LINUX Unplugged - 401: Own Your Mailbox

Episode Date: April 14, 2021

Do as we say, not as we do. This week we're setting off to host our own email. We'll cover the basics, what's we're using, and why. Plus an update on Jupiter Broadcasting going independent, community news, and more. Special Guest: Martin Wimpress.

I think Pipewire just landed the killer feature that's going to make every desktop Linux user want to transition to Pipewire. What is it? In version 0.3.25, you can now use your phone as a microphone to your Linux box through Bluetooth. What? Yeah, and that's an awesome feature. I think that's something we need to try. It makes it now possible to use A2DP as a source input device. So you can then use your phone as an A2DP microphone.

And, you know, Wes, like some of these phones actually have decent microphones on them. This could be like a seriously great way to get audio from a guest or somebody. You're just tired of dragging that remote mic around. Yeah. You think Joe will let us get away with using our phones for doing the show? I think we better try without telling him. Hello, friends, and welcome into your weekly Linux talk show. My name is Chris.

My name is Wes. Hello, Wes. And can I just say, your new 400 after-party face tat you got is, well, it's going to heal real soon. Aw, thank you. But you know, you still owe me. You said you'd paid for it if I got the show's logo on my face. Right. Yeah, we'll get to that. But first, I want to thank ACG for sponsoring this episode of Linux Unplugged. They are the leader in learning for cloud, Linux, and other modern tech skills. Hundreds of courses, thousands of hands-on labs. Get certified, get hired, get learning at acloudguru.com.

Well, coming up on the show today, it's the show that wasn't supposed to be. We were going to take the week off, and we thought about it a little bit. We thought, well, actually, there's a lot to catch up on. We have a lot that we want to cover. There's things from 400 we still have to talk about. And so we're going to kind of have the no-show. It's like a no-show this week. From the wackos that brought you running Arch on their server, we're going to talk about hosting your own email server. This is another one of those don't do as we do kind of episodes, unless you

really want to. We're going to cover some server basics, what you need to be aware of, and how we're going to try to make it all work smooth for us and what our use case is going to be for self-hosting our own email. But we'll get to all of that. There's a lot to take in there. The truth of the matter is we just have a lot going on and we wanted to do something kind of fun with you guys this week. And we'll take time off in the near future. We'll take a week off because we do have a great interview all about an individual who didn't know anything about construction, didn't know anything about diesel engines, didn't know

anything about Linux, didn't know anything about automation. And he went out and bought himself a school bus and built it, rebuilt it from the ground up as a mobile home that he worked from with his family on the road, all powered by Linux, using automation to like manage the tanks and the batteries and all of it. And so it's a great conversation. If you want inspiration or, you know, just a little, like, you can get your project done, that's the interview right there. Yeah, it's so inspiring how somebody who didn't know something just decided, like, so what? I'm just going to go learn it. And he just, at every turn, he had to just go learn something. And even when it came to, like, setting up a Linux server, he didn't sway.

He just said, you know what? I've gotten this far. I'm going to learn it. So we will run that story in the future. It's just today didn't feel like the right time because we had a lot we want to do, and we just had an episode that was honestly a little off topic. So today we're just going to have some fun. So before we go any further, let's bring in our virtual lug.

Time-appropriate greetings, Mumble Room. Why are there 25 of you here? We told you for two weeks in a row we weren't doing a show today. I don't know if they weren't listening or they were listening when we said we were going to do a show, like today? I think everyone's still drunk

on beer from episode 400. That's what I think. I am. You know, speaking of episode 400, the votes are in. Our special 400 brew was named by the community. It's officially named the See You Next Tuesday beer. The See You Next Tuesday.

And I'm drinking a little Tuesday right now, Wes. Let me tell you. Oh, you got things plumbed up again, huh? Drinking it in my beer stein too, Wes. In my brand new 400 beer stein. If you ordered them early, they were already starting to arrive. Oh, that's great. It's actually pretty special in a way.

You know, I mean, I'm sitting here in my studio drinking a Linux unplugged beer and a Linux unplugged beer stein. Talk about self-hosting, right? Right. And we're doing an episode that, because we just wanted to. We could have taken the week off.

We said we were gonna, but we just kind of wanted to. So yeah, the beer steins are shipping, which is really great. Yours is here, Wes, waiting for you. Oh, great. I've been jealous. I mean, I've already seen a few community members

posting theirs on Telegram or on the Matrix server, which is awesome. Please keep doing that. They look good in your hands out there. That's true. I do love seeing that, and I think they do look good. I'm really happy with how they turned out,

and I'm glad that they're shipping right away because Coda Radio listeners will know that I recently had my butt kicked by a robe. I was schooled by a robe and didn't like that very much. It's been a minute, Wes, since I really was schooled, you know? I went to the hard knocks of business school for a couple of weeks. I think we might need a sandboard clip for that one. Really? It was bad. But I, you know, I did learn some fundamental lessons. I'll say that I just learned them the really, really hard way. And so towards the end of the show today, after we get through the news and the email server stuff, I'm going to share some, some very new and first steps that I'm taking to grow Jupiter Broadcasting as a business.

It's kind of been just about a year-ish since the plans to go independent were kind of beginning and the thought and the process was starting. And so I want to kind of do a bit of an update for you all and share that with you. We have a couple of big updates there and some lessons I've learned, some humble pie that I have to eat and all of that. But we'll get there. But first, let's do some business and let's get to the community news because we have an update from Google. It seems Google has become concerned about the correctness of code,

as they put it in the Android platform. And they say correctness of code is a top priority for the security, stability, and quality of each Android release. Unfortunately, memory safety bugs in C and C++ continue to be their most difficult-to-address source of, quote, incorrectness. Not bugs. Yeah, here we're talking kind of about the Android underpinnings. You know, not like the Java or Kotlin or whatever framework you're using to develop your Android apps, but all the stuff that runs at the lower levels. And we should probably give them credit because, I mean, Google and Android developers more widely do invest a lot of effort and resources trying to detect, trying to fix, or just mitigate this type of bug up front.

