2025 Selfh.st Survey Results: 4K+ Self-Hosters Speak! (w/ Ethan Sholly & Matt Foxx)

push that separately.

It shouldn't end broadcast. Okay, that's

really strange. Maybe. There we go.

Okay. All right. Let's try this again.

Uh, hi everybody. Thank you for joining.

Um, we're we're experiencing some def uh technical difficulties completely because of this guy right here. Uh,

hopefully you guys can hear me. If I can get an audio check, that would be a great way to get things started. Um,

just because uh streaming on Discord is harder than I thought. Um, so if we can get an audio check, that would be cool.

Um, hopefully everybody can hear me.

Audio is 100% okay. Outstanding. All

good. Awesome. Okay. Well, thank you guys so much for hanging out while I figure out this nonsense on my end. Uh,

this is kind of our annual thing that we do where we take a look at the survey uh that Ethan puts out from the self-host newsletter. And uh I think without

newsletter. And uh I think without spending too much time uh going on with uh with my nonsense, I'm going to uh switch over here and uh we're going to

jump over right there and uh welcome everybody to the live stream. Ethan, if

you'd like to jump in and uh run the show here, that'd be awesome.

>> Yeah, sure. Thanks, DB, as usual for hosting the live stream. Some of the technical issues we had getting up and running is the exact reason I don't like to host this live stream. So, thanks for letting me leverage your resources.

>> Of course, >> DB, I I think you and I touch base very regularly throughout the year and this is pretty much the one time we chat other outside of just Discord chat. So,

uh first off, very nice to to speak face to face or as face to face as as video is. This year, we're switching it up a

is. This year, we're switching it up a little bit. Uh we also have Matt Fox

little bit. Uh we also have Matt Fox joining us. Uh we figured uh maybe

joining us. Uh we figured uh maybe everyone might get a little bored of hearing uh DB and II drone about the same results every year year-over-year.

So Mix It Up asked Matt to join us.

Matt, for those of you who aren't familiar with him, is the developer of Multistcrabler. It's a self-hosted

Multistcrabler. It's a self-hosted application for scrabbling music, listens from various platforms um into a centralized place. Uh that is of course

centralized place. Uh that is of course how he asked me to introduce him, but I know him from a lot more. If you click into his profile, he has a blog where he

writes amazing blog posts that I include in the newsletter every week when I can.

He also uh is a bit of an afficionado.

You'll find his uh home lab diagrams and networking uh occasionally shared to Reddit and always blows my mind. He's

been he's become somewhat um I don't know if he even knows this of a confidant for me in the last several months. if if something pops up during

months. if if something pops up during the week that I'm trying to include in the newsletter and I don't know much about, I defer to him often. So, Matt,

thanks a bunch for joining us today.

>> Oh, thanks for thanks for having me.

Glad you let me weigh in today.

>> Hopefully my opinions aren't uh too strong.

>> Yeah. Yeah. Looking forward to it.

>> Hey guys, I need to pause for just one second.

>> Sure.

>> To fix something that uh that Matt pointed out to me when we did this the other night. Uh, and that is that I

other night. Uh, and that is that I didn't have the right um the right thing going here. And of course, it's going to

going here. And of course, it's going to be like that, isn't it? Oops. Nope. I

should have I should have just left it alone. Um, Discord, why? Okay. Well,

alone. Um, Discord, why? Okay. Well,

then let's Oh, we're Oh, god. There.

Let's just do this. You know what? Why?

Oh, come on. Are you kidding me? It's

not letting me. Oh, because I'm stupid.

That's all.

It won't. There we go. Okay. Uh, sort

of. Sorry. I was hoping I could get this window to pop out and hide everything else, but now Discord isn't giving me the option um to to switch to the uh to

the thing that we just had window capture Discord the self-host. There it is.

Okay, I think that's better. Okay.

Sorry, guys. I just I wanted to get rid of all of the other stuff that was in the window that was extraneous. Thank

you for your patience. Have at

>> Yeah. So, we have a lot of good results uh from this year's survey that we're going to jump into today. One thing I do want to take just a really quick sidebar before we start. If any of you follow uh

DB throughout the week on any of his socials, uh you probably know that he recently hit 100,000 subscribers here on YouTube. Congratulations. And I believe

YouTube. Congratulations. And I believe that came with a nice special delivery of some sort this week. I don't know if you have that >> handy or >> I um I I got I got sidetracked on

another project. It's still sitting in

another project. It's still sitting in the kitchen, but I did get my silver play button. Um real excited about that.

play button. Um real excited about that.

Um thank you to everybody who's helped uh get me get the channel to that point.

Um real real excited to be there and um and be and be kind of part of a community uh that we've built over the last several years. So really excited about that.

>> No. Well, congrats again.

>> Thanks.

>> With that being said, I I think we can jump into the survey here. Uh just a a quick intro, I guess, before we jump in uh too far. The this year's survey was

facilitated by uh using a platform called Formbbricks, which was new to me.

They also sponsored the survey, but I usually try and find a sponsor survey that or someone to sponsor that also has a platform I can self-host a survey just to get let people or familiarize people

with the various platforms out there.

Um, and I I you know kind of sponsorship set aside had a fantastic experience with Formbricks. I think it was the most

with Formbricks. I think it was the most successful survey I've had so far in terms of people having issues with the technology behind the survey. there's a

lot of complex logic going on behind the scenes. When people selected one answer,

scenes. When people selected one answer, I tried to divert them to another question if the remaining ones weren't pertinent and and stuff like that. So,

super happy with form bricks. Um, I I'll try and talk to some of the nuances as we go through the survey. I know in some of the social posts I had shared yesterday after publishing the survey,

um, had some questions about when certain people were and weren't asked specific questions based on previous answers. So, we can chat about some of

answers. So, we can chat about some of this. But without further ado, let's get

this. But without further ado, let's get started. Uh, first question, uh, why are

started. Uh, first question, uh, why are people self-hosting at a very high level? And so, obviously, responses were

level? And so, obviously, responses were personal, enterprise, or both. As usual,

every year, this is the first question I kick off with. The trend looks almost exactly the same, at least in terms of my audience. Most people are

my audience. Most people are self-hosting for personal purposes. the

there are just under 700 people who are doing it from an enterprise perspective.

So I assume they either have small businesses they're running where they're self-hosting for those purposes or they're working for another company and they're self-hosting that company's

infrastructure as well.

That jumps in that that leads into the next question and this is kind of the first caveat I have. Um in prior years I had just let everyone answer this question and there's always confusion.

Oh, when I when I select the answer, the response for how many unique users I serve, is that for my personal infrastructure? Is that my enterprise

infrastructure? Is that my enterprise infrastructure? And so, specifically,

infrastructure? And so, specifically, only people who selected personal were served this question. And so, we're looking at the results in front of us.

They're a little different than they had been in prior years, surprisingly. Uh,

previously, people uh the two users was the most popular answer, and that was overtaken by one user this year.

Everything else I think trends somewhat similarly. I don't know if it's just the

similarly. I don't know if it's just the additional people who answered the survey this year were just a bunch of people who only serve one user. Uh but I was a bit surprised to see that pop up.

We can hear you. Yeah, he's he's working on it.

Try that. Uh, okay. All right. How about

now? Can you hear me now?

No sound. Um,

>> is there no sound?

>> Still no sound.

>> I'm not sure.

>> You're good. Okay, that's that's that's really strange. Okay. Well, uh, as I was

really strange. Okay. Well, uh, as I was saying, I think um I was comparing what I was trying to say anyway is that I last year I compared these the numbers from this year to last year and some of

the some of the changes didn't surprise me, but I think some of them did. Um,

and I think we're seeing more and more people getting into self-hosting, uh, for a lot of reasons that we'll cover later, but I think I think because we're getting new people in the community, the the one user doesn't necessarily

surprise me. Um, that it jumped so much

surprise me. Um, that it jumped so much from last year.

And I think part of that is also just people jumping from one to two users is a is a huge deal especially if you're you know it it goes from being a hobby where things can break and and be down

for an indefinite amount of time to somebody actually is depending on you for for something and uh you know going back from you know the the the partner approval factor or the friend approval

factor of this is you told me this would work and now it's not to okay well I'm just I'm >> I'm backing off on that and I'll I'll figure it out kind of thing is uh I'm sure partially that that vector as well.

>> Yep. I agree with that. Yeah.

>> Yeah. I think this is also a tricky question because if I think of the number of servers my Plex server uses or Jellyfin or whatever, it's a vastly different number than the number of

people who are utilizing my Volt Warden instance or something like that. So,

there's a little more nuance to it than I think and I'm not quite sure how I can capture that. Completely open to

capture that. Completely open to feedback. Um, and I I I also like to

feedback. Um, and I I I also like to plug in the YouTube description of this video, there is a link to my survey feedback form. So, if at any point you

feedback form. So, if at any point you hear something we chat about or or have some feedback about the survey itself, feel free to click that and leave some feedback, I will absolutely take it into

account for subsequent uh surveys.

>> Cool.

>> So, jumping to the next question, reasons for self-hosting. This is also a tricky question because I think I give too many answers, too many responses and I think if I take this survey on any

given day, I I answer slightly differently based on just some of the nuance between the the various answers.

But the top two this year are the same as they were last year, hobby and privacy, which I don't think is surprising, especially given so many are are self-hosting for personal purposes.

It makes sense that those two would surface to the top. I know a handful of my containers are are privacy focused.

As much as I love the hobby of self-hosting, I'm also very privacy focused and you can see that in my infrastructure for sure.

>> Yeah, I think when I started um it was a lot of it for me was cost. Um but as we've gone forward over the past five or six years, a lot of it I think it went

from cost to hobby and then to privacy for me. um cuz I needed something that

for me. um cuz I needed something that wasn't Netflix for a number of reasons.

Um and then I started making content and it turned into a hobby and then from there um just the way the way everybody's handling data these these days terrifies me as far as what they're

collecting whether we like it or not.

