Feature: 2023 Self-Host User Survey Results

Hi, and welcome to the Self-Host Cast, a podcast dedicated to all things self-hosted. Today, we're discussing the results of our recent user survey with YouTube content creator David from dbTek. Thanks for listening. Hey everyone. My name is Ethan and I'll be your host today. For those who don't know me, I run an independent publication newsletter about self-hosting, links of which can be found in the show notes.

I am super excited to be kicking off the opening episode of the Self-Host Cast with David from dbTek, who's agreed to sit down and talk through the results of our annual self-host user survey. David, thanks so much for joining me. Thank you for having me. I'm actually pretty honored to be your first guest on your first podcast. I've never been given this opportunity before. I think we're going to have a good time. Just a quick little bit about me.

Like you mentioned, I run a YouTube channel where I talk about self-hosting primarily in Docker, but we cover a few other self-hosted solutions. And I've always got something new coming out. So if you guys want to check that out, you can look me up as dbTek over on YouTube. Awesome. And for anyone interested, I will also drop a link to David's YouTube channel in the show notes and would encourage everyone to check it out. With that being said, we do have a lot to discuss today.

And so I'd like to take just a little time to provide a little context so we can jump right in. About a month or two ago, I hosted my publication's first ever self-host user survey, which was comprised of 34 questions related to self-hosting in some capacity. I received about 1900 total responses to the survey, which I published via charts and other visualizations in a blog post just a few weeks ago, actually, which I'll also link to in the show notes.

I've been sitting on the idea for a podcast for some time and thought maybe a review of the results would be a fun way to launch it. Since launching my publication earlier this year, David and I have had a few occasional interactions and so I asked him to join me in the discussion given his technical background and my lack of technical background. The survey itself is divided into five categories with the first being environment. So all questions relating to self-hosting environments.

And this leads us to our first question, do you self-host for personal or enterprise purposes? Right. And it looks like looking at the data here, you know, like 80, probably 80% of people do it for personal reasons. And honestly, I'm not super surprised by that. I am surprised that there was, maybe I wasn't surprised that there was also quite a few people that also do for enterprise purposes. You know, their job is probably a sysadmin at some tech company.

And then hopefully because they love it, they bring their work home with them and they translate that into their own little mini home lab or data center, whatever the case is. But it's interesting to see how many people who aren't necessarily doing this for a tech job or doing this for their own personal reasons. I absolutely love to see that. Absolutely. I feel the same way.

It's interesting seeing how large that slice of both is, I know enterprise by itself is small, but there is a significant slice, as you mentioned. It's interesting as I browse the self-hosted communities I'm a part of, I'll frequently see people endorse software that is enterprise focused. And I'm always asking myself who is on Reddit self-hosted subreddit looking for enterprise software. And this really points to there's probably a larger audience for that than I'm aware of.

And my non-technical background that probably speaks to that a bit and why I might not have been as aware of it before the survey. There's something that we'll get into later, but there's also, I've noticed a little bit of gatekeeping when enterprise software gets introduced into a home lab as to whether or not you're actually truly a self-hoster. We'll get into that a little bit later, but I've noticed that that seems to be a little bit of a trend there on Reddit as well.

Absolutely. So I'm going to jump to the next question here. With self-hosting for personal use, approximately how many unique users are registered across your applications? We had a lot of really interesting responses to this. David and I talked a little bit about this beforehand. We were joking about some of the responses we got in this particular question because we had, I don't know, somewhere between 10 and 15 responses of zero users.

And we weren't sure if people were deploying self-hosted software that wasn't being used by anyone or if those were one user instances and people were answering as if they were being asked about how many users aside from themselves. I've had some time to think about this since our last discussion. And I think it may be one of those things where in some places you've got the ground floor and then the first floor and then the second floor.

Whereas here in America, we're on the first floor or the second floor or the third floor. And I feel like that it could be exactly that. It's like I am the ground floor and everybody else is a user beyond me. I feel like that's absolutely what's going on with those zeros is it's just the sysadmin, just the home labber that's talking about they're the only user, so they count that as zero. So that's the only logical conclusion I can come to from that. The rest of the data is interesting.

So two users seems to be the most popular response and then kind of whittles down from that. But interestingly enough, it spikes back up at 10 users and makes its way back down. I'm not sure if just how round of a number 10 was as people sat down into this survey rather than thinking through, oh, do I have 11 or 12? They just thought, hey, I'm just going to plug 10. It's a nice round number and that's what it is. Yeah. I think that's probably exactly what's going on there.

Yeah. So David, just an interesting question. How many users do you have using your self-hosted deployments? I think my default answer would be 10, but I think if I were to actually stop and think about it, probably seven or eight, I think is kind of somewhere in that neighborhood given password managers and things like that that people don't want to have to rely on, a third party application that might be more easily compromised because it's well known.

I think it's a bit of a loaded question actually, and it might not be entirely fair. When I think of my own self-hosted software environment, when I look at something like Plex or Jellyfin, I might have 12 or 13 users across friends and family, but anything else I host like password managers like you mentioned, the number of users is much, much smaller. Absolutely. Yeah. I would say that I've got the same thing.

Most of my users are sharing my Plex or whatever, but I do have other users that are still using some other stuff on my network. So if you were to take Plex or a media server out of the equation, it's probably three, but everybody likes to have a friend with a Plex server. Yes. So moving on, next question, why do you self-host?

This was a multiple choice question, so we gave a couple of different selections, and I will list them in the order of number of responses starting with the greatest number of responses. So we had hobby, education, privacy, flexibility, cost, business, and then other. I don't think any of these are super surprising, at least the order they came out in. David, does anything stick out to you when you look at those responses? No, no, not at all.

I'm actually really happy to see that education came in after hobby, because I love the idea of people learning for the sake of learning. And of course, that then translates into the hobby, and then you get the side effects of flexibility and cost. Cost has its own caveats for different reasons, but I love that education is so high up on that list just because I love to learn, I love to teach people. So seeing that so high up really makes me happy.

Absolutely. What I think hobby is encouraging for myself, as I mentioned, I don't work in a technical field, but given my self-host experience, I find myself leaning more and more towards a career in a technical field. And I'm always a little concerned of, hey, I really enjoy doing this stuff in my free time, but if I did it for a living, would I not enjoy doing it as much? And so we have a question later that speaks to how many of our responders in this survey work in a technical field.

So it's encouraging to see that all these people do work in technical fields, yet continue to embrace this as their hobby and continue it when they get home after their normal working hours. Yeah, absolutely. I 100% agree with that. Yep. And then privacy, that's not surprising that that's so high up. There are a couple of privacy related questions that we can talk in further detail when we reach those.

Sure. Flexibility makes sense, again, given the hosted paid offerings that are out there and how many great pieces of software have been developed for the self-hosted community that are completely free of charge, and then cost kind of lumps into that sentiment as well.

Like you mentioned, there isn't necessarily a cost to using a lot of the free services out there, but one of the things that we had talked about was if there's a project that you use a lot that you're passionate about that you want to see succeed, I think there is a cost associated with that if you want to donate to these developers, these organizations to help them keep building the applications you love.

And I know that that kind of gets into a weird area for some people because they want the free open source software, the FOSS. But there's got to be a cost. You have to support these people in order to make sure they can keep doing their jobs. And that's them in return supporting us. Absolutely.

Absolutely. Sometimes I think it's not obvious the amount of effort developers and other people put into the efforts behind developing and providing the resources that goes behind some of this self-hosted software.

