LINUX Unplugged - 386: Perilously Precocious Predictions

Episode Date: December 30, 2020

Friends join us for a special edition of the show to review last year's predictions, and forecast the future. Special Guests: Alan Pope, Alex Kretzschmar, Brent Gervais, Drew DeVore, Joe Ressington, and Neal Gompa.

I'm looking over my predictions before we get started, and I think my Rocky Linux prediction is just too weak. I predicted they'd ship, and I'm just disqualifying that one. Oh, I've got a good one about that. Oh, you do? Yeah. Do you want it now or later? No, no, actually, you know what? Save it for the show. Hello, friends, and welcome into your weekly Linux talk show, episode 386, one of my favorite episode numbers. This episode is brought to you by a cloud guru, the leader in hands-on learning.

The only way to learn a new skill is by doing. That's why ACG provides hands-on labs, cloud Linux servers, and much more. Get your hands cloudy at cloudguru.com. My name is Chris. My name is Wes. Hello, Wes. You brought a big crowd with you today. So right off the top here, let's welcome back to the show, Joe, Alex, and Drew. Hello, gentlemen. Hello. Good to have all of you back. This has become an annual tradition now where we make our predictions and review how we've done gentlemen. Hello. Good to have all of you back. This has become an annual tradition now where we make our predictions and review how we've done. Sometimes we break it up into multiple episodes. This year

we're going to try to just put it all into one show. So just get it done. This is the last production of 2020. Thank goodness. And to help us celebrate and throw in a few predictions of their own, it is our virtual lug. Time-appropriate greetings, Mumble Room. Good morning, Mumble Room.

Okay. That's a big crowd in there and a lot of energy in there. You know, it's that pre-show. We get them all warmed up with the pre-show. Carl, I see you just snuck in under the radar there. Hello, Carl. Good to have you back. Howdy, howdy.

And, Pelby, good to have you back too, sir. Hello, hello. Greetings. Now, I want to hear all of your predictions here in a bit. So get thinking. And I think you have the luxury of not being held accountable for your predictions. We do write them down, right, Ben? You're going to write these ones down, right?

Yeah. We're counting on you. There will be a document with your prediction written down, but we won't play it back and shame you for it like we're going to do ours. So let's get right to it because I want to just do this in a couple of pieces. We'll start with the reviews and then we'll do the predictions. And just like last year, Joe is here to play judge, jury, and coach, I guess, ref. Joe, are you ready to get started with the reviews? Should we start with Alex?

Yeah, let's do it. Oh, I don't know if I'm ready for that. I know, but let's just get, it's like a band-aid. We're just going to rip it off, okay? Okay. All right. So you made a prediction about hard drive sizes. There will be a three and a half inch hard drive available to consumers of at least 20 terabytes in capacity.

Now, I don't know if that's actually happened, Alex. Ding, ding. There's 24 terabytes on us capacity. Now, I don't know if that's actually happened, Alex. Ding, ding. There's 24 terabytes on us now. Oh, no kidding! Really? Do you have a link for that? Can we verify that for the judge really quick? We may have our winner.

He has strict standards, that judge. That's great, Alex. I think, oh, we're starting off with a winner. That's fantastic. I love it. Great, Alex. I think, oh, we're starting off with a winner. That's fantastic. I love it. At 100 terabytes, it's the largest solid-state drive right now, the Nimbus Data Exadrive. 100 terabytes.

Wow. Can you actually buy that, though? In September, they said it would be here by Christmas. Any idea on the price, or it like not actually listed yet? 100 terabytes. I mean, imagine essentially your entire storage capacity, Alex, in one disk. Yeah, that would really be something. So yeah, I guess you definitely got that one right. We're giving that one to you. Imagine the RAID rebuild times on that mother. Jeez. And the trickery they're pulling off, I'm sure just to get that possible. Okay, Alex, two more for you. The next one has a good shot, too.

You made a prediction about Jellyfin. I predict that MB or Jellyfin or Olaris will become a viable alternative to Plex by shipping apps on one of the major platforms such as Roku, Android TV, or Apple TV. Okay, a Jellyfin app on Roku or Android TV. I think that exists, but I'm not positive. Does anybody... I mean, you can find Jellyfin over in the Roku store, it looks like. There's Jellyfin for Android TV on Google Play.

Alex! Two for two. Hell yeah. I mean, see, you thought it was going to be rough, man, but I'm saying... Can I be the referee on that terrible sound and say, please don play that chris yeah i don't like that i don't like that i hate it too i just found it all right alex your last prediction let's see what you said about wire guard so i predict that ubuntu lts 2004 will ship with wire guard not as a dkms module and also as a second like half prediction that pf sense will ship with wire guard okay i think the ubuntu part's right although we may have known at that point do you know ref do you know when we when we found out

uh that that was going to happen or poppy maybe you might know when that decision was made? I'm pretty sure we knew at that point. No, we definitely didn't. I wouldn't have been so callous and predicted something I already knew. So what about the PFSense part? Also, can we check on that? No, so PFSense is still just a plugin.