And those efforts are pretty effective. I mean, yes, bugs get through, but a large number of bugs are prevented from ever making it into an actual Android release and then from me complaining about it on air. But memory safety issues are still a big contributor, even given all of that. Something like 70% of Android's high-severity security vulnerabilities are due to memory safety issues.

Oof. Yeah, that's not great. So to help combat this, Google is excited to announce their words that the Android open source project now supports the Rust programming language for developing the operating system itself. Rust?

On my phone? Wow. Google notes that adding the new language to the Android platform is, quote, a large undertaking. No kidding. I mean, that comes alongside Java and Kotlin too, right? They say that there are tool chains and dependencies

that needed to be maintained, test infrastructure and tooling that had to be updated. And then here's a big one. Developers needed to be trained. And for the past 18 months, Google has been adding rough support to the Android open source project and has a few early adopter projects that they'll be talking about. They say scaling this to more of the OS is a multi-year project. So stay tuned, and there will be more updates from the Google Security blog. And didn't we see that the Bluetooth stack, Wes,

was one of their first kind of go-out-there-and-fix-them projects? Yes, it was. You know, I hope Mozilla feels secretly good about this. You know, I mean, I think this is Google admitting, Rust is good stuff. And they're not just talking about it in, a small way, they're talking about it alongside Java and Kotlin. Yeah, I mean, it sounds like this is going to be a first-class language you can use to implement whatever features you might need in the base OS itself.

And, you know, at least once tooling is up to speed, once all the build infrastructure is there, but that's a significant investment and endorsement. I love that, you know, we kind of put our eye on this about a year ago, that the Rust hype was really building, and Google's been, for that entire time, working on this. Also, speaking of Rust, listener Mike Kelly was just looking for an excuse to play around with Rust and wanted to celebrate Linux Unplugged's 400th episode, And so he combined a little bit of Rust with a fun project and created a mini website at lup400.com. And you know what, Wes? It's all powered in Rust. Hey, and it sure looks better than any of our websites. Right? I didn't really realize you could develop websites in Rust.

Tells you what a noob I am. And then the other thing he did, which is really adorable, is he embedded episode one, which nobody should probably ever listen to, in there. So, Mike, I guess we should get in contact. It's time for you to rewrite the entire Jupyter Broadcasting website in Rust. How on brand would that be? That's what we need. Thank you, Mike, for doing that.

Yeah, we'll have a link if you want to check it out. Of course, it's open source. How great. So this episode we thought maybe, and maybe one of the reasons to go live, is that we'd have an update on that Linux-powered helicopter on Mars named Ingenuity. It was set to take its first test flight, and then by just about now, as we'd be recording, we'd have the return data from the surface of Mars. But it turns out that helicopter flight was delayed during a high-speed spin test of the rotors on Friday. The command

sequence controlling the test ended early due to a watchdog timer expiration. And as you can imagine, those watchdogs are there to catch things that aren't working very well, and so it triggered a process of stop, we have to diagnose and figure out what went wrong. But Wes, it seems like maybe they actually figured out what the failure was. Yeah, that's right. I mean, at least so they say. The Ingenuity team has identified a software solution for the command sequence issue. I guess over the weekend, they basically considered and tested all the ideas of what this could be and how they could address it.

And they concluded that a tiny little modification and reinstallation of Ingenuity's flight control software was the most robust path forward. Although that sounds pretty tough when this thing is on Mars. I know, it's like, there's nothing straightforward about reinstalling a software package on a tiny helicopter in the belly of a rover. Actually, no, it's not. But you know, like, there's just nothing, nothing simple about that. And that's why it has to go through extensive testing. The software update will modify the process by which the two flight controllers boot up, allowing the hardware and software to safely transition to the flight state and meet the time expectations.

Yeah, and so it sounds like they actually know what this change is. It's not going to be that big of a deal. You know, they've got mock-ups here on Earth. They've definitely experimented with it before, but they've actually got to transmit all those updates all the way there, you know, to the rover itself, get it transferred over to Ingenuity and then actually run the update command. And I mean, we've all updated things on servers and sort of waited for the prompt to come back. But can you even imagine? No, Wes.

No, I am not the personality type for this kind of stuff because I get anxious when we reboot the Arch server in the garage. Yeah, I think this is one of those measure twice, patch once situations. Yeah. So, okay. So they have to get it tested. They have to get it qualified, as they put it. And then they have to upload it.

And then it has to install. So they don't really have a new test flight date. They say our best estimate is a targeted fluid date. We're working towards achieving these milestones and we should be able to set a flight date next week. So possibly by next week's episode, we'll have an update on when the first Linux powered helicopter will fly on Mars. Here's hoping. And really, we should expect these kind of things, right? This is all brand new science. A few things are bound to go wrong.

new science, a few things are bound to go wrong. Linode.com slash unplugged. Go there to get a $100 60-day credit towards a new account, and you go there to support the show. Linode is our cloud server provider. I was just messing around this weekend. I actually love it for setting up multiple servers and then trying out different distributions and different software packages, and then picking a winner, and then that's the Linode that remains standing, and I just destroy all of the other ones. They're the world's largest independent cloud computing provider,

and they make it easy to get your creation on the Internet fast, no matter what your skill level is. If you're brand new to Linux, you're brand new to servers, or if you've been doing it for a decade, they've got a technology stack that'll work for you, and they can help your ideas come to life on the web. And if you run into any trouble, Linode comes with amazing, outstanding 24-7 customer support by phone or by ticket. This is like the number one thing I hear now from our audiences. Man, I was

in a jam, and Linode stuck with me through the entire thing. I've got one $5 a month server, and they still help me out. On top of that, they have hundreds of guides and tutorials to help you get started and dig through what you need to know or understand some terminology. And they have things like how to secure phpMyAdmin if you're going to use that, because honestly, I understand it's a good tool, but you got to be a little bit worried about putting that out in the public. They have guides to help you manage things like that. So that way you have some confidence, some peace of mind. That's what's great about their guides. They give you that kind of confidence to put this stuff into production. And they have