Yeah. If I can kind of plug something I've got coming up real quick that lends itself to that, I have got a video coming up soon that I'm going to show how to do a security check on a Docker container so that it will actually go through and look for all of the vulnerabilities that are known for that container. And I did kind of an A-B test earlier for a kind of a one-man operation who's getting into developing Docker containers.

And I did the same test on a container from Linux servers developers. And the number of vulnerabilities comparing A to B is scary. And I want to say that the bad, I'm using air quotes there, application is a bad application. It's just there's more security thought when you've got a bigger team and you've got somebody who can specialize in patching those vulnerabilities. So I've got this video coming out. I just started kind of looking into it today. It's coming out soon.

And I think it'll kind of give a good indication of why it's important to have a team of people often working on a project and why it's important to support that team to make sure that your containers are as safe as possible. Absolutely. Absolutely. Looking forward to it. Moving on to the next question, we have, this was again a multiple choice question and this was more of a multi-select understanding that people might have multiple options.

It is select the hardware option or options that you use to deploy your self-hosted software. And again, so we have a number of responses in order of those responses that receive the highest amount of votes. We have low powered device, desktop PC, custom built enterprise, server cluster, laptop, mobile phone, and other. And so this one's interesting to me. This is one of the more interesting responses I think we received in the survey.

So when I think about laptop and mobile phone, those being at the bottom make complete sense to me. I know people occasionally will recommend laptops. You just plug them in constantly and just hope the battery doesn't light on fire or explode eventually. Right. And obviously mobile phones aren't necessarily ideal for self-hosting. I was very surprised by both low powered devices and desktop PCs being so high in number of responses.

I'm actually not terribly surprised by the low powered device being the top response actually. There are so many mini PCs that you can get out there or single board computers that have got a max TDP of like 15 Watts. And you can get them relatively inexpensive, a couple hundred bucks if you don't know where to shop anyway. And I think it's a great way for people to get in to self-hosting without messing up their power bill too much. My first server was, it actually was an old desktop PC.

It was an AMD FX8350. It had 16 gigs of RAM and it had an Nvidia GTX 770 in it. And that CPU was power hungry. I think it was a couple of hundred Watts for its max TDP. And don't quote me on that. I don't remember. It's been a long time. But I think these low powered devices, they're getting inexpensive. They're easy to get a hold of and they don't use much power. And I think it's great that people are going that route versus recycling an old power. Power hungry device.

Of course, then you've also got the issue with also not wanting to fill the landfill full of e-waste, right? So it's kind of walking that line of finding the right device for your setup. And I think with most people getting into self-hosting, they're doing password managers. They're doing low footprint as far as power is concerned. They do these low footprint, not power hungry applications that can get by on something like that. So I'm stoked actually to see the low powered stuff there.

The desktop PC stuff is again, recycling hardware that would have otherwise gone in the landfill, but also may have more power than one of those low powered devices. So you can probably push more intense applications through them. The one thing that I thought was interesting that I didn't see on here that may have fallen into other was NAS devices being a self-hosted device.

You know, I've actually, I've got a Synology back here that it was, that's where my primary Docker stuff resides for my in-home stuff. So I wonder if that's where some of that other in this category came from was NAS devices because it wasn't specifically listed in there. Yeah, absolutely. And I should probably add that as a note for next year to include that as an option because to your point that other, the number of other responses is actually fairly high.

And just jumping back to low powered devices, I think I might be just personally biased against them because I'm that person that probably throws way too much at the raspberry pies I have in my basement. And I blame them when things aren't going, working as expected or things slow down or all my memory is being consumed. So yeah, that's probably more me than it is a critique of anything else.

Sure. Well, and just one little last thing on that is that when the world shut down for a couple of years, raspberry pies were hard to get a hold of. And if you could get hold of one, they were five or six or seven times the price of what they should be. Whereas you could still get some of these low powered devices relatively inexpensively. And I've got half a dozen raspberry pies kind of strewn about the house doing different things. So I get it.

But I kind of think that's maybe where these low powered devices picked up traction was when the pies just weren't readily available. Absolutely. And to your NAS point then, moving on to the next question, do you deploy a network attack storage device, a NAS for storage in your self-hosted environment? But this was an interesting question, but I wasn't surprised by the responses only because, and I think we'll talk about this a little bit later, at least from a personal self-hosting perspective.

And again, I'm biased because I'm not as technical. So maybe the communities I spend my time in aren't either, but so many people get into self-hosting for media streaming. And what do you typically need for media streaming? You need storage and a lot of it. And so a NAS is normally how you achieve that. What's interesting is I've got a NAS that specifically, well, its primary focus is storage, whether it's for the containers that I'm running on that device.

But it also runs NFS shares for a Proxmox cluster that I've got. But I've actually got another, well, I've got a second NAS that's just backup of all of that. We're not going to count that. But I've got just a little inexpensive TerraMaster device. They're not sponsoring this, but that's what it is. And its whole purpose is to, it is my Plex server. It is my Plex device because it's got one of those Intel quick video sync processors in it.

But it actually pulls all of the media from the first NAS. So there's nothing actually stored on it other than the Plex application. So it's a total waste of so much. But it's what I had. I needed a Plex server like right now. So that's what I used. And I don't feel like trying to move it to a different device yet. That's okay. I don't think self-hosting is necessarily synonymous with efficiency. And I think engineered is often a word I would toss around when describing self-hosted environments.

So absolutely. Sometimes you got to do something janky to get something to work and then it works. And it's kind of that saying of it. If it works and it's stupid, it's not stupid. So I think it's kind of the same thing. I got this working. I'm not going to touch it again until it doesn't work. And then I'll fix my jank here. And I think that's kind of a pretty common thing in the self-hosting community.

I think we all have that one machine in our basement who has like 3,000 days of uptime because at one day a long time ago, we finally got something to work and we're too afraid to touch it. And so we just let it do its thing and just dread the day where one day it doesn't do that thing anymore. Yeah. So one of my NASs, I was disassembling it because I wanted to put some more RAM in it and I wasn't paying attention and I broke off the control board to do the power and the reset and all of that.

And then I realized that I couldn't turn it on and off with the button anymore. So if it powers off, I have to, I've left the whole shell off of it. I have to do a bios or a jumper reset on the board and then plug it back in so that it will power up properly. Yeah. So I was devastated when I broke that piece off and again, it works. I'm not going to touch it, but it's on a UPS now. So that's likely that it going down is very, very slim. And I couldn't be more thankful for that right now.

I love it. I have to say that's somewhat encouraging because I feel like online I read stories like that all the time. And to sit here and look at someone like you, who's a very successful content creator and I'm looking at your home lab sitting behind you right now in your setup and to know that you also have your own assorted machines that are just very jankily put together and maybe might not or might stop working at any point in time.

It gives me a little bit of hope that maybe I'm not alone when that stuff happens to me. No, absolutely. Like I had mentioned in our previous conversation leading up to this, the lighting I've got here makes everything look a lot better than it really is. So it's very dark in that corner and there's a reason for that. But no, I think everybody's got their jank.

It's just whether or not they're willing to admit it and even in some cases show it on the internet, whether it's Reddit or YouTube or wherever. So yeah. Right. That sounds good. I'm jotting down in my notes colored lighting for my equipment to make it look better and more impressive. Yep. And you just got to pick your favorite color and just run with it. Okay. One color, not RGB. Got it. Okay. So moving on to the next question. This is somewhat related.