>> I was was surprised convenience was so high in this list. Um I don't feel like self-hosting is a convenient thing to do at all. And can you think of any

at all. And can you think of any instances of software that's that is more convenient in the self-hosted instance than it would be as a cloud service?

The only thing I can think is that and cost and convenience are are surprising that they're so much lower than hobby and privacy because

you know for for as much as we kind of like to avoid this topic every now and then uh piracy is such a large reason for self-hosting and acquiring media and cost and convenience I think are two of

the drivers for that. It's the cost of not paying for a million treating services and then it's also the convenience of being able to find everything in one place. So, I I kind of link those two together. I'm I'm

assuming that might be what drives that, but it's a really good point.

>> Uh self-hosting for me is not convenient in any capacity.

>> Well, Colins actually brought up a good point in chat saying that it's convenient once you have it up and running.

>> Like once once the infrastructure is built and everything is running smoothly, then it's convenient.

>> Um until an update breaks something. But

that that's my own personal little addition to that.

>> Yeah. Thank you. answering my own question as I thought about just a little bit there. I' I'd say probably people think convenience, they probably think Pi Hole or some other kind of site

like networkwide ad blocker that's extremely convenient, but it is as Colin said >> convenient once it's up. Getting there

is is maybe less convenient.

>> Yeah, absolutely. Jumping to the next question, we have hardware options. This

has changed a little bit since last year as well. Uh especially relative to the

as well. Uh especially relative to the change from last year versus the prior year. Top two options are still consumer

year. Top two options are still consumer hardware and low powered devices like a Raspberry Pi or something. They actually

have swapped spots this year. So low

powered device uh took number one spot last year. This time it's consumer

last year. This time it's consumer hardware, low power device following after it. The one I wanted to uh call

after it. The one I wanted to uh call out specifically was uh the desktop PC.

I notic that's kind of falling the rankings throughout the years. I don't

know if if if it's just the familiarity people who are reading my content and responding to the survey have with self-hosting. They get a little more

self-hosting. They get a little more complex into uh or into the more complex hardware. Everyone kind of starts out on

hardware. Everyone kind of starts out on their desktop PC, turns it on when they needs it, when they need it, turn it off, and then realize that's a little too inconvenient, jump to the server, and then from there it's just kind of

hog wild with all the options.

I also wonder if people aren't having uh as much um just as much hardware laying around as we did say five years ago. I

think with the way the prices are with, you know, with consumer or PCs, uh graphics cards, those sorts of things. When you're building your own

of things. When you're building your own custom PC, it's getting more and more expensive. And I think people are less

expensive. And I think people are less likely to have spare hardware laying around from a recent upgrade. Um whereas

you can pick up, you know, a UG green NAS for a few hundred bucks, throw a couple of drives in it, and have your your infrastructure up and running in a lot of cases cheaper than using an

old desktop PC just because of how expensive parts are right now.

>> Yeah. Uh this was this was definitely the I'd say the most interesting question for me on the whole survey. Um

I mentioned this to you a little earlier. Um, I would really like to see

earlier. Um, I would really like to see a year-over-year change on this kind of information because as things like uh Penglin become more convenient and

easier to install and there are more VPS offerings specifically for self-hosters like Rackner and stuff. I I would be interested to see what like the the

change on cloud VPS versus low powered devices or consumer hardware changes because that becomes an option that is easier or more cost effective for

people. I'd also be interested to see um

people. I'd also be interested to see um what people are considering to be low power devices versus consumer hardware because I imagine people think low power devices like Raspberry Pies, but they

might be putting like those thin client Lenovos or HPs like off of eBay in the consumer hardware or in the low powered devices. It's it's a little vague, but

devices. It's it's a little vague, but it'd be interesting to see what the difference between people thinking, oh, Raspberry Pi is for self-hosting versus I'm going on eBay to buy old

old uh you know, office hardware. Like,

what's the what's the difference there?

>> Yeah, that's >> Yeah, and I think you >> Go ahead.

>> Yeah, I I think you actually hit on two of the questions Jake uh Collins had brought up in the chat. Uh asked

specifically about the change in VPS, and it actually was pretty substantial year-over-year. CloudVPS was 1300 almost

year-over-year. CloudVPS was 1300 almost 1400 sponsors this year and >> 97 last year.

>> 97 last year.

>> That was 97 for cloud VPS last year.

>> Penglin is a huge percentage of that.

>> I think it's >> Penglin jumping in. Also, there are a lot of services like Pikapods. They they

sponsor me a few times throughout the year and various things. I think those are what people where where they're kind of like oneclick install applications.

There are a bunch of them um that I can't think of off the top of my head. I

think people are also classifying those as as cloud VPS VPS as well in this category.

>> Uh Collins also asked what the difference was between um a desktop PC and consumer hardware. I think what it's a really good question. I should

probably explain a little further next year's server. I think in my mind like a

year's server. I think in my mind like a desktop PC is exactly what you think it is. A desktop PC. Consumer hardware to

is. A desktop PC. Consumer hardware to me would be uh I guess like pre-built hardware that's not quite enterprise hardware. I'm thinking like Dell

hardware. I'm thinking like Dell Optiplex like the Lenovo Think Center something that's not >> so like an old workstation.

>> Yeah, something something along those lines. But it's a it's a really good

lines. But it's a it's a really good point. I should consider that for next

point. I should consider that for next year how I how I can rephrase that to make sure I'm capturing how various people view those things.

>> Yeah.

>> Yeah. Yeah. And then John brings up a good point as well. Uh, as far as Nooks being low powered, are people classifying those are low powered or are they sticking with things like Raspberry Pi?

>> Yeah. Are you talking about low compute power or low like actual wattage power?

You know, it's a it's a it's a large category.

>> Yeah, absolutely. Something to consider for next year, right? But along the hardware front uh network attached storage just again always include this question how many people are deploying

to NAS for whatever purpose I I think just the trends in across other questions it's pretty easy for me to tell that the people are doing this are also the people who were oriented towards the media streaming questions

and responses to other questions and so obviously that's a big thing but there's a lot more than just media that people are storing the NAS on that was actually kind of my first big build when I got into self-hosting and trying to find

some way to store things, whether it was my own personal photos and videos or if it was other content I had. So, not

surprising to see there are so many more people who deploy than don't deploy.

When what's I think what's interesting to me is like the numbers from this year and last year aren't all that different.

Um like this year we've got, you know, like 20 2700 that do. Last year was only 2,300 that do. So, it's like 400. And I

wonder if that's just that many more people having answered the survey because the the number of no responses

was only like 40 different. Um so I just I wonder if part of that bump has to do with just more people answering the survey the this year.

>> Yeah. And Collins is bringing up another good point that chat. I think I just need to hire him to help with next year's survey. Uh how many people

year's survey. Uh how many people branded NAS versus a do-it-yourself NAS?

And this is actually a really good point because my NAS is also a server. The

only difference between it and some of my other machines is that it has 10 hard drives in it. Um, and maybe, you know, some hardware to support all that storage versus the other. But aside from that, when you look at the number of

services I'm running, it doesn't really differ from my other uh machines.

Hollins, then let's talk.

So, uh, jumping into the next question.

insensitive data. Uh this isn't surprising. Um also isn't uh different

surprising. Um also isn't uh different or too different from prior years. How

many people are avoiding self-hosting sensitive information? I have listed

sensitive information? I have listed here and kind of my takeaway with each question I've been including for security and backup reasons. A lot of people provide some context behind their

answers and that seems to be why the people who are avoiding self-hosting the sensitive information aren't self-hosting it because they either don't trust their their backups or they don't trust that their infrastructure

can keep things like their passwords or their sensitive photos um safe.

Yeah, >> I want to go ahead >> I want to emphasize what you what you just said not not to jump ahead in the survey at all but the the amount of

people here who are not avoiding but then also only have zero to one backup of their data is is a bit a bit unnerving as uh as Scarlet in the chat

said there it is pretty scary to see the lack of backup but are also hosting things that are clearly or you know trying to save things that are clearly you know you should have more than one backup up of is that

>> right?

>> It's a bit concerning.

>> So, next question. Um, operating

systems. This is almost always the question I get the most feedback about because I never list every specific distribution or flavor of Linux or

>> what other operating systems people are using out there. This year I I was very specific in trying to target the the really like home lab kind of media

center geared operating systems. That's why you see things like open media vault, truness, unrade, the things that I know are absolutely popular in the the crowd that I wanted to capture. I

understand so many of them are Linux based and could be categorized as Linux.

Um I think last year I phrased it as generic Linux to try and separate the two. But I did I did leave a note in the

two. But I did I did leave a note in the in the commentary on this survey. Um

based on user feedback, I will include a subsequent question in next year's survey that allows anyone who selects Linux to actually select or provide

further information about their actual flavor and then we can hopefully provide some helpful and interesting stats on that as well.

>> Nice.

One of the things that I'm noticing, uh, I've got I've got last year's survey open in a separate window here and, um,

disc station manager from Synology. Um,

the the num that number has is is four times what it was last year. Uh, which

honestly kind of surprises me after the stuff that Synology pulled this year to to see that number go up so much. It

genuinely shocks me with as angry as people were at Synology for for their different reasons. Um, to see that go up

different reasons. Um, to see that go up so much just it just kind of stood out to me.

>> Yeah. Yeah, I almost need to include a question that's something like did you participate last year so I can tell is it are these trends across people who are you know the same people who didn't

use Synology last year are they this year or do I have a different subset of people that seem to be using this because home assistant OS is also very similar if you look at last year's

survey there were 16 responses of people who selected home assistant OS or the home assistants operating system in this question and this here. It's the third

most popular option with 1,200 responses. I don't I I don't run Home

responses. I don't I I don't run Home Assistant OS. I run Home Assistant in

Assistant OS. I run Home Assistant in Docker because I'm just I enjoy chaos.

But I I don't think Home Assistant uh OS has changed that much since last year that would that would have done anything to encourage mass adoption other than if they have some hardware that maybe they

they've released in the last year that might have encouraged people who are buying it to install it directly on there. I have no idea what the trend is.

there. I have no idea what the trend is.