easy to use one click applications for like a LAMP stack or WordPress, but also some other really kind of really more sophisticated and complicated things to set up from gaming servers to entire e-commerce stacks. And with S3 compatible object storage, you can find all kinds of great usage for things that don't require a server but still need storage in the cloud. They have super fast networking and 11 data centers worldwide. Plus, they support shows like this and events like Linux Fest Northwest, All Things Open, the Kubuntu project, a lot of Linux media out there. I mean, they're a pretty special company, and they have a great service. They've been around since 2003, and they've really dialed it in like nobody else. And now they're just dedicating to make sure that if you want to run something on Linux,

it's not going to run anywhere better than it runs on Linode. It's going to run fast, it's going to run stable, and you're going to get a bunch of great information and fantastic customer support and a company that loves Linux behind it. And you can support one of your favorite podcasts at the same time. So go to linode.com slash unplugged. Well, it looks like someone spilled beer all over in here. I think it's time we clean up, Chris.

Yeah, we could do for a spot of housekeeping around here. So I just want to mention that if you want to get a little more show, one of the things you can do is subscribe to the membership program. That's a great way to get more show. We have two feeds available, one that is a limited ad feed and one that is the full everything we did live feed. I mean, sometimes it's two hours long, but there's a lot of content in there.

We try to make it entertaining, and we make it available as our members as a way to say thank you. So if you'd like to become a member, go to unpluggedcore.com, keep the Unplugged program independent, and reduce the ad load needed to make the show actually sustainable.

And check out jupiterbroadcasting.com. We've got a batch of great shows, like a bouquet. Yeah, there you go. A bouquet of podcasts. A bouquet of podcasts over at jupiterbroadcasting.com. And last but not least, join the LUP Lug on Sundays. They get together. They self-organize and they hang out and talk about, well, things you love.

That's in our mumble room. You can get details at linuxunplugged.com. And then you join Sundays, noon Pacific, 3 p.m. Eastern, and it's just right there in our Mumble lobby. And the great thing is, then you're all set up with Mumble. And you can come to the live show if you get a Tuesday. Maybe, I don't know, maybe you get the Rona. You want to take some time off from work. They're not going to ask any questions, you know? You just come show up in the Mumble. We won't ask any questions. I mean, how do you think there's, now there's nearly 30 people in there right now.

You think one of them, at least one of them must be calling in sick. We're on to you. And you could do that too. Details at linuxunplugged.com. That's probably our best housekeeping ever. By far. I think so. I think so. I think, you know, that was good. You know, we connected with people there. Well, and you decided to just double down and instead of cleaning up the beer you spilled, you just spilled beer on the places that it hadn't previously reached. So it's consistent. Yeah, I like the way it smells. So don't host your own email server.

But you know what? I don't really feel like going into a lecture of reasons why you shouldn't. I feel like we've covered it, but super briefly, it's hard to get it right. And it's something that people really rely on, and it's really easy to get sucked into a black hole where people start rejecting your emails. And not to mention there's a lot of security considerations. Yeah, and for some reason, email is like internet infrastructure, you know, businesses, banks, whatever. They expect that like emails go through and you probably expect that you receive

your email. So if you are going to play with this, at least keep that in mind and maybe do it on a domain that you can experiment around with and not your main one. Well, you got to figure people in our audience, you know, they maybe have a project they're working on or a small team of people. You have like our Lug, right? Wouldn't it be kind of cool if our Lug could have an email address and members of our Lug could have their own email? And we were thinking for 400, that'd be kind of a fun thing to do for our Lug is create a Lug email address. And so that's kind of what we were thinking is that's a great example where you got a small group of people. You don't necessarily want to be paying five plus dollars a month per

mailbox to host it somewhere. And you just want to whatever, like you're just going to, you just want to pay whatever the cost is to run the server and create as many accounts as you want, three to a hundred or whatever. And that's the goal we're kind of shooting for is creating a humble mail server with, that's my word of the day today is everything shooting for, is creating a humble mail server with, that's my word of the day today, is everything's just, I got a lot of humility after getting my butt kicked by a robe. Mine's not going to be humble.

Mine's going to be bold, brazen on the internet. Good for you. Everything is humble pie with me. Because once you've gotten your butt kicked by a robe, you don't walk around with your head hanging high. Let me just put it that way. But something simple, not too complicated. And there are actually other reasons, I think, to host your own server.

Privacy and control being big ones. You have full control over both the server and your email itself. You can view the logs of incoming and outgoing messages directly. You can pick the applications you want so you can mix and match the server software and the client software. And you also are implementing your own data retention policies. Perhaps you want to keep things forever, or maybe you don't want to keep things past 30 days. You get to choose. I mean, it's also probably worth saying that email is, for all of the difficulties we were just mentioning, it is still one of those things that like it just interoperates, right? You don't have to go

officially like register somewhere. There's not one party controlling it all. If you want to host it, if you get it right anyway, you can. And that's cool and worth celebrating. Yeah, and really, I think there is the cost aspect too. Right. This is something everybody has to find a balance on, but you're never really going to pay more than just the cost of running the server in your time,

where with a lot of hosted solutions, you're going to pay per person. Yeah, or I mean, it can kind of vary, right? Like they get to choose how they want to monetize off that service. And so it might align with your use case. I've been using Zoho for a while for some of my, you know, non-Gmail or other hosted domains. And it works pretty fine. I get like aliases. But if I ever did want to add more users or say I started a small business with some friends and I wanted to add them on there or for an open source project or something for the show, it's not very flexible, at least without paying more money.

And that has had me thinking about self-hosting for a while. Yeah, and there's just sort of like, can I still do this? You know, like, can I still make this work? Because I used to do this, but I haven't done it in a long time. And in fact, one of my very first jobs when I got into IT, like my second or third job, I can't remember, but it was early into it. They just let me go on managing their mail server. And it was a deep dive.

And I really kind of became a, I became a pro. I really got into it for a while. And I'm wondering, like, does that skill set still exist? Will those, will that come back as we do this? Is hosting your own email server just like riding a bike? You know, speaking of that, I'm not sure I've seen you ride a bike either. Oh, well, you know, if we could come up with a show reason, you'll see me do it. That's all it takes.