I think we talked a little bit about this earlier. Do you typically avoid self-hosting sensitive or important data? And so there were about, I don't know, maybe 25% of the responses said yes. And most people said no. And this is an interesting question because keeping sensitive information close to us and out of the hands of others is actually a primary reason for why many people are self-hosting.

Absolutely. And I think, you know, it was my passwords was one of the, like after I did my obligatory Plex server to get me into the hobby, my first real big jump was self-hosting a password manager. I kept having issues with all of the commercial options that were out there. And then shortly after that, I started actually backing up my phones to my local machines. And I've gone through several different iterations of how I do that.

And like you had mentioned earlier, like you got to keep your spouse happy. You've got to keep the solutions simple so that they don't hate having to participate. And so like I've got a couple of different solutions for like backing up my phone and backing up my wife's phone and they just work. And I'm honestly like going back to the NASS question before this, I'm honestly surprised that more people aren't. I know it's almost 80% of people are. Oh wait, do you typically avoid?

Yes. Okay. So like 80% of people are backing or are storing passwords, are storing photos, that sort of thing. Yeah. I'm surprised it's not closer to 90% based again on the number of people that are using NASS and the privacy question you mentioned earlier. I'm surprised those numbers aren't higher on this particular question. I think you probably hit the nail on the head.

If I had to guess, I would think an adequate backup strategy is probably what scares the most people about hosting that information. If I'm the only one hosting my passwords and my, God forbid my house burns down, do I have access to all my passwords? I think all the tools out there, if you go to some of the common password manager, GitHub pages and all the documentation out there, there's not, or at least I don't have a concern that people are going, that my data is going to be compromised.

It's always, is it backed up? What happens if I lose, if my server goes down or something happens? And I would guess that's probably other people's concerns. Although if there are people newish to self-hosting and they're not quite comfortable opening ports for reverse proxies so they can publicly access this data, I can see that being a concern, but I would guess it's probably mostly lack of confidence in a decent backup strategy. Absolutely.

And that's actually one of the reasons that I switched from using, when I first started, I was using Open Media Vault as kind of my Docker host and my NASS solution in the whole bit. And then somebody in my Discord encouraged me to try out Proxmox and I didn't understand why until I deployed it. And then I realized, oh, I can also deploy a Proxmox backup server and I can make those two talk to each other with just a few clicks.

And then I can run daily backups where I can run whenever I want to run my backups of my Proxmox stuff over to that backup server. And if I've got a container or a service that goes down, again, with just a few clicks, I can bring that service right back up to the way it was yesterday. And the amount of data that I'm going to lose in the course of 24 hours with my setup is very, very minimal.

But again, there's a cost associated with having your primary server, it might be your Proxmox server, but then of course you're going to want a second piece of hardware for that backup. And then you've got hard drives and you've got storage and you've got... So there's a cost associated with that. I think that actually maybe now that I'm talking myself through this, maybe that's part of the reason. Like you said, backups are important and there's a cost associated with on-site backups.

And sometimes it's just easier to let Apple or Google backup your photos. Absolutely. So I want to keep moving along here, jumping to the next question. Select the operating systems currently installed on your servers. So again, multiple selections, you could have selected multiple options based on the fact that people probably have multiple machines or some people do.

So starting with the most responses, we have generic Linux, Proxmox, Raspberry Pi, OS, True NAS, Windows desktop, Unraid, Windows Server, ESXi, BSD, Mac OS, and then other is actually higher than the last few options, but I usually include that at the bottom since it's not a specific option. So just a couple of notes from facilitating the survey.

I think one of the biggest points of feedback I got as I was facilitating it is people were a little upset that I didn't break out all of the distributions of Linux and group them all as generic Linux, which I had contemplated, but they're just way too many to list. And I think that probably would have fractured the results a little bit and maybe hidden some of the other things we have going on here that aren't just generic distributions of Linux. 100% agree with that.

I've done both generic Linux, as you put it, typically with either Ubuntu or Debian. And then I've played a little bit around with True NAS. I need to look into Unraid. So many people have recommended me looking into Unraid. But what's funny now is like I mentioned, I use Proxmox, but in my containers, I just install Debian in a container and deploy Docker like I normally would on bare metal. But again, it's all about the backup strategy that Proxmox affords me.

So I'm kind of doing a hybrid thing with that. But the numbers that I'm seeing here don't really surprise me even a little bit as far as generic Linux, because like you said, there's so many different versions being the highest and then Proxmox being right below that. None of this really surprises me all that much, except for maybe Windows desktop being higher than Unraid. That one kind of surprises me. But that's only because I don't like hosting on Windows.

So that's my own judgment against hosting, is I just don't like hosting on Windows. Yeah, I'm a little embarrassed to admit that the first three or four months of my self-hosting journey I did try and host a server with Windows desktop. I had heard for so long, hey, use the operating system you're comfortable with. That's the best one for you. And so I've been using Windows my entire life. And so that's what I had used. And that was not the right option for me. I had an awful experience.

It was just not efficient. And so I ended up switching that particular machine over to Unraid. I've been very happy with it. And I'm also surprised to see how Unraid is kind of towards the middle of the pack, but it's really all those responses after the first few don't have many responses at all. I spend a lot of time for the newsletter that I facilitate each week, exploring it and looking for new software and new applications that are being released.

And inevitably, I feel like with all new software, there is a GitHub issue where someone is asking for someone to create an Unraid community application template for this software so they can run it. And so I think Unraid attracts a lot of new self-hosted users who want to use the convenience of something like Docker containers or something else. And it becomes a little bit of a crutch for them where they don't necessarily learn how to write their own run commands or their own compose files.

And as a result, very consistently, I see people asking for these Unraid templates. And there's nothing wrong with it, but it feels my experience out in the real world feels that Unraid should maybe be a little higher than it is in this survey. But again, I'm biased and the communities I frequent are biased. So it might be a result of that. Yeah, absolutely. I think the other thing that might be turning people off from Unraid, at least to get started, is the initial cost of the license.

I know it's not much, but I think if you're just getting in, you want to tinker around, you want to try new things, you want to have the lowest barrier to entry cost as possible. And I think that's why a lot of people have looked at Raspberry Pi OS or Proxmox or generic Linux or whatever. But one of these days, I'm going to put Unraid on one of my NAS devices over here and actually give it a go. Yeah? I'm excited to do it. Like I said, I've tried TrueNAS Core.

I wasn't a big fan, but honestly, when I started Open Media Vault, I wasn't a big fan. And it was just because I need to learn the processes for how each of those things work. And I think I've got enough stuff off my plate right now that I can dedicate some time to Unraid. So I think I'm going to sign up for the free trial, give it a go, see what happens. And maybe I'll start doing some Unraid content here real soon.

And the last point I want to make on this question, a year or two ago, I actually probably would have checked the box for Mac OS, and not because I necessarily used it for self-hosting the traditional software, but I'm an Android user.

And so I spun up a virtual Mac OS machine on Proxmox so I could host Air Message and Blue Bubbles and join the iMessage chat with the rest of my entire family who constantly nagged me for having the wrong phone and not being able to be included in their conversations. But it's a little hard to keep up with as the Mac OS upgrades come out, so I actually ended up abandoning it. Now I just have to listen to the complaints of my family every other day. Yeah. I also spun up a Proxmox VM.

I found a site that provided them, spun it up, messed around, was real impressed how easy it was. And it's still said there, and I haven't touched it in so long, but I love being able to use that to make people think that you're using an official iOS device to get the red-colored bubbles and whatnot. I think that's great.