I'd be interested to hear.

>> I I think a big part of that is there are so many more companies that are um that are classified as works with home assistant um to help kind of ease that

that transition into committing to it because there's I think people are seeing that like companies are taking this seriously now and I and I feel like they feel like they'll have fewer issues getting this new piece of hardware to

work. So I think they're more willing to

work. So I think they're more willing to commit to to a dedicated system like that.

Yeah. Uh, congrats to them though. Home

Assistant's doing great things. We're

going to talk about them more in the survey. So, it's cool to see that that's

survey. So, it's cool to see that that's growing in popularity. I'm sure that's helping a lot. It does streamline running the platform. So, I'm sure with it being a lot of people's kind of first

foray into self-hosting. I'm I'm sure it's it's made a world of a difference to have that ready and, you know, completely different from the old days of YAML and all that other stuff that you need to get Home Assistant up and

running.

And then Matt, I think you just chatted about this. Uh you brought this up a

about this. Uh you brought this up a little bit ago, backup strategies. Um

how are people storing or capturing their backups? Uh and then again, I kind

their backups? Uh and then again, I kind of just dug into the data a little bit on my own on the back end and noted that of the responses only 34% store data in

a single location.

I think there might be an element of people again I'm going to lean into the the like Plex like sonar radar crowd

that a lot of their data is data that's completely recoverable like that would be easy to get back if it's just a bunch of movies and TV shows that they're comfortable with maybe like an offsite site backup somewhere and then in the

worst case scenario where they don't have anything like it's not the end of the world like there are places to get it online I assume that's playing a role here. But honestly, I'm not 100%

here. But honestly, I'm not 100% positive.

>> I guess there could also be a a uh some missing data here since there's not a question on how many backups you have.

You could have on-site, but you could have three copies on three different, you know, backup drives. So, um that might be missing from this as well.

>> Yeah, it's a really good point. Yeah,

that's that all all of my backups are onsite and I know I know you're supposed to have at least at least one off-site and I haven't done that yet. Um, but I think u when we get to our next question

and and I say this every year um you know people are people are using the off-site option and I really need to join that crowd.

>> Yeah, I do a little bit of >> Oh, go ahead, Matt. I have a uh I have another friend who's also light into self-hosting and and you know Plex kind of stuff and we have we are backup

buddies. We each have a a little

buddies. We each have a a little container that holds hard drives and foam and we we switch them out every like every six months or so. And I keep mine at his and he keeps mine at at

keeps his at mine. So uh it's it's nice to have people around who are also into the hobby.

>> That's a really cool idea. I dig that a lot.

>> Dream scenario actually. That makes me wonder if a good future question would be, do you know someone else in real life who selfish?

>> Yeah, >> cuz I'm not sure I have someone in my life that I can do that with.

>> Yeah, it's definitely a it's definitely a privilege to have someone who's, you know, he's like five minutes down the road from me. So,

>> Oh, nice.

>> H very jealous. Very jealous.

>> Uh, next question digs into those uh those who selected cloud in that backup question. who are you using? There are

question. who are you using? There are

always a ton of responses and I try my hardest to kind of go through and streamline because everyone spells something slightly different and it's hard to aggregate things and I think I've captured everything. I know the

other bar is huge. I almost thought about just completely removing it from the chart and then just making the note down here. Um, but there were a lot of

down here. Um, but there were a lot of options and I wish I had included how many other pro providers are in this other bucket because it's a lot. Um, and

it's a lot of providers I've never heard of. But when I come to the non other

of. But when I come to the non other responses, all of those that had 50 or more responses, none of them are super surprising. I think Back Blaze and

surprising. I think Back Blaze and Google Drive are have been at the top as long as the surveyy's been around. It's

not surprising because they're two big names, but it it is really a nice mix of what I would call enterprise level services versus consumer offering. So

very smartly people are leveraging you know their Google Drive storage that they get for whatever reason subscribing to some other Google service or similar for one drive with Microsoft. Um so not

super surprising.

>> Yeah. I um some some of these I've never heard of. Um and Collins brought one up.

heard of. Um and Collins brought one up.

PCloud.

Never never heard of PCloud.

He says it always screams scam to him.

>> Yeah, I just I've never heard of it. And

it like it's I guess I'd have to do some more investigation, but I think year over year when when we review these, Back Blaze is always at the top. Um and

and I really just need to bite the bullet and figure out how much it's going to cost me to to have a backup of my stuff there.

Yeah, I use I used Back Blaze myself and I think it when I had done a cost analysis in the past um at least across some of the more enterprise level, they made the most sense price-wise. I should

probably revisit it, but I think there's a reason you hear them spoken about so much. So,

much. So, >> um and they they of course have uh other stuff. They they publish their quarterly

stuff. They they publish their quarterly drive stats and stuff like that. So, I

think they've also done a they do a lot of other uh content and interactions with the community that have built that trust there. So, again, not super

trust there. So, again, not super surprising.

>> Yeah, >> I I think I know the answer to the to the question, but I think it would be interesting to see what percentage of people are backing up everything versus, as you mentioned earlier, like just

sensitive data because obviously you don't want to back up, you know, 4 terabytes of of Plex movies when you can just redownload them again if you need to recover. Um, I mean, I think it would

to recover. Um, I mean, I think it would be a very small percentage backing up everything, but it would still be an interesting metric to see.

>> Yeah, I've got a fairly small amount of stuff, you know, like my phone backups, those sorts of things, my wife's phone backups that I'll hit my NAS back here and just kind of live there. Um, and I think that for me would be the most

important thing to have an off-site backup of just in case. But like all of my media, >> it's whatever, you know, and and because

I I recently um downsized my home lab, I could just if I had to uh if something happened, I could just grab that box and run and I would have my entire home lab

in a cube and uh it would be it would be it would be feasible to take it with me.

whereas like the the 45 the yeah the the the HL15 not super friendly to move around. Um so yeah, I think I think if I

around. Um so yeah, I think I think if I could just back up my my family photos and videos and that sort of thing, that's all it would really need to have offsite.

>> Yeah, I think another interesting question would be, okay, for those who back up their data, how many people test their backups with some sort of regularity?

is be a very scary number, >> right?

>> Absolutely.

>> I can't tell you I I'm like very uh very prescriptive about my backups. I do

everything very manually. The number of times I've mistyped a a a database password or a SQL dump command and I've just backed up an empty SQL file for

weeks until I went back and noticed is more times than I can count. So,

>> uh if you're listening and don't back up or test your backups, please test your backups. Yeah, it's it's a lot of

backups. Yeah, it's it's a lot of discipline for a hobby.

>> Yeah.

>> Okay. So, jumping into the next section of questions, thus the the color change here. Um, containers. Containers are so

here. Um, containers. Containers are so popular at least along amongst the the home lab crowd. I I would I would assume in in professional environments they're they're popular as well, but I can't

personally speak to that. Um,

unsurprisingly, the overwhelmingly people use containers in some capacity in their hobbl infrastructure. It's

interesting to me. I I still come across people on the Fediverse or Reddit who are adamant about not using containers and I'm not quite sure I get it and and

they're probably a lot better at bare metal and other things than I am that that they don't mind it as much. Um but

sometimes whether a a container is available will make it or break it whether I deploy a project or not.

>> Yeah, absolutely.

>> Unfortunately.

>> So I guess my question here was there was there some sort of um I don't remember. Um was there some sort of a an explanation? Are we do you think

people were just talking about like Docker containers? Are we talking about

Docker containers? Are we talking about like Linux containers as well as part of that?

Yeah, that's a good question and maybe I should elaborate in the question next year. I do

year. I do >> there there's another question that asks specifically what like container technology people are using, right?

>> But that's only for people who have already selected yes to do they deploy containers. So okay,

containers. So okay, >> if if there are people who are thinking this is one versus the other and clicked no, then they would have missed that question, >> right? Okay,

>> right? Okay, >> so good call out.

Um, next question is, okay, if you deploy containers, do you do you prefer them to bare metal? Um, most people do.

Uh, it looks like there's there's a fair amount of people who it depends, which I think makes sense. I think some people like to run databases bare metal and then everything else in a container or for whatever reason. So, um, I think

that's that's fairly similar to responses from prior years.

>> Yeah.

>> Um, container orchestration. Uh I was surprised by this question. The the

amount of people who are using an orchestration platform is much higher than than I expected. I I would have expected fewer than 10%.

>> Yeah, this was this was um we something I pointed out earlier as well is I'm wondering how what the percentage of was there a question on this last year as well? Um, yeah, there was and there was

well? Um, yeah, there was and there was also a fairly >> fairly the same amount of people and I'm wondering >> how many of those are from

>> methods that simplify the orchestration like K3 or um like Talos OS where it's just a full VM that you or you know a full operating system that just does

everything for you and all you have is the control plane because it is it seems fairly high considering how difficult Kubernetes can

Absolutely. I I I think in a previous

Absolutely. I I I think in a previous year I had asked if people were what they were using and I cannot remember what the results were but but that may be something for next year to ask what people are using if they are using some

sort of >> platform for that. Uh the next one container platforms um another repeat question very similar results Docker and

Lexe um being the top two responses. Docker

overwhelmingly are what people are using and then of course as as things like Proxmox continue to grow, Alexe uh continuing with strong representation

there as well. Do either of you use Podban at all? By the way, >> I've looked into it. Um and and I've heard that it's more secure because it's

rootless. Um but as far as actually

rootless. Um but as far as actually deploying it, I haven't.

>> I I also haven't. And I've also looked into it um pretty pretty extensively.

Um, but it's it it just seems like a there's a much higher barrier to entry to use it and it's uh, you know, a lot of the I have heard the security as well, but there's a lot you can do on the Docker side to to get pretty close

as far as mitigation as far as I understand it. So, I mean, it's it's a

understand it. So, I mean, it's it's a good it's a cool technology, but it just seems not as easy to use as an end user.