Every mail server out there can really kind of be broken down into three separate software components. You have the mail transfer agent that relays mail between your server and the wider internet. You have the mail delivery agent that takes mail from that MTA and then sends it to the individual mailboxes. And you have the interface that clients will connect to, which is often like IMAP or even used to be POP3, which lets users check their email from their preferred client. And then there's like additional stuff you have to kind of consider, right? Because that's the basics of a mail server, and you have to pick those software components

that you think do the best job. But there's also security components to setting up a mail server today. And of course, there's DNS. Yes, our old friend, the domain name system. You're going to need to set something up so people know how to reach your email server. You'll need at least MX record, SPF, and pointer. Starting with MX there, that's going to tell people how to send you email, right? Like, I've got an email, Wes, at jupiterbroadcasting.com, and when an outgoing server looks that up and says, how do I send email,

they go find jupiterbroadcasting.com, look for the MX record that tells them the server to point to, and then they can actually connect and start sending email. And the important thing to consider here, too, is there's security things around DNS as well, like SPF records, which help establish the legitimacy of your mail server and reduce the chances that somebody's spoofing as you. The SPF record for your domain tells other receiving mail servers which outgoing servers are legit to kind of receive from. So it's a lookup for them to do to see if the sender is a

legitimate sender or not. And don't forget to set up reverse DNS. Yeah, I know. Sometimes we just get lazy. We just do the one direction. But the reverse DNS for your mail server really should match the host name of your server. So people go to query, hey, what's this IP? Who's sending me this mail?

That needs to match. Otherwise, you might get some ugly rejections, things saying, hey, the reverse DNS does not match your SMTP banner. And you're not going to have fun with that. Oh, man, this is really bringing me back. There is a lot for you and I to consider. And, you know, I mean, I would imagine we'd never probably have more than 50 email accounts

on this box. But even this isn't getting into like the antivirus and spam prevention stuff, because one of the things I did back in the day was I'd have to route mail through Clam AV and scan it on the mail server so that way hopefully I wasn't delivering viruses to Windows users. Right, and I mean these days I think with web mail being so common,

especially hosted by big parties like, you know, Googs or Microsoft, they do all of that for you. You probably expect that. And if you suddenly have an unfiltered inbox, especially if you've pasted that email around the internet, that could be a nasty surprise. And I'm glad you mentioned webmail too, because that's something else I think you and I should consider for this project. Because if this is kind of like a vanity account, it's kind of like an extra account for people who participate in the community and they want it or whatever. Like, I don't know, I haven't really

thought through this completely other than I kind of picture the use case being LUP Lug members who want to reach out to guests or organize events for the Lug and want to come from an email account that kind of seems sort of semi-official and connected, like jupitercolony.org or something like that. But I don't really envision that being more than a handful of people and then whatever you and I choose to use it for. But even there, it's like, well, okay, so it's not going to be their primary email account. So how can we make it accessible to them? And webmail is probably one of the ways to do that, but there's a lot of webmails to choose from. Yeah, I'd like one that doesn't suck, please. I mean, at least if I'm going to use it and not just have to set up a client because it's so bad. Yeah, this is definitely one I could use some advice from, from anybody in the mumble room or people that are listening that want to go to linuxunplugged.com slash contact. Assuming we set up kind of a standard mail server setup with IMAP

on the back end, what would you recommend we use for the webmail front end? Because our goal here is to kind of get everybody's feedback from this episode and go build something by the next episode. It's a little ambitious, but that's kind of what we hope to do with episode 402 is say, okay, the email server's up and running. Here's how we built it. Here's kind of what we considered. Here's the webmail clients we're going with. Here's the desktop clients we're going with.

And then give away some email addresses on it and stuff like that is kind of what our intention is for 402. But I think the number one thing I'm hung up on right now of what our intention is for 402. But I think the number one thing I'm hung up on right now is what webmail client to go with, because part of me is sort of leaning towards simple, like squirrel mail. Classic. Right. And you know, it works, right. And then part of me is kind of leaning towards something like round cube or something that's maybe a little more modern, but I don't even know what that would be anymore. Yeah, that's just it. It's like, it's been so long since I've really played in this space, but you know, our talented audience out there, there's, there's

email self self-hosters among them and they've surely got some tips. I hope so. I hope so. And then I'm going to take all of this experience and, and, um, see if my current bias that I still think people probably on the whole shouldn't self-host email. I, email. That's how I feel today. We'll see if that's how I feel after we do this experiment, because I felt that way about Arch, too. Exactly. Let us be your lazy admins. We'll try it out. See if it works. See how big of a mess we make. And then you can judge, is it worth the time? Yeah. And you know what? Honestly, I thought NextCloud would blow up on our face too, and here we are like almost two years into using it now, and we've had to make modifications to

how we use it, and it hasn't been perfect in its syncing, but once we learned kind of how to use it, it's been fine. Yeah, yeah. I mean, we definitely like to complain about it sometimes, and there are improvements it could have, but we haven't kicked it out of our infrastructure. And more often than not, it's a dependable part of it. Yeah, and then we can update you on how it's going to and stuff like that. And I'm funny, I kind of enjoy the idea of my prejudice being flipped. I like the idea that I'm coming into something thinking we shouldn't be doing this, it's not going to work, and then have it be successful and have to kind of go, oh, okay, it was worth doing. And then kind of try to come to terms with, well, why was it worth doing? What did I get wrong? What was I missing until I tried it?

And I think that's a good experience to share with the audience too. So that's what I'm hoping we get out of this. Yeah, this is definitely one of those areas where you see a lot of different mixed advice from folks. Some people have no problems doing it. And then you have a lot of, for good reason, people chiming in saying like, look, this is just hard. You might have difficulties. And if you don't want to spend hours on your Saturday troubleshooting, don't try, but you can't know until you actually try it for yourself. I'm coming into the back end of this conversation. So I'm not going to ask why you want to do this. Hello, Wimpy. Hello. But you could potentially save yourself some heartache by capitalizing on things like mail-in-a-box, where there is a community maintaining a deployable, full-stack mail environment, which pulls on multiple people with experience in that space.