Yeah. It sounds like maybe you and I might have had different experiences because that virtual machine was one of those, I somehow got it working, and now I'm never going to think about it or touch it because I'm afraid I'm going to break something. Okay. Anyway, moving on to the next question. Are you utilizing hosted cloud storage for off-site backups?

So prompting for this question was really every time I see people talk about off-site backups and how they do that, ideally, I think everyone would probably have their own physical hardware that they could host at friend's house or family's house or somewhere where they still have control of their data and they don't have to worry about prying eyes on it. But I think for many of us, that's unrealistic.

And so you'll often see things like Backblaze and Google Drive when it had its unlimited storage come up. And so I was really interested to see the results of how many people actually do that versus either not having off-site backups or having their own machine sitting somewhere else. And the responses are about split. There are more people not using hosted cloud storage than are, but not by many.

Yeah. It's like 45% and 55%, respectively, there for yes and no. And honestly, I'm ashamed to admit that I am in the no category here. Like you had mentioned, the ideal situation is to have a friend with a solid internet connection and no bandwidth limits and you drop a piece of hardware at their house and you just connect to it. But that isn't the reality for a lot of people.

And I think one of my next big projects, at least for my own home, not necessarily for the channel, I'll probably make content about it, will be setting up an off-site backup probably with either Backblaze or Amazon. I got to see which one's going to be, if I'm being honest, the easiest and most simple to maintain so that I don't have to put a lot of headache and time into getting these backups set up. Yeah. And this is one of those things where the cost absolutely comes into play as well.

I do utilize hosted storage for backups, but I very selectively choose what I'm backing up. I have a media server. I don't back any of that up. I think it's probably recoverable. It would take me some time if I lost it, but not worth paying $100 or more a month just to have backed up in the case that I do lose it. So I utilize actually Backblaze. I have an encrypted Arc-Clone drive connected to Backblaze where I shovel my passwords and stuff like that.

And I think my Backblaze bill is about $1.80 a month. Because it's really just a sensitive stuff. So even the couple of dozens of gigabytes of photos I have and passwords and other config files really don't take up a whole lot of data. So I'm like halfway there with my all-site backup strategy. I'm not all the way there, but I think I'm there where it counts. No, I dig that. I think that's... Honestly, I don't know why I hadn't even considered not backing up my media.

Like you said, I can reacquire that media through ripping DVDs and Blu-rays and that sort of thing. So I don't necessarily need to backup the however many terabytes of that media that I've got. But like you said, the photos, I've got my YouTube video projects. I should probably back those up somewhere as well. So I'm still going to have a few terabytes worth of media to back up.

But taking my Plex server out of the equation makes that all of a sudden way more manageable in my brain and of course in my wallet. When you're giving me that number or giving us that number as far as what you're paying for Backblaze, I think that's going to be my first foray into an offsite thing. I think I'm just... That was... I needed somebody to give me a good reason to pick AWS or Backblaze. And so Backblaze wins at this point. I'm going to give them a shot into what happens.

Yeah. I think this is probably a good segue into the next question. If you're utilizing hosted storage for backups, offsite backups, which provider do you use? So Backblaze, easily the largest followed by Google Drive, OneDrive, S3, Dropbox, Mega. And then we talked about this a little bit, David. The other bucket is huge for this question. It's actually larger than any of the responses.

And when I got the responses and started analyzing, I was really worried that there was an obvious selection that I missed and we were trying to think about it a little bit. And I actually don't even remember what we had come up with, but I'm still not sure we had a definitive answer for what that other bucket was comprised of. I think a good chunk of that had to do with what we were just talking about, having a friend with a solid internet connection where you could drop a piece of hardware.

I think that that's probably a good chunk of that. Or that maybe they're not using S3, maybe they're using the deep storage on Amazon. Or it's really hard to say without putting in a, well, if you clicked other, then fill in this box and then trying to parse that data. Sometimes other is just easier to parse. Absolutely. So we're one section in, we're moving on to our next section. I don't know what that means for our time here, but we'll keep moving along here. The next section are containers.

So a lot of people in the self-hosting world, or at least part of the self-hosting world that I reside in, like to deploy containers, usually Docker containers for self-hosted software. And so I included maybe four or five questions about containers. The first one, do you ever use containers to deploy self-hosted software? I think it's about 90 to 95% of the responses are yes, which isn't surprising, I don't think. No, not at all. In fact, we talked about this.

I actually just took something that was bare metal and switched it to a container just because it was easier to maintain. And obviously I think as more people are getting into self-hosting, they're realizing how easy it is to deploy a container and maintain a container and use a container. It's often easier than trying to deploy something that's 30 commands to deploy the same app application that you could do with a Docker CLI and just be done with it.

And I think containers are definitely the easier way to go with most things. Absolutely. And there's just a ton of support for containers out there if something does go wrong. So that makes them preferable as well, I think. There's also a few different content creators whose entire channels are about deploying containers. So if they've ever got questions, there's lots of video resources for that kind of thing. So just have to throw that in there.

In my newsletter, I always try and feature, I do a content spotlight where I just pick one piece of software and focus on it for the week. And as I started doing that section, I started to get some feedback way back when and almost always, it was people asking if I could add a blurb about how that application or software can be installed. And most of them were really just interested in whether it could be installed via container or not.

It's bananas, I think, how hyper-focused some people are on only using containers and are very adamant about sticking to software that's super easy to deploy and tear down what needed. Which I can't say I blame them, but it's interesting how passionate people are about them. Absolutely. Next question, do you deploy Kubernetes to manage your self-hosted software? The responses for this are almost the inverse of the previous question.

Most people don't, which I don't think is surprising because it's a little more advanced and complex. Any comments on that? I don't think we... I've never used them, so I don't have a whole lot to say. I'm looking into Kubernetes and the automation with Ansible and those sorts of things. But I feel like if you've just got a few devices laying around your house, even if it's in a rack or whatever, it doesn't take that long to log in, run your updates, and log out.

Or to spin up a new Proxmox container, I don't feel like there's enough... I'm involved in doing that most of the time to require a Kubernetes setup in a home lab. Now there are some people that do it for the sake of doing it, but I'm not that guy. I don't have that much patience with trying to learn that right now in my life. Yeah. That's probably all those education people from the previous question on why they self-host.

Yep. So jumping into that question, I feel like we've actually almost already covered this. Do you prefer container installations to bare metal? It isn't quite as skewed towards yes, although it still is somewhere around 75, 80% of people do prefer containers. And again, I completely understand why they're so easy to build and to tear down. And documentation is always just... There's a ton out there. People are always...

If it's not the developers themselves, there's someone if you Google how to deploy X using a Docker container, it's out there somewhere. Or YouTube. Right, absolutely. And I love the community that does that, that makes the content. And it's not even just me saying that as a content creator. It's looking at finding blogs, talking in depth, thousands of words about how to deploy a container and all of the little nuances to it. And these people aren't getting paid for this.

They're not being sponsored by these people. It's like, I love this so much. I want to spend my time helping other people do it. I think it's just a great community to be a part of. Jumping into the next question, if you ever use containers to self-host, which container platforms do you utilize? So I'm not entirely sure why I included this question. I remember coming across a conversation somewhere where people were talking about it.