Yeah, it's interesting that when I see people talk about it, they they speak about it as if it it's a no-brainer that you should use it instead of something like Docker, but I I've just never

chased it myself personally.

>> Yeah, I I I I've seen the same kind of sentiment, but then uh I I don't know. I

think these people are they're just smarter than us maybe, and they'll realize that how much more difficult it is, you know? like one of the one of the main arguments is always that it's um it's demonless that it doesn't need like

a a background service running but then that requires you to understand systemd well enough to set up like quadlets and stuff and like what's a quad lit and how does that how does that translate to

docker compose and then by the time you have that set up you're you're basically running a demon with systemd anyway so there there there is a lot of uh a lot

of proitizing around it and kudos to M for for being more secure, but I I it seems like a false equivalency to say that it's just as easy as Docker.

>> Yeah, that's a lot of complexity for people who just want to share their movies with friends and >> right >> automate their smartome hardware.

>> Yeah.

>> Uh next two questions are management software. So if you deploy containers,

software. So if you deploy containers, do you use some sort of software platform to do it? Um this is actually fairly close to a 50/50 split and not

super surprising. Um, one because I know

super surprising. Um, one because I know Docker, Docker Run and Compose can be intimidating to people when they first start. And then two, there are just a

start. And then two, there are just a lot of good platforms that that streamline this very easily for people.

I'm thinking even something like um Unrade. And honestly, I don't even know

Unrade. And honestly, I don't even know if I mentioned Unrade as a platform, but it almost is because you don't deploy containers in the same way you do if you're running Docker Compose or just

Docker run commands. Um but the the the top three selections uh received most of the responses for what platforms are people using Portainer, Dockage, and Komodo. Um I

don't think that's super surprising to me. Portainer continues to dominate the

me. Portainer continues to dominate the responses. Not sure if that will always

responses. Not sure if that will always be. Um I listened to uh John and the

be. Um I listened to uh John and the servers at home guys talk about this a little bit yesterday. Um and Portainer is a great platform. will always have support because it's also

enterprise-based, but it was also kind of like the first one in the space. And

so I think some of its popularity will always it'll always retain that for that reason.

But I think one of the one of the issues with with Dockage is that it hasn't been updated in eight months.

Um, Lewis Lamb has been working on um, Uptime Kuma 2 for a really long time and has focused a lot of his energy there.

And I think as a result of that, uh, Dockage just kind of just had to sit in in an idle state for a really long time.

Um, so I'm really hoping now that um, now that Uptime Kuma 2 is more ready for prime time, I guess, uh, that some more effort can be put into keeping uh,

Dockage updated. I think that would that

Dockage updated. I think that would that would be a huge help for those people who want to step away from Portainer because they are very like enterpriseleaning even in their their

community edition stuff. Um, and I actually really like uh Dockage, but I can't trust its security if it hasn't had an update in eight months.

>> Yeah, absolutely. Uh, Matt, I know you're you're a Komodo guy, right? And

it's funny, we were chatted about that uh and I saw some comments earlier when we we brought it up. The I worked with the Komodo you guys a little bit on on or connected with them a little bit earlier this year. And they actually

recommended I connect with you, Matt.

This was before >> we had ever spoken. So, uh those guys know you and think very highly of you if you weren't aware of that. Um not that the rest of us don't, but um I I know

Komodo is I'm glad to see Komodo on the list. I'm glad to see where it is

list. I'm glad to see where it is because Komodo is relatively young compared to some of these other platforms. I think Dockage is too. I

don't think Dockage has the maturity that Portainer does. Um, but I would expect next year to see that Komodo grow number grow a lot because it's doing a lot of cool things for a lot of people.

Yeah, I I won't I won't gush too much about it, but uh there's a the V2 release is is soon for Komodo and it's gonna it's going to really blow V1 out of the water as far as what it can do

and it's it's really becoming more of a full uh home lab management system with like dual like by connection between periphery and and core like kind of

taken over as as like an SSH agent and terminal stuff. Wow. just great software

terminal stuff. Wow. just great software and it also um from my experience really forces you to think about deploying and configuring your home lab in a way

that's scalable that uh you know dockage is is fantastic for getting people right into it with a UI that makes it easier.

Um, and Portaino's Portainer's got a little bit of that scalability in in thinking about your your your whole topology, the infrastructure, your home lab in a way, but Komodo really

I don't know, it's just it's a whole new way to think about it and it's it's great.

>> Yeah.

>> Yeah. And I see in the chat uh just Elixia mentioned Arcane. I actually have that in the commentary down here because it was one of the other more popular responses. just didn't quite cross the

responses. just didn't quite cross the threshold that I felt should be charted.

Um, but was one of the few that I isn't really on my radar or if it if it if I've seen it, it's just in passing in the past and I meant to check into it because enough people brought it up. Um,

but it still beat out some of the other popular options like Uno Host and Cloud John which are on there as well. Um, so

I'd be anxious to check that out as well for sure.

>> Oh well, one more thing about I wanted to mention about Komodo. Sorry not to get on too hard.

>> Sure.

>> Who didn't know that V2 is coming? Um,

it's going to support Docker Swarm, which is huge. Oh, nice.

>> I think it's the only I think Portainer also supports Swarm to some extent, but it's definitely of the things that are listed here, it's the only other tool that's going to support Swarm. So,

that's pretty huge.

>> Wow. So, that might also take up next year's orchestration platform number as well.

>> Wow.

>> Yeah.

>> Yeah. It'll be interesting to see if there's a correlation between those two responses.

>> Yeah. Um, and I think that's actually a good leadin into the next question about update methodology because I've seen I know there's a really great blog post that's made its rounds around the

communities I frequent about using Komodo and its capabilities to automate how they're how they're updating based on and I and I I have not um used Komodo

extensively myself so I can't really speak to it but I know it's it's uh been a gamecher for a lot of people but just kind of taking one step back and looking

at the overall results. I'm fairly happy to see the results as they are. Uh most

people are doing manual updates. Um

which I prefer anything to blind automatic updates. I don't think automatics are bad as as long as you can do some sort of version tagging where you can ensure that automatic updates

don't take you to a breaking version like if you can tag specifically with v2 or v2.1. Um and maybe there's that

or v2.1. Um and maybe there's that that's a good differentiation for next year's survey. But uh I think the people

year's survey. But uh I think the people who are uh just flying by the seat of their pants with with Watchtower uh I think that crowd is decreasing more and

more um every year.

>> Yeah, I definitely follow or fall into the both category. Like if it if it's say my Adguard container, I don't care.

it'll take me two minutes to spin it back up and run an Adguard sync back to the setting of the one that didn't crash on me. So, I think there there are

on me. So, I think there there are definitely certain containers that I don't care if they crash. Um, but yeah, for for the most part, I do I do almost everything manually.

>> Yeah, partially. I think I mean m I'm also mango for the most part and I I think less of it is because I don't trust like watchtowwer or komodo auto update but like you know you never

developers don't have any they can sever or they can not and even even then they can break versioning and you just don't know if like an actual minor update is going to be breaking changes or not. So

it's always good to to review before you update.

>> Mhm.

>> Yeah, that's a good point. the the

versioning only makes a difference if developers know how to use it properly.

>> Completely rebuild their database schema on a on a two instead of a full release.

>> Yeah. Yeah.

>> Sounds like that may have happened to you recently.

>> Not not recently. Um but it was what definitely put me off from doing automatic updates is is I had something um that I was just like screw it, it's fine. And then it wasn't and I lost

fine. And then it wasn't and I lost everything because I didn't have a good backup strategy in place at that time either. So I I had just set myself up

either. So I I had just set myself up for failure across the board.

>> Yeah, sometimes experience is the best way to learn unfortunately.

>> Um jumping into the next section. This

is all about networking. Uh, Matt, I know you had offline some questions about the firewall question here and and I probably should I either drop it or elaborate on it more next year. I think

last year I asked what people are running if they're running something like PFSense or Open Sense or or what they're running on their routers. But

it's really just trying to see how many people have have jumped into that networking game where like one minor change is going going to completely wipe their internet until they figure out how to get it back versus just like the

simple port forwarding feature from your, you know, Verizon router or whatever.

So, I'm an open sense guy. Um, I've

brought my network down an unfortunate amount of times, but uh, I I do feel better about my self-hosted infrastructure after having gone that.

So, gone that direction. I I would definitely recommend it if people haven't looked into it.

>> I I'm I'm not much of a networking guy.

Um, it it's it just doesn't make a lot of sense to me. So, I'm actually really glad that I've got a unified system that I just I tell it kind of what I want it to do and it just does it. Um, and I

don't have to worry about trying to configure a lot of that stuff because like I've looked at OpenSense and PFSense and it just it is it is a completely foreign world to me when I

get into that level of things.

>> I'm also in the same boat of of networking is I understand it pretty well, but I prefer to leave that to the experts because I'm already I've already got my fingers in so many things. like

the last thing I need is is for myself to to shoot myself in the foot on the internet. So, uh

internet. So, uh >> I I actually reminds me I have I have um I followed a guide to bypass my AT&T router like a months ago. So, I'm just

going straight to the fiber drop from my UniFi gateway and I don't think I have a backup of that configuration. So, you

know, I got I got, you know, egg on my face for not having the the the really important stuff backed up at this point and meddling the network thing that I really shouldn't be meddling with.

>> Well, best of luck to you for that.

>> Right.

>> I uh I I'm primarily for my for my job a remote worker and so I it makes me a little nervous every time an open sense update comes through because if my network goes down, I also can't work,

which is also an issue. So, I obviously save those updates for the weekends.

But, um, someone really quickly in the chat going back to the the update, someone made made a really good point about maybe a future question. How do

people get their updates or how do they know when releases drop so they know when to update their containers? That's

a really good question because I know a lot of great tools have popped up in the last year. Um, I use a command line tool

last year. Um, I use a command line tool called Dock Check that I'm absolutely in love with, but I know there are a dozen others that work really well. Um it'd be interesting to see how the the community

is split across those.