Wes, we've gotten a lot of recommendations for Mail in a Box over the years. I think, you know, probably at least a dozen. I know there's some competitors, but that's the one that sticks out in my mind, even though I've actually never tried it. Yeah, I haven't tried it either. And neither have I. And I only cite that one

because it's the name that came to the front of my mind, but there must be other things like that. It was interesting, both of you saying, it's been ages since you've done something like this. I can't, well, in fact, I can remember the last time I did anything to do with mail servers was in 2016, when my last job was to maintain a fleet of massive mail servers for a CV application

that was sending out like 27 million emails a minute, like 24-7 or something. I love those kinds of memories. You know, looking back at that, thinking that is crazy today, like something you would never catch yourself doing now, right? I assume you just would never build that again. Well, I mean, I didn't build it. I inherited this sort of, you know, I inherited this sort of, you know, of course. Oh, those are even better. And, you know, things have moved on in the intervening six years that would mean that you just wouldn't do it quite like that ever again. But, yeah, I mean, interesting experiment.

It will be. We shall see. And I guess we know who to ask for help. Yeah, you're going to be our go-to mail admin. I am tempted to try Mail-in-a-Box. I'm very tempted to try that because we have gotten it recommended so many times. You recommend it yourself, Wimpy. It might be the way we go. Like I say, I've never used it.

It's just the name I know. The orange one, you have a suggestion for us as well. Yeah, years and years ago when I made the terrible decision of hosting my own email, I used an alternative to Mail-in-a-Box called MailCow, which is sort of this dockerized setup so you just run a command it runs everything for you and that i found to be really nice it uses a alternative to round cube called uh sogo which is sort of materially fancy looks a bit pretty but basically is webmail and calendar and all of that stuff. And from what I

used of it, it was really quite nice and just sort of got out of your way and just worked. How did you spell that? Sogo, S-O-G-O. Oh, okay. That's easy. All right. I will definitely look at that. Yeah. The webmail thing is like, I'm trying to chew on that one right now. I'm not sure if mail in a box provides a webmail client. I might be worth looking into. Maybe they've already picked something there. The other option is that you could go super simple and just go full on like Zimbra. I have run Zimbra for a couple of clients.

I use Zimbra also as a mechanism to back up exchange servers. And I recall Zimbra had a fantastic search. So if you're somebody out there looking for kind of a suite of tools with really good search and like a SQL backed email database, Zimbra is really good. I don't know if it's still the case, but certainly back in the day when I was deploying Zimbra, my citation for successful deployments was Red Hat. I know they were using Zimbra internally. Yeah, I've come across it a couple of times. I'm surprised it didn't take off. I feel like it was a better exchange. Yeah, it was designed to be exchange compatible, stroke compliant. And it was one of these weird things. It originally grew up out of, you know, Yahoo acquired it. It

got kicked around between, you know, different organizations over the years, but it was always very solid. Yeah. I had some, I had some heavy hitting clients that really used email and, and we ended up switching to Zimbra for their archival and search solution because they just simply couldn't find it in Outlook when they were looking for stuff fast enough. And Zimbra was just faster. So that is a really good one, but I think that's probably too much because that's like the kitchen sink approach to email. It's everything. Yeah, it's all the things, without a doubt. That'd be a good, like, if you were looking to replace Google Apps, you didn't just need core email, but you needed like a whole suite of functionality,

including calendar and collaboration space. Indeed, yeah. You combine that with something like Mattermost, and you'd have a really good self-hostable internal collaboration suite. Those two things together would be pretty good. So I guess we'll wait for the emails to come in. We'll review our options. And then next episode, we'll announce our decision on which way we went

and if it's up and running or not. We'll see if we get it going. I hope I haven't overcommitted us, Wes. I may have oversold, but we're going to get done. Yeah, let's find out. Hopefully, at least, we'll have some email. We can get some tests

and see how many, you know, audience mails actually make it into the show. Yeah. All right. Well, before we go, I really got my arse kicked by a robe recently.

I thought I'd share that tale with you because we're just about a year into things getting serious. We're not at the year mark of being independent, but we're a year mark into the process beginning. And, you know, I thought when I was envisioning one day back this, you know, at this stage, I was just fantasizing, thinking to myself, you know, if I were to go independent again and try to survive, especially during a pandemic, I know that I'd have to do more than just sponsors. And the membership program was part of that. And so thank you, everybody who has become a member, because that was a critical part in making it to as far as we have at this stage. And I figured advertising would be a critical part of that, and it has. And that's

why, you know, companies like Linode coming on board, literally, I'm not exaggerating at all when I say made it possible for Jupyter Broadcasting to exist. But I also knew that I couldn't just rely on those two things. I've been doing this for 14 years, and that's like 50 years, at least maybe 100 years in internet time. And one of the things I know that's absolutely true is if you have one or two income sources, you're going to die. on advertising the first time around. I've been around when Amazon, which represented like 20% of our revenue at the time or 30% of our revenue at the time. I was there when Amazon just one day pulled the plug on us for doing stuff that still other networks

and podcasts and YouTubers are still doing today. It's really, you know, but you gotta diversify there. So I knew when I was thinking about, boy, do I wanna make this transition? I knew I'd have to figure out my merchandise and swag game one day. And it's been just embarrassingly too long since we've had anything new. It's been over two years now. And it's something that I enjoy. And it's something, it's another one of those physical manifestations of what we make becomes a physical product. And we recently went on kind of like a archaeological dig here in the studio, and we

found a bunch of old swag. And I had an emotional reaction to finding this stuff. And I don't normally have an emotional reaction to things, except for Star Trek ships, that I, those I like. But mostly, like, you know, I, one of the reasons I moved into a class ARV is I wanted to get rid of everything. I don't really want to have attachments. I'm not big on that. But when I saw this swag, I felt something. I don't know what you humans call it,

but I felt something. And it was a good something. And I realized this stuff is special. We have some really special items. And I want to do more of that because it's like, I can look back and I remember I see an item and I think of an event or I think of a show milestone or I think of the way that the audience responded to an item that we did. And it makes like the past real in a way that I've kind of lost touch with over the last couple of years, especially the last year.