And I know at some point I'm a Proxmox user and so inevitably containers in a VM versus an LXC, what's better, what should I be using always comes up. And so I thought it would be interesting to see the results. But it's by and far Docker is what people use. It's Docker, LXCs, Podman. And then I have other, which only looks like it has maybe 40 or 50 responses. So I'm not sure what else people are using for containers.

There's probably some other containerization software out there I'm not aware of. I know. I think there might be something on the Microsoft side like Azure, but I don't have enough experience with Azure to speak on that with any intelligence. But I've looked, obviously my channel focuses primarily around Docker. I've used some LXC stuff. I know a bit about Podman and its name. It is basically a root list Docker and I've never actually used it.

But again, these are numbers that I'm not at all surprised by, especially with Docker being so much farther ahead in the numbers than anything else on that list. I look forward to being educated in the comment section of the communities we share. That's when it launches. I'm sure someone will have something to say. One of the things that I've learned is even though I teach this stuff online, there is a good chunk of my audience that is smarter about this stuff than I will ever think about being.

They have forgotten more than I will ever know. Somebody today actually presented me with the idea of doing the vulnerability scan thing and I'm going to make an entire video on it. But they're just like, it's super easy. Here's what you're looking for. And it was just today it changed how I look at deploying a Docker container by one comment from a viewer on my channel. Interesting. I think that's why the self-healing community is so interesting. Incredible. Yeah, 100%.

Jumping to the next question. This is an interesting one. If you use containers to self-host, do you use the management platform to deploy them? When I initially asked this question, I was thinking Portainer or Yacht used to be a big one. I haven't seen much on that anymore. And the responses are almost evenly split. It looks like maybe it's... Actually, I probably should have pulled the data up. There's a 10 vote difference between these two. Oh boy. Yeah. I'm pretty close.

I actually think I was thinking about this. There are probably some Unraid users who are classifying themselves in that because Unraid almost pretty well is a container management platform. At least it supports managing containers more similarly to Portainer than probably someone who just spins up Debian or something and runs a Docker compose file for their containers. So I assume that's contributing.

But given the split again of non-technical versus technical people in the community, I suppose that's not surprising. I appreciate that this is fairly evenly split. When I first started, I actually learned Docker through Portainer. I learned... Originally, I went through Portainer and I would fill in all of the little individual boxes for what is the image, what is your path, what is your port. I would fill in all of these boxes. And I was like, there's got to be an easier way to do this.

And then I learned about Docker compose and of course then Portainer's version of that, which is stacks. And it was Portainer that gave me kind of the base knowledge to get into deploying a Docker compose or a Docker CLI without a graphical user interface. And if somebody wants to get in and wants to commit to self-hosting with Docker, I encourage them to learn both a GUI, a graphical user interface, as well as command line.

Because there are just certain things that are easier in command line to find or deploy or look at than it is in a graphical user interface. And I feel like you get a better understanding of what's going on if you do it in a shell or in a command line or whatever versus just, like I said, typing stuff into a box and clicking go. I feel like your understanding gets better when you get into the shell a little bit. Yeah. And I'm the same way.

I learned Docker through Unraid, as I had mentioned, Unraid early on, and now I use compose files to manage it. But I completely recognize that there is a place for software like Portainer and HomeLabs, even maybe for people like me. I know a couple of years ago they started offering free business licenses. Maybe they were limited per user. And it actually comes with some pretty cool features. I know they have indicators next to images to tell you if there's an update available or not.

And we're going to get into that with the next question here when we talk about automatic versus manual updates. And I guess this is probably about as good of a segue as ever to jump into that. Next question is, do you prefer manual or automatic updates? And this is very much like a yes or no question or one or the other. And in reality, I did get some feedback and it was fair. It probably isn't black or white.

And for most people, there are probably containers and software that are completely fine updating automatically every night if there's something going on. And then there are other containers that you absolutely would not back up, like a database or something like that. And so it was wrong to give users or responders in this survey just the two options and probably not like a both option.

Yeah, I think a both option would have been good just because speaking from experience, I am in the both category. Like you said, I will not have a database automatically update. I did that one time and I didn't have a backup and I lost everything. And the data was just gone. And since then, I've implemented backup strategies and I've implemented update strategies. And like you said, if my pie hole or my ad guard or any of the Rs, those just automatically update. I don't care.

If that crashes, I spin it back up and there's no data lost. But if my password manager doesn't get an automatic backup or an automatic update, those sorts of things, I want to go in and make sure that the update makes sense or should I wait till the next update to fix something else. So I kind of pick and choose what automatically updates and what doesn't. I think next time you do the survey, definitely have yes, no, or both. Yep, I definitely will.

And I think that's reasonable, especially given how many people are doing this as a hobby and maybe don't necessarily have the time to monitor every individual image they're running or they have a container built on. So it makes complete sense. Okay, moving on to the next section. Next one is networking. So we have probably four or five questions here. The first one being if self-hosting for personal use, do you deploy a firewall aside from the router provided by your ISP?

And I listed OpenSense and PFSense as options. And this was pretty, I'm not going to say even, it was definitely skewed more towards yes than no, but it was probably more, I know, 55, 45 or 60, 40 or so. More people said yes than I realized. And I think that's predicated on the fact that I built an OpenSense box last year when I moved into a new house and I found it incredibly challenging. I thought the learning curve was steep and it was much more difficult than I anticipated.

And I remember thinking there can't be that many people who are doing this because this has been really hard and my internet has been down more than it has been up for the last week. As I try and configure my five VLANs and everything else. Right. Well, and like we kind of mentioned earlier, you've got a family, you've got a spouse, you've got to keep all of those people happy while you're trying to configure this stuff.

And if you screw up a firewall rule, well, then half your network or your entire network is going to go down. Everybody's going to be upset. I've never actually implemented like an OpenSense or a PFSense. I've never implemented a dedicated firewall piece of hardware. However, I was on Unifi for a while and their security stuff is pretty good as far as blocking incoming stuff and DDoS attacks and that sort of thing.

But I had my own separate issues that a lot of people have with Unifi and I've since switched to an Omata setup. And they also got really great built-in protections that are basically a firewall that I didn't have to configure. And I can just say, if these conditions are met, block that traffic and I don't have to worry about that. And I can still do VLANs and I can still do some of that stuff.

But with this question, I would still answer no, because I didn't deploy a firewall specifically, but my software, my router has that kind of functionality built in. So, it's definitely one of those kind of yes, no, or something else, I think kind of things. But I don't know how you would word that something else. Yeah, I'll give it some thought, but I completely understand. I actually have Omata devices myself. I run it behind OpenSense. And at one point I was like, do I really need OpenSense?

Omata seems to have most of what I need. And at one point I considered abandoning it, but I stuck through and I am glad I did. But it's a good point, I guess, for anyone listening who's interested in something like that, but maybe doesn't want the complexity of a full-fledged firewall. There are self-hosted solutions that don't require going all out and becoming a networking expert, although I'm not calling myself a networking expert at all. I'm trying to figure it out.

Next question, how do you access your self-hosted software remotely? This is a multi-select. I included three options. There's VPN, there's reverse proxy, and then only via LAN, so only if you're at home. And so, VPN won this one, although it was very close with reverse proxy, and then not many people are doing it only via LAN. David, what are you doing with your services? I've got most everything available to me from wherever I need it.

That said, I'm actually, and we'll get into this in a future question, but I'm actually using Cloudflare Tunnels for most of my remote access stuff. And I've got several services that are for the community. I've got blogs, and I've got resources, and that sort of thing that are available to my community or anybody who does a Google search and finds it. They're just easy to access. They can go get the information they want. Pages load quickly, not a big deal.

