I there have been discussion around for the past few years about, hey, what don't we integrate some AI tools into Thunderbird to help speed up your inbox or help translations and things like that? And I was always the first one to say, no, absolutely not. We're not going to don't even mention it.

We're just going to destroy our reputation because of all the problems that I've been exposed by these large corporations. They're heavy, unreliable. They're always they need a remote server. So your all your data, you just send it. You don't know where it goes. You don't know who reads it, how it's used in which bucket it ends. If it's going to be used for other training sessions, who knows, right?

Like it's all a massive black box, but now with the conversation that we had and all the products that Mozilla is doing and the effort that Mozilla is putting into these into these these section, I have the confidence that we can actually do useful things. Even the simple exploration of let's make a model the smallest possible target and only on one single thing that can do right and can be run in client. And it's so small that it doesn't destroy your CPU. Hell yeah, let's absolutely do it.

That opens up so much for Thunderbird in terms of what kind of feature, what kind of extra things that we can do that yes, our competitors will have because they will do, but they will not never be able to claim that they do it as good as we do in terms of ethical respecting and privacy approach for your data. Jason: They'll never be able to claim that it's not a black box. Alex: Exactly. Yeah, we are. We will be able to show the user this is what's happening behind the curtain.

If you want to see your we're reading your email to give you a summarization or give you a translation. The data stays in your computer, the data stays in your client. And if you close it or uninstall it, it just goes away forever. Like we don't keep any record of it. No one will be able to even remotely claim any of that because it's not in their interest. They only need to make money, so they need more data. That's the only thing.

Jason: I feel like that's that's really at the core of most people's privacy concerns is, you know, whatever kind of product we're talking about or whatever kind of software or service. The core problem is always my data going out into the cloud. I'm using air quotes for the cloud. You know, it's really going to into a server and not having full control over. What happens to it and not really not understanding.

Even if you read the 35 page terms of use and privacy agreements and things like that. Not even understanding like I think I think actually. So Mozilla has a blog called Privacy not Included where they review, you know, products and services to to analyze its. Well normally analyze its lack of privacy and raise some raise some red flags and warnings and things like that.

And I think the most recent newsletter they sent out said something to the effect of, you know, we asked nine privacy experts about Microsoft's new privacy policy as it concerns as it as it relates to AI. And none of these experts understood it. So how do you stand a chance of understanding it? Ryan: This is truly different because a lot of these companies are using these the data that they collect through interactions with you and your data.

So when you talk to an LM, when you interact with whatever data that AI model is touching, it's your data. Oftentimes there's a clause in there like we reserve the right to use any of this data in order to train future models. And sometimes for other reasons. And so and that's that's the core difference, isn't it? You know, we went to all hands and what we learned about the rest of Mozilla's is that we all carry that as a core tenant, which is like it's just kind of a no brainer, like.

Hey, you use this product, you use this software, you interact with us in this way. We still don't own your data. We don't own you. You are not the product. The I was very encouraged by the fact that there was an acknowledgement of not only do we need to make these things, but we need to make them impactful. We need to actually get them in front of people and get and show there's something different. There's another option because I think normal people, they're past this at

this point. They're like, the options are all crap. You know, I don't have any other options, so I'm just going to I'm just going to stop thinking about that. I'm going to stop thinking about Facebook, Instagram. Are they grabbing all of my data? You know, we know how many permissions they ask for. We know. Mixed: All of them. Ryan: From all these.

Yeah, exactly. And we know from I don't know, I wouldn't call them leaks, but reports on Facebook data usage that it's all used like it's used, it's all sold. Like you've got the Facebook app on your device. It's tracking where you're going and it's selling that to advertisers. You may not think about that, but that's what it's doing. So like, yeah, people just accept it because they have to accept it because they, they feel like it's the there's no other product in that space that they can use.

Et cetera. Et cetera. Et cetera. So I was excited by the fact that I thought all these efforts we heard about are grounded

in like, reality and have a chance to shape the market and people can use them. Alex: So nobody liked the Montreal All Hands. Basically. It was like it was awful. It was awful. Yes, it was. Yeah. Jason: Another confession, I guess if you want to if you want to categorize it as such. I don't feel like I'm working in a silo anymore. Alex: I don't.

Jason: Yes, I and I don't mean that in terms of I'm over here doing my marketing and communication thing and you guys are over there doing your engineering and your UX and you know, but I mean, I came out of there wanting to devote some of my time to promoting other Mozilla projects besides Thunderbird. Like I feel really part of. I'm really inspired now to, to try to affect some change and some education and just try to make a difference beyond just, you know, our work here at Thunderbird.