In fact, I don't even know if it, because I've switched to OpenSense now, because of WireGuard support. It was right on the line last December, right? It had been pulled into the NetDev tree, but not actually into Linus' tree around this time. But the Ubuntu one was definitely not certain until 2004 shit. Yeah, I think I'm going to give it to you.

I'm going to give it to you, I think. I mean, Ref, you can talk me out of it. I don't have to warm the point, because you get the call. I don't mean to overstep. But I actually, I'm thinking to give it to you. I'm going to give it to you, I think. I mean, Ref, you can talk me out of it. I don't have to warm the point because you get the call. I don't mean to overstep. But I actually, I'm thinking about it, and I don't know that we knew for certain it wouldn't be like DKMS or it wouldn't be some sort of weird. If I recall correctly, it was very much targeted at that point,

but it hadn't been 100% decided. So it was a bit of an easy prediction, but I think we have to give it to him. All right. Good job, Alex. Jeez, that's all three for you, Alex. You got them. Three, yeah.

Jeez, now the pressure's on. You really better not disappoint this year. Do I get a prize, by the way, for getting all three right? Yeah, but you've got to come to Washington to get it, unfortunately. Well, you've got a 3D print it for yourself. Yeah. Thanks, Joe. All right.

I think I'll hold mine. I think I'll hold mine. I think I'll hold mine. I'll do mine next. But let's do Wes, because I've got Wes. You queue it up next here. I guess we're not doing alphabetical anymore. Oh, boy, Wes, I really wish this one came true. You made a prediction about BcashFS.

I predict that in 2020, BcashFS makes it into the Linux kernel mainline. Oh, man. No luck for me there or for any of the rest of us. Ironically, I just made that my prediction on Linux Action News. We have another set of maybe more professional picks on Linux Action News or predictions, I should say. And that was one of mine. Hint, hint. Yeah, I feel a little bolstered by that.

You know, like it's not unreasonable, but at the same time, I'm sort of happy to see things continuing to develop nicely. But, you know, with still a few rough edges and not getting rushed, I'd like it to be a file system I trust once it's ready and in the kernel. I want to give some latitude to the ref on this next prediction because I think it's kind of like half and half. It was about me and what distro and desktop environment I would be using at this point. half. It was about me and what distro and desktop environment I would be using at this point. I predict that by the next predictions episode, you, my friend, will be

on Manjaro. Now, I think you also, in the less shortened version, said I'd probably be on Plasma, too, if I recall. I did predict you'd be on Plasma. Yeah, but that wasn't but what was in the locked version is Manjaro, and while I have Manjaro on a device, I'm actually on a mix of Fedora and Arch.

Or I guess you could say Endeavor OS. Does that count? No, that doesn't count as Manjaro. What's on your upstairs desktop in your office and what's on your laptop? I think it's got to be one of those two. So the laptop's Fedora and then the desktop upstairs is Endeavor OS or Arch. That was close though,

Wes. Yeah, you know, I think I was in the right vein of things in that I think I didn't think Endeavor OS would be where it is today and really be an option. So I think Mandara was like, that's the Arch that Chris would actually use. And it turns out there's a slightly, you know, closer to Arch version of that. And okay, that makes sense. And I think the part that didn't make it in the locked in prediction,

because just for the rules, for those that haven't listened to the previous one, we have to lock in the prediction. So when we get to our predictions, we're going to work them out as a group. And then we decide on a locked in version, we make it nice and concise. And then we lock that in. And in our conversations, Wes, what I recall was that you were doing the plasma math and you were saying, I think you're going to be on plasma. And if you're going to be on plasma, you're probably going to be on an arch. And at that point in time, it was just, it was really likely going to be Manjaro. The curveball is, is I didn't appreciate the fact that the Fedora spin of plasma has

very frequent KDE neon style plasma updates. It's, you know, you're really, you're getting fresh plasma in there. And that was the part that was kind of unknown to us back then that I only discovered, you know, a few weeks ago when Neil told me. So there's that. And I think that threw you a curveball

in the whole thing. Hmm, so you're saying maybe there's a new prediction for next year buried in here. You probably ought to give us, you could give a stab at it. You never know. I'm not going to reveal, you could give a stab at it. You never know. I'm not going to reveal, though, which one I think it is. All right, you got one more, Wes.

And it was about Google Stadia. It doesn't make me happy to say, but I predict that by the end of 2020, Google will have announced the sunsetting of Stadia. Gladly not. Yeah, yeah. So I didn't realize they were in year two of this thing. But I guess we're entering year two of this thing. But I guess we're entering year two of Stadia. I wonder too, it seems

like it's, you know, it definitely felt kind of stagnant, not a lot of news, but it being one of the better places to play Cyberpunk, I think has bolstered it a little bit, at least in the past, you know, couple weeks. True. That and the fact they gave it away for free whenever you bought a Chromecast and stuff like that for a while. Yes.