However, I've got password managers, and I've got media servers, and I've got all kinds of other stuff that I can still access remotely very easily. But there are multiple requirements that have to be met, whether it's an IP address, like I've got my VPN IP address, my dedicated IP address there, and I've got my home IP address. And if you're not on one of those two networks, you can't access my stuff.

Or for certain things, it's an IP address, and you have to log in with one of my Gmail accounts as in the form of authentication to prove that you're me, and you actually have the permission to be there. So it's finding the right solution that works based on your needs, whether it's just publishing a website or locking something down for very critical reasons. And we talked about this a little bit.

I don't know if Cloudflare Tunnels would be a VPN or a reverse proxy or a hybrid of the two, but that's what I use for everything. Prior to that, I used NGINX Proxy Manager, just as a standard reverse proxy with ports 80 and 443 open, and that always made me feel uncomfortable. So instead, I hand my data over to Cloudflare and hope that they don't do something nefarious with it. Right. Yeah, and they're audited regularly, so you hope you don't have that concern.

I was sitting here as you were talking, I was like, did I make this? Why did I not include a multiple or both option? But it was multi-select. So I think there are people who probably, like you, do a multiple or have a multifaceted response to how they manage or strategy to how they manage. I would hope so. External access, which is somewhat similar to what I have as well. Sure. So moving on, do you use a custom domain for your self-hosted applications?

And this is heavily skewed towards yes, I don't know if it's maybe 80%, 80%, 85% yes, and most of them no. And then the next question, if you own a custom domain, who is your domain name provider? And so you and I talked about this a little bit. I remember being a conversation in the Reddit comments when I posted the results. Cloudflare is the largest provider here. And I know they do provide domain name purchasing and they offer them, but I didn't realize how popular it was.

Or I think maybe people purchased their domains elsewhere, pointed them to Cloudflare, and are maybe giving Cloudflare the credit in this survey for being their domain name provider or who they purchased it from. Which is maybe I should have elaborated better on the question.

Sure. And here's the thing, I think last year I made a few different videos talking about Cloudflare and then some other creators around the same time also started releasing a lot of Cloudflare related content for tunnels and certificates and things like that. And I think as a result of that, I think it kind of got more eyes on Cloudflare because they realized what Cloudflare could do for them. And I've been using Cloudflare for more than a decade now in different capacities.

But it's only been over the last couple of years that I've started using them in my home environment. And I think that people would like the opportunity to just buy their domain name from Cloudflare because then the DNS change if they need to do something as instant or relatively instant compared to going with kind of a no name registrar that might take 48 hours to update your DNS records.

I think if people can just can buy a domain name and do all of their DNS or their tunnels or everything from one provider, I think that a lot of people are leaning that direction just for the sake of simplicity. I agree. Overall, we had a note in the commentary, there were 192 individual providers submitted as responses to this question. I'm assuming they're all real providers. I didn't fact check all of them. So I suppose someone could have put in one to be silly and I wouldn't have caught it.

But even within I've charted maybe like the top 15, it looks like many of those I personally don't recognize myself. The top are Cloudflare, Namecheap, BorkBun, and Google domains, which I feel like I see mentioned frequently. So that's not surprising. But there really was a diverse set of responses to this question. It was really interesting. Maybe it's a regional thing. Maybe it's where you live, what country you're in, what on then. I think that probably plays somewhat of a role as well.

But that was a very interesting question. And maybe more interesting the next question, if accessing your self hosted content through a reverse proxy, which web server are you currently deploying? I love seeing this question, people coming in asking what they should use. And there are 10 different answers because there are a lot of really great web servers or proxy servers that you can use for your content. In our survey, NGINX won out, which isn't surprising.

I think a lot of people use plain NGINX. I think a lot of people use NGINX proxy manager. And I think a lot of people use swag, which is the Linux server's team implementation of NGINX. And so that's not surprising. No, not at all. And in fact, this is kind of one of those where we were talking about gatekeeping earlier, especially on Reddit.

If you use Cloudflare tunnels, there's a good chunk of people who will tell you that you have sold out, that you're not a real self-hosted because you're using somebody else's servers to transmit your data and they've got access to your data and they're looking at your data. And I think if it works and you're comfortable with it and you're not doing anything that's going to get you in trouble, do what works for you.

Like I said, I've been using Cloudflare for a long time, even before they had tunnels, even before they did domain names. They were a DNS provider for my website clients for years and I've always trusted them. I've never run into any nefarious stuff with them. I've never had an account. In fact, I used to make my clients use them to obfuscate the IP address of my server from their domain as another layer of protection for both of us.

And that's why I've been using them for so long is they've always worked and I've never had an issue with them. Yeah. I mean, I think the company and I think tunnels are great. I don't think the dig on anyone using them personally myself either. I think no one can self-host every aspect of their infrastructure. So even if you don't consider that self-hosted, I don't think that makes you a not a self or makes someone not a self-host. Right. I 100% agree with that.

In the responses, traffic was second, which we talked a little bit about. I think traffic is for masochists or maybe people who are a little more willing to put in the time to learn. I tried it once and struggled and ended up with caddy and I love caddy. I use it for my internal search so I can get HTTPS at home without having to have a public sub domain that can be accessible outside of my network. But again, all of these are great.

The order for, I guess for anyone not looking at these charts, it's Nginx traffic, CloudFlare tunnels, caddy, Apache, HA proxy, and then there's a small other bucket. So a lot of interesting responses there. Yeah. So we've arrived to the next section, so we're the second last section. We are making it through this. And second last one is software. So some more targeted questions, maybe specifically about software and types of software. The first one, are you self-hosting your own email server?

This one was surprising and only that there are a lot of people who said no. Most people, it looks like maybe 80% of the responses were no. I'm surprised not more than that were no because I feel like I always see people discouraging others from self-hosting the email, given the complexity. Yeah. And I think this goes back to your previous comment of you kind of have to be a masochist in order to actually self-host an email server in a home lab on premises, whatever.

There are so many different things that can go wrong and it can take a considerable amount of time to fix them for things that are outside your control. If somehow your IP address gets compromised, you either have to get a new IP address or you have to go and prove that you didn't do that. And both of those can be time consuming or costly.

And again, like I mentioned earlier, if you can find a solution that isn't hosted, if it's a third party, whether it's Google or any of the email providers that are out there, if it works and you're comfortable with it, that's the route that I like to go because there's a time and place for certain things. And if you've got that much time and effort to host an email server, I'm proud of you. But I don't.

I don't have that much brain space to keep on top of a server like that and make sure that it keeps working and I don't miss an opportunity that I might get via email. It's just, it isn't worth it to me. Absolutely. There's just a lot of important things flowing through email and missing out on something is definitely to me also not worth the risk.

What I find hilarious is when someone online in whatever community will ask if they should self-host email, you'll always get maybe 90% of the answers of people saying, hey, don't go through the effort, it's not worth it. And then you'll always get one user who comes in and is like, yeah, I've been self-hosting since the 80s and I've always self-hosted my email. I spend five minutes a week doing it and I have no problems with it. You should be fine. And it's just hilarious.

There's always that one person who doesn't seem to have those problems, but they're probably the one in a million. They have a very special mindset that it just makes sense. It's kind of like that, it's in the movie A Beautiful Mind where all the numbers just kind of flicker across the screen and that it just, that's email hosting or people who self-host email, they've got that going on, I'm sure of it. Yes, I agree. Moving on to the next two questions both about media streaming.