Like the, the larger picture. Um, I just felt, yeah, I don't know, it's just warm fuzzies. That's going to be the name of this episode. Yeah. All hands, warm fuzzies. Alex: It felt the overall week. It felt invigorating, refreshing. And it gave me I think it gave everyone a more hopeful view of the future. Currently, the Internet looks more like a dumpster fire every day. Jason: There was there was a comment made at one of the sessions. The session was called AI Boot Camp.

And, you know, he made a good point. The presenter, he he says that we tend to look at things like AI from the lens of, you know, post-apocalyptic world. And Yes. And, you know, all of the negative negative outcomes because as as humans. We tend to our technology, our creations tend to mimic the fiction. That we read and most of the science or that we create, you know, movies, whatever. And most of the science fiction revolving around AI is super negative, but it doesn't have to be.

Ryan: I remembered. So I didn't go to that talk, but I remembered you telling me about that Jason and it shifted me a little bit. Just hearing kind of that take in some of the examples you gave of that session. It's good to remember that technology is neither good nor bad. Usually it's both in its, in its implementation, in its, in how it's rolled out into the world. And that's because you have people leveraging it who have both good and bad intentions.

And so we shouldn't shy away from a technology that is. You know, coming of age and that that has good applications because we know that so many people are going to use it for bad or overhype it or oversell it. Like the blockchain is like neutral.

Like it's a great way to have a distributed ledger, you know, And if you can look at it like that and not think about, you know, the good and the bad ways in which it's used, you can just say like, Oh yeah, that's a great, interesting piece of technology that might have some purpose, which, you know, with AI, if we can save people time around managing their inbox, if we can provide essentially democratized like the an executive assistant

who helps you sift through all the email and figure out what's worth the most time, what's needs your attention, that's good. That's good. That's a good application of AI if it actually works, if it's not just hype, if it's actually is able to surface things that need your attention, if. Jason: It's, you know, if it's under your control and it's running locally and and it's personalized, personalized for you. Ryan: I will admit I wasn't going to admit this in the podcast, but I'll admit it.

I went in with some level of skepticism just in general, like like Alex, I saw some of the talks. It's not that I didn't believe Mozilla could do make a positive impact in these spaces. I kind of had a cynicism around the fact that I, I was worried no one could make a positive impact in these spaces. And as I watched and met these people, I realized, you know what, Actually, I think they can. I think we can, you know, because it's very clear eyed.

I don't you know, without sharing too much, I think everyone in the in the whole event was very clear eyed about where Mozilla can actually make a difference. And things that are just because of the players in the space are maybe too big to have an impact right now. You know, I think there was a really just like I said, clear eyed view of like where we can have impact and where we can make a difference. And and I'm so glad I'm so glad that my cynicism and skeptic skepticism was just

eroded. And now I'm just like you, Jason. I'm just excited and want to tell people about all the cool stuff that I saw.

Alex: I think this is the perfect note to end this episode, I guess. Jason: Yeah, I think. I think so. Alex: I think positive and warm. No negativity, Warm fuzzies. Yes. Jason: So yeah. So that's going to do it for episode four of the Thunder cast. Thank you as always for listening and thank you for tolerating our tangents and rabbit holes and enthusiasm. Yeah, because you know, it's not always about talking about Thunderbird. Sometimes it's about embracing.

The culture of the company we work for are just talking about issues that matter to people like you, people like me who want to see the the internet become a better place. If you want to keep up with us, if you want to see what we're up to, you can visit our blog blog at Thunderbird dot net. You can follow us on a whole bunch of different social platforms like Mastodon, and we have matrix channels.

If you need some community support or if you just want to hang out and talk to the community about whatever you want to talk about. If you have a question for us or have a suggestion for the podcast, you can send an email to podcast at Thunderbird dot net and I think that's going to do it. And oh, do we want to tease anything for the next episode? Ryan: Well, I think there's some fun stuff to tease for the next episode.

I talked about our future plans, which some Thunder Cast listeners will be familiar with. Some of those plans that we've announced. And and we have even more in the works trying to make email better in the world and stay tuned for that. I hope by next episode we have some some really good details to share on that. Otherwise I would encourage folks to, if they haven't already updated to Supernova to go and grab Supernova, you can download Thunderbird. Net next time.

I think we should spend some time talking about what was good about ye old internet. We've touched on it before, but that's my topic. Submission is like we should talk about. Mixed: Okay. Yeah. Oh yeah. Alex: I have like Italian internet experience so that's going to be a culture shock for everybody. Mixed: Yeah. Ryan: That's great. Perfect. I look forward to it. Mixed: Yeah, so me too. Jason: Me too. Mixed: Cool. Jason: Alex Ryan, thank you so much. And everybody out there listening.

Enjoy your day, enjoy your week. Take care and take care of each other. We'll see you next time. Thanks for listening.