And during COVID, it was free for like two months. Yeah, yeah. I think the whole COVID and the lockdown has boosted a lot. Yeah. And honestly, there was just no way I was going to upgrade my PC to play Cyberpunk. I just won't get the return on that kind of investment right now. And I just was not going to consider it. But when I realized it was on Stadia and I had like some sort of founder's credit or something. So I basically

got it for a super cheap price. And I was like, okay, well, that'll do. And I still wish to this day that if I buy a game on Stadia, it came with some sort of Steam redemption code or some other platform. Because if I'm paying full price, it really makes me tap the brakes before I use Stadia because just like Wes, I did in 2020. I think you could predict this year they spin it down. It's still not a guaranteed lock for me because it's Google and I don't feel comfortable spending full retail price for a game that's in Stadia. But if I can get discounts here and there from Google, I've taken action on them. And as a result, I've got Red Dead Redemption in there. I've got a couple of freebie games that they gave out. And I've got Cyberpunk 2077.

And I know people aren't going to believe it, but I have to tell you, I sit here in the studio at this very computer, which is not a really nice computer anymore. And I play Cyberpunk on the live stream. And I live stream playing Cyberpunk with graphics that I think look great. And I have zero issues with latency, with how it plays at all. And I'm not on a fantastic connection. I'm on a fast for Comcast connection. And it's real. And it means I can sit here and do it with free open source drivers. You don't have to worry about it, right? It's just another service that you have access to. You can pop on, play some games, and then go back to editing things in Nano,

if that's what you want to do. Okay. So I just came across Drew's prediction, and it threw me for a loop because it was a Drew prediction via Brent because Drew couldn't make it. That's not fair. Everyone likes Brent. I feel like I should get all my predictions read in Brent's voice just to have a better shot at winning.

Yeah. Hold on here. That's a great point. I didn't know having Brent read stuff for us was even an option. I would like to take advantage of that, too. Me, too. Sign me up. Brent is a service. All right. Brace yourself, Drew. You made a prediction about home assistants like the Alexa and Google Assistant, and it goes a little bit like this. I believe in 2020,

we will see one of the voice assistant platforms, one of the Pepsi or Cola, get a malicious exploit that is active in the wild. It's a little vague, Ref. I don't know, but what are you thinking on this one? I don't think that I've seen one. I haven't read about one. There was one in August of 2020, but it wasn't named. It wasn't huge news. So I don't know. I don't know if I would count it.

Okay, so there's actually been two different ones because I follow InfoSec fairly closely. infosec fairly closely. One allowed people to install applications bypassing your install account stuff. The other one actually allowed someone to put a malicious app on the Amazon Alexa and turn on the microphone remotely. It actually got a CVE, if I remember correctly. If you got a CVE, then you get the point.

Right, right. Yeah, boy, it'd be great to find links for that, but that actually correctly. If you got a CVE, then you get the point. Right, right. Yeah, boy, it'd be great to find links for that, but that actually does sound like he might have got it, ref. Yeah, I would say so. Woo-hoo! Nice job, Drew. I think that was one of the more kind of, I'm not going to say risky picks, but more out there ones,

just because it was even kind of debatable right now if it happened. But if it's got a CVE, I say, there it is right there. Boom. Popey has it. Yeah. Ha-ha. Good job, Drew. You only got the one via Brent, so unlike the traditional three, you just had that one,

but you can come up with up to three today if you'd like, but start ruminating now because it's going to be time shortly, but first, let's get to my three and see how I did, and I remember feeling like some of these were a long shot. So we start with my prediction about Home Assistant, which I actually didn't know made it into the show, but I guess I locked it in. In 2020, we will see the introduction of a Home Assistant box, either by the project directly or a third party bundler that creates a Home Assistant appliance. And this just recently happened. The Home Assistant Blue.

I can't believe the timing on that, Chris. Yeah, in fact, not only did they announce it, Alex, but a few people have started receiving their units in 2020. The Home Assistant Blue. Yeah, good job. It's based around the Odroid N2+, I think. Well, I don't remember making that prediction. So I was totally surprised when they did it,

which is funny because usually my predictions, like I hold on to them all year waiting for them to happen. But this year I just threw it out in the universe and let it go. And look what happened. All right. So, Ref, if you agree, that's a winner. Yep. Okay.

Uh-oh, this doesn't look good. My next prediction was about Plasma and Chrome OS. I predict in 2020 that by some means it will be possible to run Plasma Shell on top of Chrome OS before the end of the year. I mean, it seems like it should have happened with all the Linux stuff they were doing and running apps. But as far as I know, you can't actually get all Plasma Shell up on Chrome OS still. Damn shame. That should have happened. I think that's a loser, Ref. Yeah, I think so. actually get all Plasma Shell up on Chrome OS still. Damn shame. That should have happened.