And also, getting this figured out is just a good business strategy, really. I mean, that's just obvious on its face. It's critical to making a long-term business successful. So all of this, knowing all of this, it created this self-inflicted pressure. Like I've been stressing out really bad to get swag right. And I want to do things that are unique and fun that aren't just like the stuff you find on every fulfillment website out there. And so I thought, let's shoot for the moon

and let's do a fabulous embroidered robe called the Coder. And it'll be a limited item. It'll be a robe that we can sell on the Coder Radio program to celebrate 400 episodes, because Coder Radio just hit that milestone as well. What a glorious dream.

I mean, you know, a custom swag worthy of the epic history of that show. Yep. Like Tyler's saying right now in the chat room, he says, Chris, last, I still have my challenge coins. Right? Right? I mean, Tyler, I love the challenge coin. It was the best.

I was just about to say, I've been sorting out my office and listen to this. Challenge coin? Two challenge coins being struck together, which I found from 2017 recently. I just found a cache of them as well, about 20 of them. And I love them.

They're very special. And, you know, they're like limited in a way. Anyways, it just gave me a whole new appreciation for swag. So I wanted to do this robe, and the robe was the dumbest possible idea I could have had in like a year, really. The last year is probably my worst idea in the last three years, maybe four years. Now that I think about it, it might be my worst idea in the last five years. What I learned was that shipping a custom product domestically is tricky. To ship something that's produced, custom, hand produced domestically, to ship that internationally, holy crap. I had

like quotes of like $185 just to ship a $65 robe to Denmark. It's just basically not possible at a reasonable cost. Because when you work with these one-off shops, their business isn't procurement and shipping around the world. Their business is taking these items and doing embroidery, right? And so they don't have like these depots around the world to make shipping more reasonable at international rates.

And boy, do I have a whole new appreciation for Amazon's cheap shipping. What a massive market advantage that is for them, Wes. They can ship stuff at prices I could never dream of. No, I mean, we don't have a bulk discount. We don't have established supply lines. This is all new. We got to get that drone program up and going, obviously. That I think we have the supplies for.

They have shipping prices that are just not even achievable by small businesses like a shop like mine. But for the community and for the business, I figured I have to crack this nut. And I figured the Robe would be a learning experience. I'd jump into the deep end and I'd learn a lot from it. And I did learn a lot. But one of the core lessons that I took away is that theme of the day is be humble. I have to start humble with something like this.

I went into the deep end with a robe and that was such a bad idea. If it's going to be something that becomes part of the business long-term, I need to start small and then I have to grow it and I have to learn as it grows. I have to build on small success after small success. So I've been thinking a lot about how to get this right because I think swag and merchandise

is going to be a fun part of the business. It's going to be something that we can, we're going to, I'm going to, I want to do limited stuff like name brand stuff and, and items that you'd actually want to buy and just wear in general or things and gadgets that you'd want to pick up or used items here at the studio that we're done with that are still great, still fully functional, but we just need something maybe that has more ports or does more channels, that kind of stuff. I want to have a business that can accommodate all of that. So I'm a little nervous to do this because I think this could be, this could be bad or this could be great, but I have an announcement to make today.

I am launching the Jupiter Broadcasting Garage Sale at jupitergarage.com. I am launching jupitergarage.com today at jupitergarage.com. Go check it out. Let the fireworks go. Go ahead. I know. I know. It's exciting. I know. Yeah.

I don't think you should be selling those to kids. Yeah. No, I know. But they do so good with the fire. You know, they're responsible. Oh, whoa. Whoa. Whoa.

That got close. So here's what I'm doing is I have a selection right now of mostly used hardware items like a FreeNAS Mini, a mixer, a compressor, you know, gear that I think is really high-end and good but we just don't simply use anymore. And instead of throwing it out, I'm going to sell it for really crazy, stupid, ridiculous prices. Really stupid, crazy low prices just to clear out the space because I want to make room for the new, for the new stuff that I want to do, the new swag items, the new merchandise,

the new stickers, getting ready for events to come back, and I want to stock swag for booths and that kind of stuff. And the way to do that is to get rid of the stuff that I've been stockpiling forever. And there's things in there that still work totally good, that are great items, that I've always kept just in case, but I really shouldn't be. I'm just hoarding. And I'm just kind of creating the garage experience, the garage sale experience, but online at jupitergarage.com with the goal to make room for the new.

And honestly, quite frankly with you guys, total transparency here. I just need practice boxing, shipping, the whole thing. Like I got to learn all of this. And I need space. I transparency here. I just need practice boxing, shipping, the whole thing. Like I got to learn all of this and I need space. I need space. But here's what I'm going to do. So go to jupitergarage.com. Look at some of the items I have there. If you're buying one of the used items as space and weight allows, I'm going to stash random retro swag items in those orders. Like just as I can, I'm going to just throw some stickers in there or something like that. Or I might wrap a device in a shirt or something like that. Just sort of close your eyes

and grab something at the studio, put it in the box. Throw it in the box. As long as it's not Levi. Right, no. But, you know what I figured too? Because I found a couple of caches of retro swag and I'm not making any more of this stuff. If you just want a random bag of retro swag, I'm going to make that available too.

They're going on sale right now. They're going to normally be $25 for a bag of swag. But right now, to kick off the new thing, they're going to be $15. And I only have a small handful available. I'm doing limited quantities so that way I don't get inundated with orders. And I make sure we get them out the door and that we're getting the shipping estimates right and that if somebody orders international, we can fulfill it. And I'm literally learning with each one of these orders. So not only are you going to make a very small financial contribution to the network and get yourself some gear or some swag, but you are

basically sending me through bootcamp to learn how to do this. And we're kind of accomplishing something together. I'm clearing out some stuff here, this gear that's still perfectly usable. I'm powering it on to make sure it works and stuff. It's getting a new home and I'm learning how to basically grow a new part of my business that will be independent of sponsorships or memberships or anything like that. Although that said with the memberships going forward, right now, these prices are stupid. Like you wouldn't believe what I'm charging for a free NAS Mini. There's no discounts because the prices, they are ridiculous. But going forward, in the future, I will make a standard discount available for all of our members.