>> Yeah.

>> Um remote access jumping to this uh a very opinionated opinionated area in self-hosting.

Um I I feel like this and open source get the most people riled up whenever I speak about them in my the intros to my newsletters. Uh but asking people in

newsletters. Uh but asking people in this first question, how are you accessing services outside of your LAN?

VPN and reverse proxy are the top two.

Unsurprisingly, a strange number of forwarded ports. I don't know if that's

forwarded ports. I don't know if that's like you forward your Plex port because like for the most part that's relatively safe. It's never been a huge issue for

safe. It's never been a huge issue for anyone or if those are just people who uh are are yoloing it and don't and don't quite care.

I've seen a lot of people on Reddit when somebody's like, "Hey, how do I access this remotely?" the number of people

this remotely?" the number of people that just forward the port and they don't they don't they don't give any qualification of but be careful when you

do that because it's just forward the port and and and and so few people are willing to say hey look you can do that but um and I feel like if somebody

doesn't have the answer of how do I access this remotely they're inherently not going to understand the security implications of just opening ports on their network and When I see those

posts, for a while I would comment saying, "Hey, that's a bad idea. There's another

option." Um, and the number of people that would just jump down my throat because it doesn't matter. It's not a big deal. It's I just I quit responding

big deal. It's I just I quit responding to those posts on Reddit because the number of people that are just like, "It doesn't matter just blew my mind."

>> Yeah. If you um if you use software like any of the the the R suite like sonar radar in the last few years they've pretty much forced people to start using

the built-in authentication for it used to be completely optional but so many people were forwarding their ports. I

remember I don't know probably like five or six years ago you used to be able to go onto Shodden and just do a search for like port 8989

and all these sonar instances would pop up. I specifically remember the Rs and I

up. I specifically remember the Rs and I I don't know why and and I and like I I promise I never did anything nefarious, but like you could go in you could you could just very easily from the showdown

results pull it up and see just to get the sonar view and access that you has have as if you were logging in today.

And so they they've since changed that and like I'm absolutely certain that's why because you could still do a showden search today and still see all of them but now they're at least protected by

some sort of authentication. So,

>> I I have also seen the kind of things DB mentions on Reddit for just forward your ports, but I've also I feel like anecdotally in the last year or two seen way more of a move towards just use tail

scale or just use Penglin, which is which is exciting. like tail scales penetrated more into our little community and like while Penglin's not a like it's not a

a cure all for everything in the selfhosted lab it's so easy that it's it's easier to say oh just install Penglin on like a a $5 a month VPS than

it is for people to recommend port forwarding. So I'm I'm really excited to

forwarding. So I'm I'm really excited to see that kind of software out there now making people more secure by default without even realizing it.

>> Right. But that kind of goes into what um what was said earlier about that being a kind of a contentious question because if you're using somebody else's VPS, if

you're using if you're using a third party thing, you're not self-hosting.

>> And those people need to kick rocks because >> Yeah.

>> because there there's there's never going to be a one-sizefits-all solution for for well anything. But but but for the for people to be that pretentious about it,

it's just an elitist thing to do and I don't get it.

>> Yeah.

>> Yeah. I And it's funny because Collins mentioned in the chat, he made kind of a similar joke about Cloudflare as well because Cloudflare and Tails scale are the two that pop up when people really

lean into the oh you're not self-hosting and and like to me like sure there's an element of it. you know, both of them went down this past week and so people were impacted and that could have been

avoided if if you had been using other services. Um, but I think at the end of

services. Um, but I think at the end of the day, like we can't self-host every aspect of our infrastructure. And so I think it's perfectly okay for people to to decide what their level of comfort is

and how far they want to take that. And

for some people it's trusting tail scale and having to rely on their turn server or whatever to get their access knowing that you know if their open sense router goes down and everything else is up then they can't access it till they get home.

Like I get it. I I don't necessarily use all those services all the time myself but I understand why people do and I wouldn't give people flack for it.

>> Right.

>> Yeah. It's the you know don't let the >> don't let perfect be the enemy of good.

You know it's it's a lot of steps. It

can be incremental.

>> Yeah. There's people in the chat been saying it's a lot of gatekeeping that we we really should be discouraging. And um

I also want to just comment that Scarlet I am I am one of those weird people using Netbird and I also self-host it and uh it is a bit it is a bit

masochistic to try and get it set up and it is a bit weird. Yeah.

>> Um so the next question is interesting.

I asked about reverse proxies and bringing up Cloudflare is interesting because I realize I don't know when people when people are using Cloudflare tunnels. I I think their responses might

tunnels. I I think their responses might span VPN and reverse proxy. I think

depending on who you ask, they see them they see the way Cloudflare tunnels work in terms of not having to forward any ports. You know what it truly is. So it

ports. You know what it truly is. So it

does pop up down here. I have no idea.

I'm assuming some of its popularity is impacted by the fact that some people who use a VPN and don't use this weren't prompted with this question and didn't get Cloudflare tunnels. But

>> um EngineX Proxy Manager took the lead.

A little surprising to me because it is uh I know it just had a relatively large update, but it feels like one of those platforms that hasn't been actively developed as much as some of these other platforms have, but it's stupid simple

EngineX. So, I get why people use it.

EngineX. So, I get why people use it.

Um, EngineX is always like the default config when developers provide an example of how to proxy their services.

So, I'm not surprised to see it and all of its other implementations like uh Linux server swag is actually lower on the list than I would have expected. I

don't know if that's decreasing as things like traffic and caddy are popping up, but um it's cool to see the distribution. I'm a Caddy user. I love

distribution. I'm a Caddy user. I love

seeing it here. I don't think anything on this list is more simple to set up than Caddy, and I'll I'll fight you on that. even Cloudflare tunnels. It is

that. even Cloudflare tunnels. It is

just a dead simple config. Um, if you can get the DNS part up and running and working properly. Um, I've tried traffic

working properly. Um, I've tried traffic before. It is complex. Uh, I got

before. It is complex. Uh, I got frustrated and stopped and haven't thought about it, but I know it's probably something I should revisit at some point in the future.

>> Documentation.

>> Oh, sorry. Please, please go ahead. I

would say, yeah, traffic feels complex, but I I feel like they need a they need someone on documentation over there because it's really just their documentation that is the the barrier.

It's not nearly as as friendly as it could be. Once you once you get once you

could be. Once you once you get once you get the rind around traffic, it's like, "Oh, that was actually very simple to do." Just trying to connect the dots in

do." Just trying to connect the dots in their documentation is uh a herculean task. Unfortunately,

task. Unfortunately, >> I um I made a couple of videos on traffic fairly early on um when I when I was when I first started making content

and I've actually got a a thing still hanging on my wall from years ago where I was a traffic ambassador um and and then I discovered other

platforms that I I felt were easier. And

and I think my biggest complaint with traffic um at least last time I looked at it, like I've got all of my services split up

over each service gets its own LXC.

So to to try to manage that many remote hosts via traffic was a nightmare. Like I know you I I believe you can connect to a remote

host, but EngineX proxy manager just made it so much easier. I could point to I could point to my TV with EngineX proxy manager and and and it would be fine,

but trying to get traffic to communicate with a remote host used to be a nightmare. I don't know if it's gotten

nightmare. I don't know if it's gotten better. Um, but if you start with

better. Um, but if you start with traffic as as your your foundational point and then add labels to your containers as you're deploying them,

it's it's it's just stupid easy for everything to just automatically come up with labels. Um, but I'm not that

with labels. Um, but I'm not that forward thinking. So my my reverse proxy

forward thinking. So my my reverse proxy is always the afterthought once I've got all my infrastructure built.

>> Yeah. And something I I didn't mention in the commentary that I probably should have. I there's there's a popular fork

have. I there's there's a popular fork of EngineX Propy Manager that's called MPM Plus. I think there are just some

MPM Plus. I think there are just some people who wanted it to move a little quicker than it was and developed their own fork. I think I consolidated

own fork. I think I consolidated responses in this in this uh in the survey results here because it wasn't always obvious to me if someone meant just the the vanilla version or the

fork. So that might also be skewing this

fork. So that might also be skewing this a lot. Um but Matt, I know you've

a lot. Um but Matt, I know you've brought up Penglin a couple times.

Really cool to see them this high on the list because I'm I like I'm very positive Penglin just launched this year. Like I don't think they've been

year. Like I don't think they've been around >> before 2025.

>> Yeah. Yeah. So, it's really cool to see them so high on the list being so new.

That's not surprising. I expect them to be much higher next year >> as well.

>> They're Yeah. And their their pace of development is insane. Like the

>> they went from just that one pengal into now they have like a whole tunneling software and it's it's great. They're

they're they're doing big things over there.

>> Mhm.

>> Yeah. Yeah. Very cool. Next couple

questions are about domains. I don't

think we need to spend a ton of time on them, but really just curious, are people using custom domains? Are they

just bookmarking, you know, local IP addresses and ports? Most people are using domains, and it's a it's really a good mixture of people purchasing domains, whether it's from super cheap

registars like Cloudflare, Namecheep, Pork Bun, but also kind of like the dynamic DNS services like Duck DNS where they just give you free subdomains for your account. Um glad to see people are

your account. Um glad to see people are using that and then hopefully leveraging HTTPS along with it.

>> Okay.

>> Um I will note very specifically this year and you know throughout the throughout the the the form bricks platform that I used throughout the survey I could mark questions as

required or optional. And when it was optional it there was text on the question that said optional. And um I still got so many people who felt forced to answer the optional questions because

they didn't see that they were optional.