So the first one is, do you deploy any self-hosted software for purposes related to media streaming? And I tried to keep this broad because there's a lot. I know we talked about Plex and Jellyfin for movies and TV shows, but there are a lot of people who are streaming. Books, I would even include books in there. They have servers for their books, they have music. And so I try to keep it pretty broad and it looks like it was maybe 75, 80% of people said they are, which isn't surprising.

I think we mentioned, I think media streaming is somewhat of a gateway drug into self-hosting. And so I wonder how many of the people who said no maybe did stream media at some point and maybe just no longer do it because they lost all their storage and didn't feel like building it back up at some point or something. Like I said, I think media streaming is like you said, the gateway drug that gets you into everything else. And it just, what is it like 30?

About 30% people said no. And I'm honestly surprised it wasn't like 10%. Same. And then jumping into the next one, if self-hosting media, which media server do you deploy to stream video content? So I tried to target this one a little more. And so number of responses, Jellyfin, Plex were by far the largest. Jellyfin looks like it has about maybe 150 more responses than Plex. And then I did include Kodi.

I don't necessarily consider it a media server, but I think a lot of people use it in place of Jellyfin and Plex because they don't stream their video outside of their house. And then I have NBN, then there's another option. I know there's some other media streaming servers. This was surprising. I thought I was under the impression that Plex was probably the more popular of the two.

And in our instance, and it might be the bias showing of the communities I shared the survey with, but Jellyfin seems to have won the lead in this survey. It's funny, out of all of these that are listed, Jellyfin is the one that I actually haven't tried. I tried Kodi 10, 15 years ago on an old PC that I had laying around and then switched to Plex when I got into self-hosting. And that's what got me into self-hosting.

And then I had issues with Plex and I jumped onto the Jellyfin or the MB train and ran that for a couple of years and then started having issues with that. And then was just like, you know what, let's just go back to Plex and see if they've fixed it. And I haven't had any issues with it since I switched back about a year ago.

And like I mentioned earlier, I think people are very interested in the free open source software and that's where Jellyfin comes in, is it's very easy to do what you need to do with it. However, like we mentioned in a previous conversation, there aren't as many apps for it or for TVs and smart devices and things like that. So it's a bit of a trade-off. I think I'm honestly surprised like you are that Jellyfin is higher than Plex here for that reason alone.

At the end of the day, I think they're both great pieces of software. They both, they're pros and cons. So I suppose I would just encourage anyone who's weighing between using the two, just try them both out and pick whichever one suits your needs the best because they are both good. Next question, do you deploy any self-hosted software for smart home or home automation purposes?

I ask this because I know Home Assistant is huge in the self-hosting community and so when I engage, are there people engaging with Home Assistant or any other automation software? I know Home Assistant comes with a bunch of integrations and there's software that people use to support Home Assistant, but also on its own like Node-RED or some of these other automation tools. And this one was skewed towards yes, but not by much. There are maybe 55, 60% of people do and then the rest don't.

And I suppose this isn't super surprising. One I think the home automation aspect of self-hosting is intriguing. I think that seeing things people do with it can be inspiring and tempt people to jump into it and it's probably a bit of a gateway drug as well along with media streaming for people who are getting into it.

Also because you can do it on super low-powered devices like Raspberry Pi or some of the hardware Home Assistant I think has traditionally recommended is not more than, I don't know, $60, $75 or something to get started. And then maybe, and I'm sure the no is probably a combination of people who just hadn't considered it or who just don't want to go through the effort because there's a lot of effort behind home automation. Yeah, I think there's a lot of effort that's involved in it.

And I think also when you've got ecosystems, whether it's Google Home or Apple or Alexa, or whatever, and people are, they've got their Google phone attaches to their Google Home Assistant that attaches to everything in their house. I think for, again, trying to keep the family happy, trying to keep the members of the family who are using the service happy, I think sometimes as simple as better. But it's also pretty neat when you can manage your entire house from a single application.

That's definitely where Home Assistant would come in. But I like to keep things super, super simple here in my house, just to keep the stress levels down. Completely understand. Next question, are you self-hosting a game server of any kind? So this was a yes or no question. I don't know, maybe 20, 20, somewhere between 20 and 25% of people have or are in some capacity and the rest aren't. I fall in the no category. I've never done something like that.

I've thought about spinning up a Minecraft server, it seems like it'd be pretty easy, but I've just never gotten around to it. But it's more common than I would have guessed at 20 or 25%. Yeah, I have. I've made videos about deploying some certain or some different game servers. I have deployed my own game server so that my friends and I could all play on the same server without having to have the rest of the world involved.

I know when I game, I do it to escape people and going out and playing the Fallout online or WoW or any of those types of games adds the stress back that I'm trying to escape. So I like having my own game servers that I can get on and game by myself or game with just exclusively the people I give it access to it. As long as, of course, as long as the game supports a self-hosted server.

But I think it's great that people are doing this so that they can have their private time with their friends, whether they're hanging out in the same room or they're on the other side of the world. I think a self-hosted game server is a very cool way to hang out with your friends and share an experience together as kind of nerdy as that sounds. But I guess that's why we're here is because we're a little bit nerdy.

Yeah, I wonder if there's a chunk of people, I assume they're always users who monetize their game servers. I think it's a good way to self-host and also bring in some income if you have people willing to pay to make it onto private and well-moderated servers. So I wonder how much of an aspect there is of that. But that in itself would be daunting to me because I'm always worried about uptime with my users who don't pay me for anything I provide.

I don't necessarily owe them uptime, but I'm really anxious about it. And so if I had people paying me for my services, I think my anxiety levels would be out of this world. Yeah, I mean, you'd have to have... Obviously, you'd want a UPS to make sure even if your power goes out, that it stays up. But then you also need a backup ISP just in case one of them goes out, you've got the other.

And it's just, you start accruing that cost that we talked about at the beginning of this and doing that kind of stuff. You definitely have to weigh the cost benefit of, is it worth my time to put this much of my effort into making a few extra bucks a month? The next question was more of a personal curiosity as I was building the server. I don't think we need to spend a whole lot of time on it. It's just simply, how do you prefer to manage self-hosted databases?

And so I think when you and I were talking about this previously, the conclusion we had come to is just use individual databases. Yeah. But there are a few instances, I think, where it makes sense or is of sound mind to share them. And given they typically aren't very taxing on resources, especially for people who are newer to this, probably stick to individual databases per application. Yeah. And that goes back to the backup thing.

Again, I'll try to make this quick, is that if you've got an application that crashes and you have to restore it and it compromises that database, well, any other application that is on that same database could also be affected. So compromising one application could take down your entire infrastructure if you're sharing a database with too many different applications. So that's to speak to your point, that's why all of my databases are separate instances.

So if something goes down, I restore to the last known good point and it doesn't affect anything else on my network. The next question, I think we could probably spend an entire episode talking about, what is your favorite self-hosted application or software? This is one of the free text responses. We had just over 275 individual pieces of software applications mentioned. And within the visualization and the results on my blog, I did not list anything with less than 10.

So there probably are about, I don't know, 15, 20 listed here. And then maybe just like top five or six, we had Home Assistant, Jellyfin, Plex, Vault Warden, Nextcloud, Piehole. I don't think any of those are surprising. Those are probably, if I sat down and jotted down what I thought the top 10 would be without running a survey, I'm sure all of them probably would have made it into the list that would have created. Yeah, absolutely. And I think I agree with pretty much all of that.