I think that's a loser, Ref. Yeah, I think so because I would have noticed that if it had happened this year, but unless it's on some obscure forum

somewhere, no, I don't think you get the point. Maybe it's possible, maybe via a container, but I don't think so.

I know some people run KDE Neon in Docker, but I don't think so. I know some people run KDE Neon and Docker, but I don't think so. I'm not going to try to fight that. All right, this one also looks like it's another loser for me. I made a prediction about Ubuntu deploying Wayland. I predict in 2020 that Canonical will switch the Ubuntu release over to Wayland in their 2010 release. That also, it definitely did not happen. Although I have installed Wayland on Ubuntu 2010,

and that works just fine. Well, look at you. Yeah, unfortunately. Popey, what do you think the chances are that there'll be another shot at it before the next LTS, like the release before? You can't ask something like that right before the prediction segment. Oh, okay. All right. Fair enough. Fair enough. Fair enough.

Let me just quickly write down a prediction. Yeah. I think I clearly got that one wrong, Ref. I don't think there's any arguing that. Yeah. All right. Then one last prediction, and that was from you, and I only have one for you here, Joe. So here we go. All right. Well, I'll give you an absolutely crazy one. Chrome OS will add ZFS support.

I'm going to say a big no on that one there, Joe. Yeah, no ZFS support on Chrome OS. They did get ButterFS support, but no ZFS support. Yeah, that was just me being as ridiculous as possible at the end of the episode. Yeah, fair enough. Well, you don't get that free or meaningless internet point anyways. All right, so there we go. You know, I got to say, Alex, I think you clearly did the best.

I mean, I guess Drew got all of his right too, but that's because he only did one. We both got 100%, Drew, good job. Drew got all of his right too, but that's because he only did one. We both got 100%, Drew. Good job. Now, to be fair, I did submit like four or five, but only one of them got read. Oh. Hmm.

Hmm. Sounds like there is probably a case for you to file with the Brent as a service company. Yeah, I'll submit a ticket on that. Yeah. What we need to do is pay for the pro slack, and then we can scroll back and see what they were right yeah that's probably it that's that's how we'll do it i will unlock it um good job alex but yeah i think my my only solace in all of this is i beat west by one all right come on lucky right? I mean, it was a rough year for a lot of reasons, not just the predictions, Wes, not just the predictions. But all right.

So speaking of those, we will do that in just a moment. I want to thank everybody who's joined us live. You know, the Unplugged program will be live for you to come hang out all in 2021. It's a whole new fresh year. Why not do a whole fresh new stream over at jblive.tv? Linode.com slash unplugged. Go there to get a $100 60-day credit towards a new account and go there to support the show.

Linode is our cloud server provider. And because the price is so great, you can try them out and you can get a machine starting at $5 a month. Or with that $100 credit, you can try something pretty powerful. We spin up rigs as just test machines from time to time. Alex has a brand new fancy website that he'll be talking about soon on the self-hosted podcast. I don't want to spoil it, Alex, but he's hosting the entire thing on Linode. Listener Jeff wrote in to tell us about how he did a test migration using NextCloud. And he planned it all out, and he spun up some Tesla

nodes. He did the whole thing, read through all their documentation, and then he did it in production. And he actually discovered that their cheapest rig was more powerful than his local quad-core machine. And it's a great story of learning and deploying and just using the resources available at Linode to do all of that. And what I love about them is their passion for Linux and open source. That's what got them in this business in the first place. Probably a lot like yourself, they saw where technology was going to be really useful. They saw where something was going and realized, man, that's going to change the game.

That's what got them fired up about starting up Linode. And that's why they started three years before AWS, because they were doing this from an independent passion standpoint, not from some bureaucratic, how can we leverage infrastructure and make money? And what they ended up doing over time was building one of the greatest and largest independent and independently owned cloud companies. Now with 11 data centers all over the world, super fast SSDs, 40 gigabit connections to the hypervisors,

machines with multiple GPUs, dedicated CPU cores, and of course then things like object storage and block storage and load balancing and all of these really fancy backend features that you can take advantage through infrastructure management tools or just by getting down on the command line yourself. They're dedicated to offering the best virtualized cloud computing. If it runs on Linux, it'll run on Linode.

So just get started at linode.com slash unplugged. Go there, get a $100 60-day credit towards a new account, and go there to support the show. linode.com slash unplugged. towards a new account and go there to support the show. Linode.com slash unplugged. All right, gentlemen, it is time to get into the 2021 predictions. And I think we will start around. I think we'll just use the same route again, Alex. If you're comfortable kicking off with your predictions,

we'll start with you, sir. Oh, come on. Can we do reverse alphabetical order or something so I can steal someone else's? No, all right. Here's what, you know what we'll do is I'll start. Okay, so that'll make it, we'll just sort of set, then we'll just go around, everybody will do one.