And so as I as I'm about to caveat this that I left a note about Cloudflare specifically, I'm wondering if people actually read it now, but but I very specifically said do not check Cloudflare unless you use them as your

actual registar because so many people purchase their domain names elsewhere and then just bring them into Cloudflare for DNS and all the other fun stuff with Cloudflare. Um, so with that being said,

Cloudflare. Um, so with that being said, it still blew the others out of the water. I'm still not quite sure, it's as

water. I'm still not quite sure, it's as popular as this indicates because when I see people recommending registars, it it I feel like Pork Bun and NameCheep

are the two big ones. Like I see them pop up so much. I I see Cloudflare, but it's hard to believe. I'm I'm still uh hesitant that these aren't skewed a little bit by Cloudflare's DNS functionality.

Well, >> I mean, I go ahead. I say I I I believe it. Um uh as as Jay Collins just said in

it. Um uh as as Jay Collins just said in the chat there, uh I was also a Google Domains user and I moved everything to Cloudfare because it was all just right there. And I imagine if you're, you

there. And I imagine if you're, you know, someone tells you about Cloudflare tunnels or you're just trying to set up DNS for the first time and you see that registered domain in the sidebar, it's

it's just so convenient to have it all right there. So, um I'm I mean it makes

right there. So, um I'm I mean it makes sense to me that people are using it for everything.

>> Yeah. and and like if you look at last year's results um it kind of follows the same trend um

from like it's the same top three results this year to last year um with with fairly similar growth numbers as

far as percentages are concerned um and I think yeah I I 100% think just if you can buy your domain at Cloudflare and use Cloudflare

DNS or tunnels or whatever and have everything in one central location. I

think people are definitely looking for the simpler solution there.

>> Yep. And and that's fair. That's fair.

Maybe it's just me. I don't want all my eggs in one basket, but I I use port bun and they use Cloudflare for DNS anyway if you don't even move over move your domain over to Cloudflare. So, it

doesn't even >> at the end of the day, I don't think it makes a big difference. I don't know why. That's a hill I'm dying on. But,

why. That's a hill I'm dying on. But,

uh, >> well, I also use Pork Bun and and I've always appreciated and I I I guess I didn't realize that they by default used Cloudflare for their DNS because I

always wondered how how they managed to get my my DNS switched so quickly, but that explains it.

>> Yeah, that's probably >> I've had the same experience of also using pork button for that. I feel like you see uh Ethan I you see people loudly

proclaiming not to use the same registars to use for DNS but the reality is that it's it's not as important for people as as the people that are loudly proclaiming you shouldn't do it.

>> Sure.

>> Y and most of us are going to use Cloudflare for DNS anyway. Like the

number of people using it is probably the majority of these people if they're using a DNS service of some capacity. So

>> um and for for those who aren't aware, Google Domains, I think that was acquired by Squarespace. So that's why Squarespace popped up this year.

>> Um and I noted that in the survey, so people knew to select it. I just didn't include it because it was too long as a label. Um but if you look at last year's

label. Um but if you look at last year's results, it's just like Squarespace kind of pops up where Google Domains was in prior years.

>> Yeah.

So now I think arguably the more the most interesting section of the the survey at least for me software. What

are people actually um deploying and where are their preferences in terms of software in their home labs? First two

questions uh again somewhat divisive in the self-hosted community. Um are you self-hosting email? And for to from what

self-hosting email? And for to from what I can tell, it's always like overwhelmingly people are not self-hosting email, but there's like some guy who's been doing it out of the

back of his pickup truck for the past 30 years and has never had an issue with it. And he pops into every thread and

it. And he pops into every thread and has said, "Hey, you should do it. I've

had no problems with it. I've been doing it for 30 years." But

>> um it's complex. It's complex.

>> Yeah. And and I'm I'm looking at last year's numbers. the numbers didn't

year's numbers. the numbers didn't change from last year. Like like in the grand scheme of things, the like we're talking a minuscule amount of difference across the board on on both of these

email questions. Like almost exactly the

email questions. Like almost exactly the same numbers.

>> Yeah. And it and the I think the the sending component is um is a little hard for me to understand because I think in

my mind that's the most complex part of self-hosting email is how do you how do you not get blacklisted or how do you build the reputation to make sure all of your mail is delivered. So, every time I

include this email components question, I'm always surprised that it's the most popular, but I I know there's also some complexity with receiving to make sure you're not receiving all the spam that's coming through, but also getting

everything.

>> Um, so it yeah, it's it's interesting.

My question on on the sending part of things um are are people who are using like a like a local host like a mail hog server to get notifications about things

internally. Are they also including that

internally. Are they also including that as sending?

Because I know you can set up like a mail hog server that will give you notifications about things and I wonder if that's where those numbers are coming from.

>> It's a good question. Maybe it's

something worth diving into um next year if we can't get anyone in the chat who might have uh some experience with that.

It'd be interesting to hear.

>> Just say no to self-hosting email. Yeah.

Yeah. It's just it's not worth the headache.

>> It's like I'm just imagining like a 90s commercial or ad where just uh anyway data database deployment uh moving past

email. Um how are people deploying

email. Um how are people deploying databases? Uh overwhelming amount of

databases? Uh overwhelming amount of people are doing it via containers. Not

surprising. And then the people who are doing it containers tend to do it in separate databases for every service which makes sense. That's that's one of the benefits of containers. I will say I

do a shared container database deployment for my infrastructure.

>> So scary. um at least for services that are on and it's just like I again I love I love being so like into the weeds of managing my infrastructure like I don't

mind the upkeep and the backup associated with doing all that. I knock

on wood I've not run into any issues but I probably will now that I've mentioned it.

I think in past years DB you've you've mentioned like oh I can't believe you know people are so crazy that would do that and in the back of my mind I'm like oh I'm doing it >> like I that's why all of my all of my

different services each have their own is if in case something goes down I can fix that one Lexe based on yesterday's backup and and and so far that served me very

well but the idea of putting all of my eggs in one basket as you just put it a moment ago uh just scares the crap out of me like trying to figure out which database tables I need to restore from

yesterday's database back up to bring everything. No, not going to do that.

everything. No, not going to do that.

>> Yeah.

>> Yeah. But I I bet I'm running at least 20 at least 20 or 25 services that need a database. And I've never run into issues of database requirements

where something didn't support the actual version I was using in my shared.

So, in my mind, like I think it's a little less complex than people say, >> maybe. Um, but again, maybe I've just

>> maybe. Um, but again, maybe I've just gotten lucky.

>> I I think you've been pretty lucky. I've

had plenty of experiences with not I used to do shared as well and then I stopped doing it and I've just had plenty like image has changed their

>> their Postgress uh database a few times to support different extensions. Um, the

few like >> I'm not sure if I'm pronouncing this correctly, Dwarish the the GPS tracking, they've changed theirs a few times.

They've changed minimum versions and I cannot imagine having to export all that out and then start up a new one and

then reimpport everything again instead of just having it all separated per per stack.

>> That's crazy.

>> Yeah, I guess that's just for us masochists.

[Music] But anyway, moving into the next the next several questions are um specifically about what software are people using across a number of

categories. A lot of them are new. This

categories. A lot of them are new. This

first one media streaming isn't but actually super insightful um and looking forward to to chatting about the results and also hearing what people are using in the chat. First one media streaming

um Jellyfin number one again it has very much taken the lead over Plex versus prior years for sure. And I know Plex has been caught up in so much nonsense

this year. So I'm not sure if if it's an

this year. So I'm not sure if if it's an element of that a lot of people have migrated to Jellyfin or if again just the audience of the survey differed enough that that more people preferred

Jellyfin. Um but overwhelmingly people

Jellyfin. Um but overwhelmingly people prefer Jellyfin who responded to the survey.

I've recently um gotten to the point where I'd like to have a personal conversation with uh whoever decided to change the Plex app

for the Roku devices.

Um I feel like there should be a public lynching for that person. Um the the change that was made was so drastic that they're getting drug in all of the forums

because they moved everything from the left column to a series of sub menus across the top and makes everything much more difficult to access.

Super frustrating. My wife hates it. The

kid hates it. I hate it. Thousands of

people online hate it. And the developer is like, "Yeah, sucks to suck. Like

we're not we're not going to we're not going to go back to the way we were."

Um, and I just I can't imagine the thought process that went into let's make our UI worse. Um, but also I think one of the reasons that people uh tend

to go jellyfin um is is the same reason that people think that uh hosting or uh uh hosting something on a VPS isn't self-hosting.

Using Cloudflare tunnels isn't self-hosting because Plex is a corporation. So by using them you're

corporation. So by using them you're shilling. Um there's very much that

shilling. Um there's very much that mindset um when you look at Reddit posts and that sort of thing as far as media consumption is concerned that that if if

you're using Plex then then you're not self-hosting your own media server basically.

>> Yeah. It's closed source. You rely on them for authentication. So if they go down I think locally you're fine but your users can't access things like I I get it like that's absolutely

um a factor in that. And it's funny you bring up the Roku client. A couple

months ago, my in-laws mentioned something to me about their Roku Plex having issues. It changed in my mind.

having issues. It changed in my mind.

I'm like, "Oh, like it's probably not bad. They're just, you know," and I

bad. They're just, you know," and I remember when I saw it, I was like, "This is awful." I have no idea who who signed off on this. But it was even for me, it was not intuitive to use. So,

it's interesting to hear you bring that up.

>> No, it's it's been it's been my my wife has been asking, "Isn't there something else that we can use?" Um, and and like from my experience, like we've used we

used MBI in the past for a while, but the user interface on on the app on our on our Roku TV was was garbage. And I

don't know if it's gotten better, but um but I haven't looked at JellyFin. I

don't know what the what the interface looks like on cuz all of our TVs are Roku now because it was the easiest thing for Plex and now they've now they've screwed that up and I don't know

what the options are as far as a a jelly thin uh interface on Roku.