So if you're an Unplugged Core contributor, you're going to get 15% off every order. And you'll have more information about that soon. That's true for all of our patrons and for members of any show of ours. You're going to get 15% off any of your orders from the Jupiter Garage in the future. Right now, as I'm learning how to do this, and I'm probably going to be on the other end of profit, I'm going to be paying out of pocket for some of this stuff as I learn it. I just can't do a discount because there's no discount to give. But there will be one in the future for all of our members because my intention long-term will be to clear out some of this limited-time swag

that we have to get some hardware out there. And there's a lot more hardware than what's on there right now, so there's going to be more stuff landing. And then kind of transition to creating inventory space for new, original, high-quality merchandise, stuff that isn't just the standard fulfillment t-shirt or something like that. I mean, I will still do those from time to time because who doesn't like a t-shirt? And there is a t-shirt with the bearded tux holding the Jupiter rocket in a nice kind of low-key

gray and white on black. That's available in the garage sale right now, as well as a hat that has the same bearded tux logo on it because that's my favorite tux. So there are a couple of those items that I'll do from time to time. But my goal is, I'm eventually going to have an entire wardrobe of this stuff, because I want to get stuff that I want to wear. And when I start rolling that stuff out, we'll start doing discounts for members more on the regular. And that's going to be more as my confidence grows, as I know what I'm doing. And I know I can tell somebody that you're going to pay this for international shipping or you're going to pay this if you live in California, you're going to pay this if you live in Canada. When I get to that point and we're actually getting there, then you'll start seeing more stuff roll out. That's just sort of as I learn more.

I'm going to take it one small part at a time. But there's always going to be that core part of the Jupiter Garage that will be items from the studio that we're done with. Stuff that is still functional, and we don't buy crap. So stuff that's good and functional, but we're just done with it. We've outgrown it. That's going to be always a part of the Jupiter Garage. And my intention there is just sell it, not really at much of a profit, maybe just make 20 bucks or something off of it really. And let you have something that we're done with instead of letting it sit and rot like it does now in a closet or in the garage. Plus we got to make room for beer. Yeah, Wes, that is actually part of this is when we decided, well, we're going to have a beer keg

and we're going to have a beer keg fridge and we're going to have spot to brew beer. We got to have room. And we retire stuff from time to time. It's still perfectly good. It doesn't make any sense that it just sits around and wastes. It could go out to an audience member. They could help support the network a little bit. Who wouldn't want to connect to the mumble room with a previously used Jupyter Broadcasting audio gear, right?

I mean, come on. That's pretty much it. So there's a selection of stuff there. There will be more from time to time. Go to JupyterGarage.com to go see what's there now. And we have gone through our closets. Angela just found a huge stash of stickers that will be slipping into rando swag bags and orders that people place. And I found a stash of last challenge coins. And then independently, there's some really nice retro last swag, like the zip up hoodies

with the classic bearded logo and stuff like that, that will be going up for sale soon too. As we just kind of go through and make sure everything that's been in storage is still in good shape. And then we're going to box it up and get shipping prices that make it available at the garage sale. That's the plan. And then over time, I hope to grow that into kind of an arm of the business that will be part of sponsorships, memberships, and swag that is swag like nobody else is doing.

It's not just boring stuff, but it's stuff that you can, you know, you could eventually end up buying. One of the things that's on there right now is the mixer I used for years of remote productions. And it's perfectly good. And it's got, I tell you what I love about it too. Like it's got, it's got all, you can run it all off nine volt batteries. So if you have noisy audio, if you have line noise on your electrics and you want a mixer that's battery powered, this is like the total ultimate hack to avoiding line noise is run a battery powered mixer. And I used it for years hack to avoiding line noise is run a battery-powered mixer, and I used it for years on the road. It's still

perfectly functional. I just now do all that stuff in Jack and don't need a mixer anymore, but it still works, and it's up on the garage sale. I'm very excited about it, as you can probably tell. And I've been going on and on about it for weeks to Wes. Just a bit.

So thanks for putting up with it, Wes. I appreciate it. There really is a lot of good gear that's been in there. And especially as the setup all pivoted to like Linux focused and more professional stuff or just, you know, more channels we needed to set up the studio right to have guests and get all the crazy jack stuff going on. There's just a lot of good gear and other awesome items from history that, as you said, it really would be a shame to just see them rot away on the shelves when we're not using them. Yeah, grab a bit of JB history and stuff that you could use to make content or run right now. And the chat room's right.

I need to update the website with links and stuff. I've just been getting it all going today and wanted to have it ready for the show. But I am often just the worst at that. and wanted to have it ready for the show, but I am often just the worst at that. I really need, the whole JB website needs a refresh because my intention was to have it replaced by now, but that never came around. And now it needs a refresh.

And links to stuff like this need to happen. So hopefully that'll get done soon. But in the meantime, just go to jupitergarage.com for that. Well, Wes, that is the no-show. That's the unplanned show that we had today. It's almost a show. It was kind of a show, I think, actually. I think we just accidentally had a show.

Whoopsie. So, I know. If anything we covered today interests you, then go check us out at linuxunplugged.com and get the links at linuxunplugged.com slash 401. You heard really briefly there from the orange one. He was just recently featured in the self-hosted podcast with his setup. You can

find that at selfhosted.show. Hosting things like your own email server is the core of what we talk about there. So if you like that, go check that out. If you do the Twitter thing, you can follow this here podcast at linuxunplugged. The network is at Jupyter Signal. And the whole gosh darn podcasting network is over there at JupyterBroadcasting.com.