And now remember the format is you have to, this is so hilarious, you have to make your grand prediction, then everybody can kind of chime in, the ref is going to make sure everything kind of lines up and is actually accountable, and then we have you repeat it in a lockable, precise version. And that is what will be held account. Not the math that got you there, but the actual locked in version. And I'm going to start

with mine. I think this is a long shot, but I feel like having a little bit of fun. I think our friends over at System76 are going to look very heavily at forking GNOME to some degree. Now, I don't know if fork is too heavy of a word here, but I think when you look at the direction of GNOME Shell 40 and some of the UI fundamental changes there and how Pop!OS is kind of acting as a layer between Upstream and System76's customers, acting as a layer between Upstream and System76's customers, I think they're going to find this in their best interest to kind of take the endless OS type approach where they don't stay forked forever,

but they kind of go off in their own direction. They build there, and then they contribute back Upstream, and what gets upset gets upset, makes it in, and, you know, everybody benefits. But that's going to be some sort of transition that System76 makes in 2021. I don't know how you're going to boil that down into something that you can say concisely.

What counts as a fork? What's different enough, right? It can't just be GitHub, right? That's what I think. If we can come to an agreement on a fork, then I think the lock-in is that System76 forks GNOME. But I think the fork here has to be, and I don't really know how to articulate how Endless does it. But what I roughly understand about Endless is they essentially stabilize on a version of GNOME Shell.

They build on top of that for a while. So essentially a few releases pass them by by and they pass along what they've done upstream and then they kind of like re-snapshot off of say GNOME Shell 338. They'll build off of that for a while, maybe GNOME 40 and GNOME 41 even ships, then they kind of rebase again off of that and kind of move forward and all while sending the work they do upstream to the core project. while sending the work they do upstream to the core project? That's not exactly what they do. Actually, Endless does far less upstreaming of their changes to GNOME Shell and other GNOME components than you would expect

because they retool the UI significantly in a lot of the underlying frameworks for their particular purpose. And as part of that, you can think of it closer to the process that eventually led to the Unity desktop existing than the process that would lead to improving the GNOME code base itself. Wouldn't that actually be kind of almost poetically perfect in a sense because it was the death of Unity

that compelled System76 to launch Pop!OS? Okay, Rev, how about this? System76 forks GNOME Shell in the same spirit that Endless OS does. I think that's probably about as precise as we can get with this. We'll just have to have a fun debate and discussion next year around this time when we decide if it happened. Right. Well, because the reality is it's probably not going to look exactly

like what's been done before. So I can't just point to a specific thing and say it's going to be just like that, is what I suspect. I do think that you're right, that there will be a sort of a line in the sand of like, was this a significant change? Did it cause some development friction?

Did it represent significant work from System76 to make this change or to make the process different? And we'll have a sense of that next year. Well, you could just look at where the GNOME Shell package comes from in Pop!OS, and it comes from Ubuntu. So if you look in a year's time and the GNOME Shell package that's on a stock install of System76 comes from their PPA, then you could argue that that's a light or heavy fork,

depending upon how changed it is. But if they're using GNOME Shell from Ubuntu, then it isn't really, is it? I think what the line should be is if GNOME Shell starts shipping from a System76-owned repo, that's at least the beginning of some kind of shift there. And I think that's a line I'd be comfortable with passing or failing by. You could say that System76 starts shipping their own

or starts hosting their own GNOME shell version or something like that, maybe. Maintaining is the key word. Maintaining. Okay. All right. Okay. All right. I'm going to lock that in. I predict that System76 will begin in 2021 to maintain their own GNOME Shell packages. Good? Yeah. We're good with that one? Good job. Okay.

All right, Mr. Payne, would you like to go next? I'm going to say that Linus announces a change in his day-to-day involvement with the kernel. Whoa. What do we think about that? It makes me sad. I think he maybe should do it, to be honest with you, because he himself says he mostly just does email now and that's got to get tedious

I don't know that it necessarily means he's not involved at all anymore but a significant change to how he contributes day to day How are you going to define significant change? I feel like that's kind of vague Well I mean we'll be reporting on the show it'll be news in the

open source community right there'll be articles on the register, ZDNet, LWN will talk about it, right? So the prediction could be something like, in 2021, it'll become public that Linus is stepping down or is becoming less involved with day-to-day operation or something like that? Mm-hmm. What do you think, Judge? What do you think, Judge? Yeah, I think we need to refine it slightly more than that. We need to be more precise about it. I mean, what exactly is his job title at the moment?