>> I I uh I haven't heard about this Roku fiasco. Um well, I don't have any Rokus

fiasco. Um well, I don't have any Rokus in my house specifically, but none of my users have also complained about it. So,

>> I'm not sure if everyone's avoided it or if it's just been a non-issue for them, but um I also have to agree with um Jay Collins and Scarlet's in the chat there

that um until Jellyfin can get easier for non technical people. Um I'm I'm on the Plex train myself. I've been on it since before JellyFin was decently

usable. But man, just, you know, for

usable. But man, just, you know, for like people's parents and stuff, it's just so much easier to have them sign up for an account, a website, and then they just send me their email and they

automatically are in. It's it's just >> they don't want to have to deal with setup. And uh you know it's it's as much

setup. And uh you know it's it's as much as Plex is for me uh it for the people that rely on it in my life I'm I'm not going to make them go through jellyfin

setup and the the trials and tribulations that go with that just because of a principle.

>> Right.

>> So >> y Matt I'm I'm curious for the some of these other platforms listed on the question here especially since you've developed Multiscrabbler and I assume you stream a ton of well I know you

stream a ton of music. What does your music setup look like? Or or what platforms are you using to bring that together? And I'm half expecting you to

together? And I'm half expecting you to just say like Spotify and something, but >> yeah, it's just Spotify. I mean, I have I have a a a decently sized local

collection from prior to Spotify being a real big thing that uh I still use with um it's it's in Plex uh and it's also on Navajome and Jellyfin, but basically

just for Navajome and Jellyfin are for testing uh multiscribler. Um but I use Plex Amp if I want to listen to my local stuff because it's a great interface.

Again, it's just I know it's paid and it's not exactly self-hosted, but man, it's just so easy, >> right?

>> And then we have Plex for Plex for video, but the majority of my streaming is or audio is Spotify cuz it's just even at the price point now, it's still

I listen to so much music. It's not

worth uh uh I don't listen to albums at a time. So like the the R software

a time. So like the the R software around audio acquisition is just doesn't support the way that I listen to music.

So I just I just pay for Spotify and they have a great API and a great interface and it just works. So yeah.

>> Mhm.

>> Uh shifting to the next question. File

storage and syncing. Uh there there's a lot listed here and there are a lot of there's a lot of software that that does different things. I think I was trying

different things. I think I was trying to capture and this is multi select so people could select multiple things. So

if you're using Nexcloud to you know share things with friends but using file browser to manage yourself from the browser like you could select both.

They're just really trying to figure out what are people using to either manage their files externally or even internally. A lot of these are internal

internally. A lot of these are internal tools as well. Um and so nothing super surprising here. Similar to last year um

surprising here. Similar to last year um nextcloud and sync thing top two responses.

um >> jumped from five to three >> um and has five times the results that it did last year again, which surprises me after >> after the nonsense they pulled this year

to see it go up that much really kind of surprised me.

>> It just really reinforces there's no such thing as bad press. Although with

Plex, Plex did go down, so maybe there's bad press when you're Plex, but um file browser continues to be popular. I think

a lot of the same platforms we saw before. Um, Nexcloud is interesting

before. Um, Nexcloud is interesting because everyone loves to hate on Nextcloud, but everyone uses Nextcloud.

>> Um, so it's surprising but not surprising, I guess, to see it up there.

Uh, next question is authentication.

This was new this year. uh if people are using a centralized platform to um for authentication or identity management whatever you want to call it what are you using there are a lot of popular

ones and this this is a good again a good mixture of what I called legacy platforms and new ones so Athelia authentic keycloak those are kind of like the de facto platforms I think of

when people talk about these things but pocket ID pengalin tiny off I think all of those are 2025 launches is Pocket ID maybe was last year. I can't remember,

but I know Tiny Off and and Penglin were both uh 2025. So, it's really cool to see how how much those have kind of infiltrated the rankings here.

>> Yeah. And I really like um like Matt was saying earlier, Penguin has just come out of nowhere and and they started with a just a real simple reverse proxy and then immediately it blew up into into

basically a Cloudflare tunnel replacement service and they built authentication into it like and I know they don't have a huge team

and I just I can't imagine what kind of of of coding sessions they've got to be having to to to do all of this this quickly.

>> Yeah, their velocity is in it's really crazy. I don't know how they're and and

crazy. I don't know how they're and and their documentation's pretty good, too.

It's I >> they Yeah, it's it's really really impressive.

>> Um the ne next question is note takingaking and documentation. I almost

need to zoom out a little bit for this one.

>> U this is this is another one that there are a lot of platforms listed here that do a lot of different things there. To

me, there's there's not like a clear distinction between notetaking and like documentationwiki platforms. It feels like they're often used interchangeably. And so that's why

used interchangeably. And so that's why I've grouped them together. Um although

I'm sure some people will disagree with that. And oddly enough, I included

that. And oddly enough, I included Obsidian because uh I actually don't know if I included Obsidian last year.

>> You did. It was number one there, too.

>> Yeah. So, it's like if I didn't, people responded enough with it that I felt that I needed to to to include it as an actual selectable response because >> it the overwhelming amount of people are

using Obsidian, which >> there was some discussion in some of the the places I posted the survey results.

Is Obsidian self-hosted? Like, is a is a desktop app that you can sync the backend to different places self-hosted?

Um, traditionally I I don't think it necessarily is, but um I've never used Obsidian and I'm wondering what I'm missing out on.

>> I've had that same question.

>> That's crazy.

>> Uh, Matt, I assume you don't use Obsidian then.

>> No, I'm I've uh I've been using any type and it's been all right. Um, but I I honestly am not very organized for note-taking in general. So, this this

category of software hasn't been high on my list of things to experiment with.

You were did I mention you use some things on this list, right? Do you use a fine or is it outline?

>> I use outline on this list.

>> Yeah.

>> Um, which it's very similar to that.

It's similar to docu. Like a lot of these kind of are have kind of that similar look um to them. But again, this is also like a nice mixture of newer and

kind of older software and platforms here. Silver Bullet is one I've been

here. Silver Bullet is one I've been playing with a lot lately. It's very

extensible. Um, very almost programmatic, if that's the right word for it. Um, I don't I don't actually

for it. Um, I don't I don't actually keep a whole lot of documentation or notes in it, but it's actually really fun to play with. If if if no one's checked it out, absolutely look into it.

Um, but interesting. Next cloud notes is also interesting. I guess given the

also interesting. I guess given the amount of people who are deploying nextcloud, it makes sense that they're also using it for nodes.

>> Yeah, >> it's just great to see so many options.

Honestly, it's like so many actively developed choices here.

>> Yeah, absolutely.

>> Uh next next question. Uh Git, what are people using for Git? So Gee is the first option. I'm not super surprised. I

first option. I'm not super surprised. I

I think GI has always been a little more popular for most of the home lab crowd because it's it's a lot more lightweight than GitLab is. Um for JO, which is the

fork of Gi is growing in popularity. I

wouldn't be surprised to see that match gate or surpass it in the next year or two. Um there's been some controversy uh

two. Um there's been some controversy uh across some things Gia has been doing. I

I can't get I can't speak to it a whole lot myself. Um but but interesting

lot myself. Um but but interesting nonetheless. Um especially given how

nonetheless. Um especially given how prevalent GitHub is in the self-hosting space. It's comforting to see a lot of

space. It's comforting to see a lot of people are still using self-hosted git forges.

>> One of the uh both git t and forjo is how did you pronounce it? Is that how it's pronounced for?

>> Yeah, for jo. I I've looked it up so many times.

>> Yeah. Okay. I've been pronouncing it wrong all all this time. Anyway, both of them now support uh some subset of GitHub actions runners, which is really

exciting. Um, and they both they both

exciting. Um, and they both they both have their like their disadvantages and vantages on what they actually support, but I I I'm really glad to see both them at least supporting it. And it's

exciting to have that kind of like Git Ops functionality in the in the self-hosted scene now.

>> Yeah, man. I actually

>> GitLab as well, but I'm not sure. I only

know about git team and 4.j having that.

>> Yeah, I actually think Matt, you helped me out when I was deploying forj a month or two ago. I remember you and I were chatting about the runners and stuff and you sent me a tool I hadn't come across.

So, >> um, home automation, how many people are deploying software for smart home or home automation?

um a lot of a lot of deploys. Not as

many as I would suspect given how prevalent Home Assistant is in this space, but it's still a lot. Um I I I guess it's not surprising that there are a fair amount of people who just don't

care to dabble in the space. Um whether

it's because their spouse is just tired of their lights not turning on or whatever whatever other stigmas are associated with it. Um not a whole lot more to to ask about this question. Last

year I asked about protocols. So like

are you do you have ZigGB or Z-Wave and those results were interesting, but Home Assistant is kind of just like the the the name of the game in this space.

There are a couple other voice assistants that that have been popping up um that I've seen uh kind of take on what Home Assistant has done in voice, but nothing I I don't if there's a

platform out there that rivals Home Assistant, I'm not aware of it. No,

>> I've seen a few different things, but nothing nothing to that level. Sorry for

>> jumping in there late.

>> No worries.

>> No worries. Game game servers are next.

Uh more people aren't hosting a game server than are, which isn't surprising.

And then when you look at the games um that are being hosted, Minecraft accounts for 37% of responses, which is wild. But again, given the the

wild. But again, given the the popularity of Minecraft has been interesting if you go back and look at last year's responses, the other games mentioned like it has fluctuated a little bit. Um, you see some new games

little bit. Um, you see some new games pop up and some others drop off. So, you

can kind of see the trends, but there are a lot of games um that I had never heard of actually when I was putting this together. I had to look up to make

this together. I had to look up to make sure I was spelling everything correctly. Um, but it's cool to see kind

correctly. Um, but it's cool to see kind of how I'll say fractured, but that has a negative connotation, but just how many different games people can can host on their own.

>> Yeah.

>> And then everyone's favorite question on the survey, the hardest question to aggregate the results for because every response, regardless of how simple of a

word it is, will be misspelled 10 different times that I have to go through and scrub the data. Um but very similar to prior years um home assistant

jellyfin at the top. I actually think jellyfin was the top answer last last year. Um and so jellyfin and home

year. Um and so jellyfin and home assistant swapped which I which is kind of unusual given how much more jelly how much how much more prevalent jellyfin

seems to be versus Plex this year. Um

but it sounds like home assistant again they've had a bunch of new hardware launches since. Um, not unusual to see

launches since. Um, not unusual to see them kind of gaining traction in that space. Image, I think some I think one

space. Image, I think some I think one of us mentioned image earlier. I'm

surprised we haven't spoken about it more. Maybe there's like a photos

more. Maybe there's like a photos question. I should have asked photos and

question. I should have asked photos and videos. Um, but especially with their

videos. Um, but especially with their big version two launch. Um, expect them to start seeing seeing them grow in popularity.