And last but not least, may I just say, if I don't say so myself, picture me like a chicken who's like, you know, fluffing up his feathers right now and like getting braggadocious. I'm also picturing high-waisted pants for some reason. And maybe some glasses. But gosh darn it, if I'm not just getting more and more proud of that Linux Action News over there, if you work in the industry, don't miss Linux Action News. Every Monday morning, Wes and I break down everything that matters in the world of Linux. And something you don't know by listening to the show, but something I want to tell you is Wes and I put a lot of effort and thought into

how can we say something concisely, efficiently, and so that it's easy to understand. And we try to get you all of the news you need to know in less than 30 minutes. And we pride ourselves on delivery in a way that is understandable, processable, with the context that you need, that doesn't drag on.

And so there's a lot of stuff that doesn't make it into this show. Otherwise, this would be like a two-hour show. And those stories and that information, like the recent ports to the M1 platform, we went and got quotes from Hector Martin. We reached out to Hector Martin for contact and we bring that into the show.

And that's another thing we do with Linux Action News is the stories are sourced. We try to contact the people behind the stories, the people writing the stories. We try to get their takes and that information. We incorporate that into the show either by what we say or sometimes we'll play the audio directly from them. And that's a special kind of effort that we spend the last half of our week doing to make Linux Action News special and to have information that perhaps you're not hearing or reading in other places. Yeah, save yourself the time. Stop clicking through the clickbait in our Linux.

Just head over to linuxactionnews.com. That's good, Wes. That is a, that's good. Well, this here podcast, it's going to be live next week. Even when we say it's not going to be live, sometimes it's live. See you next week. Same bad time, same bad station. So keep the Linux rolling and make it a Linux Tuesday.

Join us live, 12 p.m. Pacific, 3 p.m. Eastern. Links to everything we talked about today, how to contact us, our Mumble server, our matrix info, that's all over there, including subscribe links for RSS feeds, etc. at linuxunplugged.com. You can get all of it over there. Go get your links, go get your show notes, and then go check out the garage sale at jupitergarage.com. It's brand new, and I'm going to school,

and you can help teach me a lesson or two. Thanks so much for tuning in to this week's episode of The Unplugged Program, and we'll see you right back here next Tuesday! so so so so so so so

so so so so so so so so

so so so so so so so so so so so jbtitles.com let's go name this thing did you guys see that the gnome project is considering moving the top bar to the bottom and creating kind of a layout similar to chrome os yeah

what i really hope not really? It reminds me too much of Windows and then also obviously Chrome. And I just like that Noma Lee sort of looks unique right now. I think that's precisely why they're considering doing it because it looks like Windows. Back end of last year, I think

it was probably around October, November time, the GNOME Foundation had like a show and tell based on some user accessibility testing that they'd commissioned. So this was putting GNOME and other operating systems in front of a random selection of people or a selection of people with different backgrounds and different computing backgrounds and what have you. and different computing backgrounds and what have you. And it was GNOME as it was.

What was slated to be at the time GNOME 40 and also including stuff like the endless OS spin on how the GNOME desktop is presented. And long story short, those people that are not already familiar with GNOME desktop is presented. And long story short, those people that are not already familiar with GNOME gravitated towards the intuitive nature of using endless OS because it had a lot of similarity with Windows, which is what they knew. This is a point that I had a conversation with Neil McGovern, the executive director. He said, yeah, that when people use more traditional UI paradigms, they got to work faster, basically.

Yeah, and also, those traditional patterns didn't come out of nowhere. They're through long-established design research and human interaction research. So it was no mistake that GNOME looked a lot like Windows 95 and Windows 95 shared some familiar patterns with macOS. All of those were influenced by, you know, the same design research and design thinking. Which shared some design patterns, honestly, with Xerox. I mean, really. Indeed. Indeed. You know, this all goes back to, I mean, it does go back a long way. It goes back to the 1960s, but this is, you know, well-researched, well-funded

research that goes back a long way. So I am not surprised to learn that GNOME are taking that feedback on board and looking to, I'm not going to say roll back necessarily, but there is some sort of stepping away from the path they've been on to responding to user feedback. And user feedback is what people are familiar with, is comfortable and intuitive. Yeah, I wonder how, I mean, I'm pretty flexible. I could probably roll with whatever,

but it would be a change. You know, there's forever. I mean, with GNOME, I've always had a lot of my items up in a top bar in one way or another. It'd be a pretty significant shift for them. Yeah, but what about that mouse travel, though? That's the part I've seen people kind of talking about the most, I think. Yeah, yeah, I just use the keyboard for everything. Yeah, exactly.

Yeah, right. Everybody knows which buttons to push now. Yeah, exactly. So Wimpy, the Ubuntu podcast is back. It's active. What else is going on? How are you? I'm very well. I hear that everyone else is very well too. Yeah. Yeah. You know, he didn't stop by the show today, but a friend of the show, Popey has recently left Canonical as well. So both of you now have left the Canonical mothership, but I assume for greener pastures. Yeah, obviously I've been at Slim now for like

five weeks, I think I've been working there. So I'm doing my thing. Don't think Popey's left Canonical yet. He's announced that he will be leaving. A bit like I did, I announced my plans to leave ahead of time to give people in the community who may have wanted some final stuff from me time to contact me and round that out, and Popey's done the same. So yeah, Popey's off to something new. I'm not going to disclose what that is. He'll do that when the time is right for him, but well poppy if you're out there listening you are welcome to join us when when you're uh when your last day arrives and you want to come and hang out with your

buddies what i will say is this some people have reacted to you know there are some people that maybe only know about alan and i and our association with ubuntu and canonical and the fact that we've both left recently they see as as the sky is falling. That's not the case. And perhaps, you know, the silver lining to this is maybe the two of us, you know, old and grizzled, you know, people from, you know, Ubuntu leaving is a good thing because there is, you know, vibrant new new life at canonical who are spearheading community engagement now with you know monica and reese and this is their opportunity to take things in new and interesting directions so i don't think people should be interpreting this as you know a negative thing but the opportunity for a new era of,

you know, Ubuntu community to shine and blossom. I like it. I hope so too.