Like benevolent dictator or whatever, head maintainer. Like you'd have to say, I think, that he's not going to be the head maintainer anymore. Yeah, okay, right. Like maybe he's still involved, but he's no longer doing the actual releases for each version. I don't know. I like this idea that, so where you were originally going with this, Wes, that I liked was this idea that Linus is still the boss, but he pulls back a bit and just lives more of his life. Right. You know, he still shows up. He maybe makes some consequential decisions or settles

arguments on the mailing list, but he's not the one cutting these releases, you know, each cycle or something. And I think that's the core of what you were going at originally. And that's the scenario that I really hope eventually does happen for Linus. I hope it's not the burnout scenario. But I wonder if maybe what is left for him

is just the parts he does enjoy. He may actually just genuinely enjoy what he's doing. Linus, clearly you just need to come on the show and talk to us about it. We know you're listening. So what do you think, Judge? Is it something to the extent of like – because here's the other bar we could go by. Instead of really deciding how we want to define less involved, what we could just define is how much reporting is there about Linus stepping back in some form? And if there is reporting, like there's a mail, if he posts a mailing list, you know,

saying that, you know, I'm going to take, I'm going to take every week off or something in a month, something that gets then picked up on LWN, I think that could be a qualifier. That's something we could judge against. Maybe, but I think it doesn't really matter who reports on it. I think the fact that he publicly announces it,

I think that's the key thing. That works. So I could say something like he publicly announces stepping back in his day-to-day roles in the kernel. Yeah. Yeah. I think that's pretty easy to say if it happened or not. All right. You ready to lock it in, Wes? I predict that Linus Torvalds will make an announcement sometime in 2021, and he's stepping back to some degree from his day-to-day involvement with the Linux kernel. All right, very nice. And then we have a little breaking news kind of related.

JJ, you had something you wanted to pass along. So I was trying to look and see if there was any other developments from Linus Torvald, and apparently when I was looking at the main page, his birthday was yesterday. Hey, happy birthday, Linus. And it doesn't look a day over 37. All right. So, Wes, that's a good one. That's I hope that it's a positive announcement if it happens. All right, Drew, are you ready to make your first prediction here on the show? So I predict in 2021 that Google will release an official Chrome OS installer for x86 machines.

Oh, that's good. And they just recently bought that company that... Neverwhere. Yeah. So how does that... What's official? Where's the gray area in that after this acquisition, I'm wondering? Well, specifically not something like Neverwhere branded. So they purchased Neverwhere,

who makes a product that allows you to install Chrome OS on any x86 machine. And so now that Google owns them, there will be an officially Google-sanctioned product that you can install on x86 machines. Well, and so there'll be an ISO that you can download from a Google domain, maybe. Yes. Wow. Would something like an official Chromium OS from Google count?

I think it'll be Chrome OS an official Chromium OS from Google count? I think it'll be Chrome OS, not Chromium OS. And it's pitched as any PC? I think this is solid. Any x86 PC. Right. Yeah. All right.

I predict in 2021 that Google will release an official ISO image for Chrome OS for x86 computers. What I really like about that one, Drew, is you've been, you know, following the news and then extrapolating from that, which is what my strategy tries. I try, that's generally my strategy is like, I got to leverage the fact that I follow the news every week. So let's just take those to their natural conclusion. And that's exactly what you did there. Hey, we're going three for three this year okay alex let's see how your first one stacks up all right so my mine is right up your street it's about the raspberry pi i predict that the raspberry pi 5 if that's what they call it uh the next major version of the raspberry pi so i think 5 is a decent guess will ship with support for proper storage. So it will have either a SATA port,

an mSATA, or an NVMe slot on it. Ooh. Oh, gosh, I hope so. I sure hope, yeah, right? Boy, so it's not necessarily a Raspberry Pi 5, you're saying, but a Raspberry Pi product. Yeah, yeah. So I think the next major release

that happens in 2021 will have support for an NVMe or an M.2 SATA or some kind of built-in storage. Hopefully not eMMC, but I'll take it if that gets me the prediction. Isn't the Raspberry Pi compute module right now, doesn't it have like PCI lanes on there that you can wire up if you... Yeah, Jeff Geerling's been doing a bunch of work around that kind of stuff. That could be step one in what you're talking about, right? That could be step one. Hopefully, hopefully.

Clearly, clearly, clearly, using the Raspberry Pi 4, now especially with the 8-gig model, the limitation is the storage options. It just so obviously is. And that's made better by USB 3 and being able to boot from those disks. And it's also, if you're willing to go with something like NFS or iSCSI, it has a gigabit NIC on there. But it's still not as great as just local NVMe storage or something. So I think that's pretty good, Raf. Do you think there's anything he's got to bang out with that before we lock it in?