>> Um, yeah, a lot of you I'm watching the chat to see if people are bringing up anything specific. A lot of these

anything specific. A lot of these platforms I'm familiar with. I didn't

see anything pop up that I hadn't heard of. Um, Pi Hole and Adguard are kind of

of. Um, Pi Hole and Adguard are kind of low in terms of ad blocking. I assume

more people would be adamant about their ad blocking capabilities.

Uh, but interesting nonetheless.

>> I think I think those those are the the kind of apps that we set up and we don't think about them ever again. It just it just works. It blends into the

just works. It blends into the background. Um,

background. Um, I went to a friend's house a while back and they had a Roku TV up and it was just on the screen saver and I didn't

think much about it until I started seeing ads on the Roku TV screen saver >> and I was like what? Like you're it never occurred to me that they would put

ads on the screen saver because I've never seen them on my TVs. Um, so it's just it was it was something I didn't think about because it's just part of my infrastructure that runs quietly and

diligently in the background. And I

think maybe that's why why those aren't quite as high as we would expect them to be for favorite apps because people just forget that they're there.

>> Yeah, it's one of those things you don't you never notice it unless it's not working, right? Type thing.

working, right? Type thing.

>> Uh oh, there's Penglin. Penglin seems to be like the penglin jellyfins see seem to be the the big theme in this year's survey. Oh, Kimot's popping up there as

survey. Oh, Kimot's popping up there as well. Um, and I know we're we're

well. Um, and I know we're we're approaching an hour and a half, so I know we'll probably start to wrap this up here soon. The last question the last section of questions are demographics.

Um, these absolutely have not really changed at all this year. Um, none none of the information is super surprising.

I will say I felt kind of bad because I I think I included up to like the 80 90 year old age range last year and no one responded beyond 69 and so I didn't

include that this year and then someone reached out and was like hey I'm over 70 and I self host where's that option and I felt really bad so I I think I'm just going to include to like 120 next year just to make sure I have everyone

captured. Uh, but it was really cool to

captured. Uh, but it was really cool to hear about, you know, a 70s something year old self hosting.

>> I hope I'm still as into this when I'm uh 70 80 years old as I am now, >> right?

>> Uh, but overwhelming amount of people between 12 20 and 50 >> years old. By the way, as I think I think I'm going to hit like the three of us as we go down this. I feel like I'm

describing the three of us in a way. Um,

which makes me think maybe we need next year a little more diversity on the uh the commentators for the survey. Uh,

>> gender 94% of people are are male respondents. Um, a little more

respondents. Um, a little more representation this year than prior years. So, it's really cool to see that.

years. So, it's really cool to see that.

I I don't know what we can do to encourage more diversity and participation from from other genders in the space, but uh if anyone has any suggestions or if I can do anything via

my content, always happy to to look into it. Uh career field, uh self-hosters are

it. Uh career field, uh self-hosters are more likely to work in technology related fields. DB, I guess you're in a

related fields. DB, I guess you're in a technology field. I was I was wondering

technology field. I was I was wondering that earlier like how you would have responded to this question.

It it's a weird question for me to answer because this is my job like like my my job is

talking about tech online. Um so so I guess I I think I responded that technically uh my career field is technology. Um but like going back prior

technology. Um but like going back prior to doing YouTube, you know, I was I was a front-end web developer. Um, you know, I've done I've done those kinds of things in the past. I've done hardware

testing in the past. So, a lot of my my history comes from actually being out in the work the workforce in a tech uh technology field. Um, but I've just kind

technology field. Um, but I've just kind of I've just kind of put myself in one because >> because that's where I ended up uh kind

of by accident. But um yeah, >> it's I guess technology just not the traditional like technology role when you think of it. But uh

>> right >> you're I feel like tech YouTubers with 100,000 subscribers are a very very small portion of the population >> probably. Yeah, there's there's a few of

>> probably. Yeah, there's there's a few of us. There's there's a few of us.

us. There's there's a few of us.

>> Uh mobile operating systems, Android's the most popular. I I when I first started my the uh self-host, I was surprised at how many iPhone users there are. I just always assumed everyone in

are. I just always assumed everyone in technology was more Android uh leaning.

But I iOS is uh not unpopular. The 32%

is uh not insignificant for sure.

>> Yeah, looking looking at last year's numbers, those they actually didn't change all that much compared to last year for between iOS and Android, but more Android users this year than than

last year by by a bigger margin. But iOS

users didn't really change all that much.

>> Yeah. Yeah. and graphine OS that that popped up this year as well. That was on last year's chart. Um, also received a fair amount of responses. I think that's more privacy focused like the uh

>> deled Android.

>> Um, next question I include just so I can shame people throughout the year.

Uh, it's a it's a good talking point to bring up when I when I'm running out of things to say talk about in the newsletter. Um, more people don't donate

newsletter. Um, more people don't donate to self-hosted projects than do donate.

Um, we definitely encourage people to to think about it. I don't know, you know, everyone's financial situation, so certainly don't want to pressure, but um, a lot of the a lot of developers are putting a lot of their own time and

effort into the things that I think we take for granted. I spend so much time deploying self-hosted software. Um, like

to make sure people people are somewhat compensated for that.

>> Yeah. Um it's it's an interesting statistic especially given the correlation between self-hosting self-hosting and open-source software.

>> I would like to see uh questions next year on people's contributions non-financially to like this this industry like uh how many people are

actual OSS developers or how many are if you're not a full developer making your own bespoke app have you have you published any Docker containers that you

use or let other people use like what kind of contributions are people making back that aren't financial? I think I think we all kind of know what the answer to that is the percentages wise, but it would still be interesting to see.

>> Yeah, I agree. I think it would help I think it would help kind of define who who the respondents are to a certain degree if they are people who are contributing to to projects or building

their own projects. I think that would be kind of an interesting um bit of information to have.

>> Yeah, maybe even you know you maybe haven't made your own container but have you contributed any like you know YAML configurations for automations in Home Assistant? Have you have you helped

Assistant? Have you have you helped other people out in a >> in a non-trivial way in a like a material way with with setting up their their systems? That would be interesting

their systems? That would be interesting to know.

>> Yeah.

>> Yeah. And John just mentioned in the chat about donating isn't just money but time to the project. There's there there is so much you can do to contribute. I

think documentation like I think so many devs would kill to have people help with their documentation uh translation things like that. Like there are things we can do to contribute that aren't just you know financially contributing.

>> Yeah.

>> So and then lastly we have education here. I should probably swap these two

here. I should probably swap these two because education is a little more uh anyway uh self hosters. My takeaway was they tend to be more educated than general population, but there's still a

nice uh variety and representation uh across different um educational experiences. I will say I got a lot of

experiences. I will say I got a lot of feedback that this was not very internationalized. Um there were people

internationalized. Um there were people in places like Europe who were like I don't I don't really know what my level of education translates to. Uh this is very US-oriented. So I have to give that

very US-oriented. So I have to give that some thought into how I can do that for next year. So apologies to anyone. I'd

next year. So apologies to anyone. I'd

never actually considered the idea that they may have a different educational ranking or whatever system. Um, yeah,

it's weird.

>> Yeah. So, that was the last question.

Um, again, I thought this year's survey was super interesting. We got just over 4,000 responses, which was uh 4 or 500 more than last year, >> which was super exciting. These results

are obviously public on my website. I

also publish um them on GitHub as as well just in a JSON file if people want to do their own things with them. If you

want to display them elsewhere, you don't need permission from me. You don't

even need to site me, although it's always nice if you do. Um you can use them for whatever purposes you'd like.

Feel free to.

>> Um DB, once again, thanks for opening up your uh YouTube home to me to to chat about the the live stream.

>> I enjoy doing this every year. like I I genuinely uh enjoy hanging out and chatting about all this stuff.

>> Yeah. Cool. And Matt, thanks thanks for joining as well. Super interesting to get some other insights um from people who aren't just DB and I and of course you also develop a self-hosted platform.

So >> um also super interesting to hear from you.

>> Thanks for having me. This was a this was a great time. I'm glad we got to I'm glad I was a guest. Appreciate it.

>> Absolutely.

>> And DB, I think you dropped some links in the description. And I think if you want to get a hold of any of us, right?

>> Yeah. There there are links to to all kinds of stuff down there. You

should definitely check out um I'm not smart enough to get there in the back end of uh YouTube here. But uh yeah, no, definitely check out the links in the video description uh if you want to get more information about anything going on

there. Um

there. Um look forward to to doing this again next year. Like I said, I always enjoy

year. Like I said, I always enjoy hanging out and going through the data and and having a good conversation about about what we're seeing across the community.

>> Yep. Absolutely. And thanks to everyone for joining us for thanks for everyone who participated in the survey, but thank you also to those who join us and have been chatting with us. Um, a lot more interesting to talk about these

results while also seeing what everyone else's thinks about them as well versus just the three of us.

>> Yeah. So,

>> have good times.

>> Cool. Thanks everyone.

>> All right. Well, thank you guys for hanging out. Um, if you enjoyed the

hanging out. Um, if you enjoyed the stream, be sure to give it a like. Um,

like I mentioned, we do this, we do this every year. If we've done this for the

every year. If we've done this for the last three years anyway, uh, I imagine as long as nothing goes catastrophically wrong, we'll be here again doing this next year as well. U, so again, thank you to everyone who stopped by, hung

out, jumped in the chat, or whether you whether you chatted or lurked, that's awesome. Thank you so much for hanging

awesome. Thank you so much for hanging out, and uh, we'll talk to you again here real soon.