I'm actually really a big fan of Bookstack. I'm glad to see that one landed on there as well. That's one of the resources I've got for my community, hosted in Bookstack, and it always just works and it's so easy to use. Yeah, they're all super interesting. Definitely encourage anyone who's listening. If you're looking for new software or things to run, and especially if you're new to the self-hosting community, check out the list.

I think everything on the list I have probably tried at some point or another, and they're all very solid pieces of software or applications. Totally agree with that. David, we made it to the last section. This is the final section. All right. It's not actually related to self-hosting in any sense of what we've been talking about so far, and it's just a section on demographics. I don't think we need to spend a whole lot of time on any of these questions, but some of them were interesting.

Some of them weren't as interesting or weren't surprising. The takeaway from this is the average self-hoster looks to be in the age range somewhere between probably 20 and 49 years of age. Those were the larger age buckets. They tend to be male, although we did get other responses that weren't nothing. They are pieces of the pie chart that you can barely see. They're not super tiny slivers, and I think that's really encouraging and great to see.

Users skew towards the US, which I don't think is surprising. We talked a little bit about my website statistics versus yours as well. Coming behind the US is Germany, which is always surprising to me when I see how many visitors I have from Germany. Probably a lot of interesting reasons behind that, but I don't even think I list it here. I chart anything with 10 or more responses, and so I think I have maybe 15 or 20 on the chart.

Then I provide a table that's collapsed underneath with everything else because I thought it was interesting to see all the countries of the responders to the survey. The list is long. There are a lot of countries, and some of them only have one response, so only one person from the survey was from that country. It's still really interesting to see how diverse the crowd is.

I would say as a total percentage, the United States is actually not as large of a percentage as I would have expected given especially the fact that my site and the survey were written and provided in English, and I'm in the US, and a lot of my communities that I frequent and posted this to are probably more US-centric. I'm really surprised and encouraged to see this. Yeah. We talked about this earlier, OpAir, looking at, like you said, United States, Germany, United Kingdom.

Those numbers line up almost exactly the same here as they do in my YouTube analytics. As far as the differences in these numbers, I think it's in that same order with approximately the same differences, and it's interesting that it's consistent across types of media, the numbers of people who are hitting these types of communities and that sort of thing.

Unless you actually come in and look at this, this collapsed menu you talked about that only had a few responses in all of these different countries, I think I counted quickly there was 54 additional countries in here that responded. Some of these places, you wouldn't think about it, but of course they're there. Why wouldn't they be? It's interesting to see that this community is literally everywhere, and I just love seeing that that's the case.

Yeah. It's not surprising, I feel like when I'm going through GitHub Read Me pages, I'm constantly seeing people or developers asking for help with translations and trying to make their software as available as possible. Clearly, the demand is there, and so I just have to remind myself, I need to make sure when I produce my own content or write that I'm keeping that in mind, that the audience is much, much broader than sometimes I'm led to think it is.

On the flip side, the next question, not surprising at all, if employed, do you work in a technical field? I know maybe 75% of the people yes, and then the rest no. I think that makes sense. Obviously, people who are technical by trade probably skew towards a technical hobby, but we do know, like myself, there are a fair amount of people who aren't technical who do this as a hobby for a number of reasons.

Yeah, and I think it's great, like we talked about earlier, I love that it's getting to a point where we're getting into a self-hosting hobby. It can be inexpensive and the barrier to entry with the knowledge that's available has become so much lower. There's so much information out there on how to do so many different things. I'm loving the fact that there are non-technical people who are getting into this, who are doing it for whatever their reasons are. I think this is great.

I think obviously, as more time goes on, we're evolving into a more technical existence in life. I think people have to start becoming more technical just to get through a day, really. Whether it's on their phone or their tablet or their computer or whatever, technology is not going anywhere and we have to learn to embrace it. Absolutely. Speaking of phones, next question, what is your primary mobile device's operating system?

About 70, 75% are Android, the rest are iOS and there's a little sliver of other. We were chuckling at this the other evening. We weren't quite sure what was comprising this other bucket. This is a little surprising only in that I would expect most technical users to be skewed towards Android. When I actually look at the technical visitor statistics of my site, it's actually fairly even between the two.

I also have a fair amount of desktop users and I'm wondering how many Android users who work in that technical trade are maybe digesting my content on their work computer or their desktop machine at home. I'm not sure if that plays into it. Again, I can only speculate with the data set I have. Yeah. Sorry, I just jumped over to my dashboard just to see if it would give me any idea about devices that people are using and not without digging into it further. I think you're right.

I hear a lot of people, even in my discord, people be at work chatting and they're in the technical field. There's probably a pretty good chance that they're also watching YouTube videos on work devices, computers, laptops, whatever. I think that's probably a pretty good chance of what's going on there. Yeah. The second last question, again, we were joking about this. I think I answered this as a question just to shame people a little bit.

Have you donated to a self-hosted project in the last year? We discussed earlier, I think it's really important to support the developers, programmers who are delivering the software and content to the self-hosted community because we rely on them for most of what we have or a lot of it. So interesting, the yes responses, I don't know, are maybe like 30% and then there's a large chunk, maybe 60, 65 are no. And then I did include a response that was prefer not to answer.

And I did it for many questions. This was the only one that I actually included in the visualization because I thought it was interesting in trying to decide whether those prefer not to answers were yeses or noes. And I think I have a hunch, but I certainly won't speculate out loud on air. Yeah. No, I think it's about 5%. I think they preferred not to answer. And I think then maybe they were just trying to be a little like, look at me, I'm funny, which I get.

But I think also, I think like you said, maybe they don't want to answer because of the shame or the disappointment. I don't want to put words in people's mouths. Maybe they're just not happy with their decision to not have donated yet. I'll just say it that way. Yep. And then rounding out the survey here, the last question we have, what is the highest level of education you've completed? I don't think this is super surprising.

I think it actually kind of skews along with the population in terms of levels of education. I don't know that there are a lot of inferences you can make looking at this data about the self-hosted community, but it's here and feel free to hop on my site and view the results if you're interested in it. David, did you have any insights or reactions when viewing this? No. I mean, most everybody these days has a bachelor's degree.

And I think master's degrees are becoming more popular with certain job fields requiring it for whatever the reasons might be. I don't think I'm actually surprised by anything other than the master's degree being higher than some college or no degree. I think I would have expected those two to be reversed, but I think beyond that, I'm not surprised by anything that I'm seeing in this. Well, with that being said, that is the final question of the survey.

I think it took us a little longer than I had anticipated, but we made it through. David, thank you so much for joining me today. It's been a pleasure getting your insights on some of the responses to the survey. For anyone interested, I will leave a link in the show notes to David's YouTube channel. If you aren't already a subscriber, please go check out his content and subscribe. And again, thank you for being gracious enough to sit here with me and chatter about some of these responses.

Again, thank you for reaching out. I really appreciate this. It's nice to be able to have, like we mentioned earlier, two different sides of the community here. The technical but not technical. Of course, you're more of a numbers guy than I am, so I couldn't have put something together like this like you did. I don't have the wherewithal to put together all the charts and the data and parse all the information.

And I've really enjoyed chatting with you and being a part of this and getting to look at the data. And hopefully, maybe in the future, we can do this again sometime with that, whether it's another survey or a different topic or whatever. I'd love to come back and do this again sometime. Absolutely. Thanks again, and thanks everyone for listening. Happy self-hosting.