I would only suggest that you put a caveat of there being an option for built-in storage because I could see them releasing two versions like they've got the four gigabyte and eight gigabyte ram ones now you might have one that just has an sd card and one that has an sd card and emmc or whatever yeah because the the compute module 4 uh has emmc on it now so yeah i think you're probably right with that yeah so i i would say something like uh it like there'll be one available that has an option to boot from built-in storage that isn't the SD card or something like that. All right, so I predict in 2021, the next major release of the Raspberry Pi,

maybe the Pi 5, will have an option for a proper storage system. So that's NVMe or M.2 SATA or just a genuine SATA port. Man, I hope you're right on that one. That would be great. Let's hope your winning streak continues. Yeah, let's hope. So since Joe is judge, he only gets to,

well, he only has to actually, I should say, make one crazy prediction. So we loop back around to me again, and I'm going to go with one that, my goodness, better happen, or else the industry is in a world of hurt. But I predict that Qualicom will come out with an ARM SOC that matches or beats Apple's M1.

And it will be possible to boot Linux or actually maybe just Android on those systems. I mean, this is totally quantifiable, but you're totally wrong about it. You guys are telling me nobody in mainstream ARM that ships on consumer devices is going to be able to match Apple, that Apple is that incredible at engineering?

It's not that they're incredible at engineering. It's that Apple doesn't have to care about supply chain and demand margins. Everyone else does. So, I mean, I think that makes it a fun pick myself. Yes, it's outlandish for sure. Yeah, the least likely the better. I support this.

Okay, should I remove Qualicom and say an ARM distributor, an ARM chip distributor? Yeah, an ARM manufacturer other than Apple will come out with a system on a chip that is as powerful or more than the M1 that's how I'd phrase it how do you measure that though is it a Geekbench score

or some kind of what's the metric I'll go Geekbench okay Geekbench is fine with me yeah that's the one that that's what the one

that mostly M1 folks are using for the people that don't know Project Skybridge AMD has Project Skybridge where they have ARM and x86 as a single platform. Yeah, I don't think that counts. I think it's got to be pure ARM. So we're going down on the record right here saying that at the end of 2020, we're looking out of all of 2021,

even though the whole industry knew that Apple was, you know, they've known about what these CPUs can do for a little while. So they've had years to start ramping up. But what we are saying at this point is we have so little faith that just saying an ARM manufacturer might beat the M1

is a risky prediction. And I love it. So I'm going to say an ARM manufacturer comes out with an ARM SoC that matches or beats Apple's M1 on Geekbench. Yeah. Yeah, okay. I'll leave it at that.

I was going to add another qualifier, but I don't think I need to put myself in a box. So officially, locking it in with this, an ARM manufacturer in 2021 will come out with an ARM SoC that matches or beats Apple's M1 on Geekbench. All right. Mr. Wes, you are up, sir. You got a prediction for us? I'm going to say Gnome fixes their plug-in architecture. We get a real plug-in API and a bad plug-in. While it might still hurt your system, it won't crash the whole dang shell.

I needed a good laugh before the end of the year. Thanks, Wes. It doesn't seem to be on the horizon for 40, but maybe it lands in 41 or 42. Something I'm hoping for anyway. I like this, though, because again, like your Linus one, if it pans out right, it's a good thing for all of us. Do you think there's anything he has to qualify there, Ref?

It just seems a little bit hard to measure is the only problem. Well, it'd be like a new API, right? They'd be making an announcement about here's the way we're building plugins going forward, something like that. Right. Okay, so not just an improvement, but a whole new API. Yeah, I think that's quantifiable then, yeah. They do have a new extensions

effort that's essentially like we're going to increase communication and make sure that people are sharing information. That they've already announced, but there's no real like, an API would truly be the benchmark. And I think we could probably be loose on what the definition of an API is even. All right, I guess I'll try to lock this in, huh? Go for it.

I predict that in 2021, Gnome finally fixes their plugin problem with a new API they've announced. That means a misbehaving plugin can no longer crash your whole dang shell. Again, with some of these predictions from your lips to the developers' ears. Okay, Mr. Drew, are you feeling like you got another one in you? Oh, I got another one. It's a good one. Okay, okay. I predict in 2021 that VMware will go independent. Ooh, this is a fascinating little insight you must have picked up somewhere recently.

What leads you to this conclusion? They're kind of leaning away from Dell. That's really the only thing is just kind of reading what they're doing. I use VMware at work and I don't see a lot of Dell influence in them anymore. So would VMware getting sold to another company not qualify then? No, I think they will go independent. Okay. The thing is on that one, though, Drew,

is that VMware are basically trying to steal Red Hat's lunch with Tanzu and their play against OpenShift. So lots of people have to pay the VMware tax right now. And I think the biggest risk that VMware have moving forward is people want to just run containers on the bare metal and use something like OpenShift or Rancher or any other Kubernetes distribution to kind of get rid of that VMware tax. So I think maybe some of the stuff you're seeing is them focusing on that kind of hybrid cloud enterprise-y type sweet spot because they can see a future where VMs just aren't needed anymore.

Oh, I don't think you're wrong there, but I think that they have a lot of irons in the fire and I don't think that Dell plays a huge part in their future. Yes, fascinating prediction. I'll be interested to see how this one plays out. It seems pretty straightforward there, Ref. Yeah. I predict in 2021 that VMware will go independent. That is that's a spicy one. I like it. Okay, Alex, you're up. You got another one for us? I do indeed. So a project I've done recently was running a virtual GPU on top of Proxmox. And in the research for that, I came across some videos by Wendell and he was talking about how the new NVIDIA graphics cards might have something called SRIOVv which would allow you to slice up

one gpu and pass it through to multiple vms at the same time now this is this technology already exists in the data center if you're willing to pay for something like nvidia grid but it doesn't really exist at the consumer level because obviously nvidia and amd and all that want to sell more graphics cards. So my prediction is this. The upcoming Intel XE discrete GPUs will support SRIOV out of the box. All right. That's pretty concise. That also won't happen. Well, then it makes it even more interesting.

Well, because they already do GPU splitting now with a different technology, mediated devices through GVT. And so that's effectively a total substitute for SRIOV. Yeah, why do you think they would change it, Alex? I must admit, I wasn't up to speed on the technical nuance there. I thought SRIOV was just a different name for the same thing so if i could qualify my prediction to be that the upcoming intel discrete gpus will support virtual graphics card slicing okay let's lock unless judge unless you have any uh anything you want i think we should lock it in yeah let's lock it in the upcoming intel xe discrete graphics cards will support virtual graphics card slicing through SRIOV or mediated devices through the GVT technology.

I like that. That's good. I'd love to see that because it just gets more value out of those machines. I'm going to go again, and I'm going to say this is my last one, so I have a couple more on here. Like I was really tempted to do one about the M1, but I'm going to save that for the mumble round. I've got one for the M1. Okay.

So I'll save it for you guys, and I'm going to say that a distribution announces that it has more users on WSL than actual physical metal installs. I can't see that happening. I can't see them announcing because that would mean they'd have to announce numbers and none of the distros want to announce numbers. I suppose they could say they have more of a proportion and not actually say what the numbers are. Right, right.

You don't think so, Popey? You don't think it's going to happen? I think it's just really hard for what Joe says. I think it's just really hard for what Joe says. It's like, so what if it's not true, then the Australian question wouldn't say it. And if it was true, are you implying, Popey, that it may already be the case, but nobody has just said it? No. If it was the case or if it wasn't the case in both ways, they're going to harm themselves by saying it.

I don't know. I don't know why. Why is that harmful to say that we have 3 million WSL users and we have, you know, I mean, like maybe that's probably too many. But if you think about it, I think you know what some of the actual deployment numbers are of some of these distributions. It's not inconceivable for one of them to get more users on Windows. It doesn't seem to me. It might be more conceivable for a smaller distro. Yeah. Because the numbers are so small that it might be actually a thing that they wouldn't mind saying.

Right. Yeah, or a purpose-built distro. Well, I feel, okay, so that I feel like is not, that's where this prediction I feel like is a little bit too easy. I can't say like- That's not quite in the spirit of what you're suggesting. Yeah, there's a couple of distros that are targeted just to WSL, and I feel like that's a no-brainer.

Here's what I think kind of gives me an edge up, is it might come from the Microsoft side. Maybe not even intentionally or something like that. Maybe it's like an offhanded comment that, oh, yeah, yeah, you know, XYZ actually has more users on WSL than they have physical installs on the wild. You know, and it's just not even something that's released by the distro. But they wouldn't know that. They can't know that number

because they are not in the infrastructure for the distro. True, true. Okay. Here's the root of what I'm trying to get out there is, I think you have a lot of middle tier distros that maybe have 30, 40, 50,000 installs, maybe as many, if they're lucky, of 100,000. And I don't think that's an impossible number

to get surpassed on WSL. And I think it's going to happen within the next year. But I guess what you guys are saying is it may happen and we'll never know. So how do I make... Because to me, it just seems like an obvious thing that's going to happen. I'd love to get on the record saying it's going to happen, but there's no way to measure it. I think it's a reasonable thing for you to predict.

And I think it's not unreasonable to think someone might let that slip or someone in a fit of pique and PR might decide to actively promote that as a thing. I think it's fair to say. Yeah. That's why I should have gone with my Rocky Linux one because I could be an easy winner. Damn it, damn it. See, I could make one up on the spot, but I don't really know.

Okay. So I think, Judge, what it comes down to is I could still go with it, but it's a small shot that it actually ever goes public. Yeah, as the ref, I can say that's totally fine to say it, but as Joe, I'm going to say you haven't got a chance, mate. Yeah, okay. I know it's going to happen if it hasn't happened already.

It's going to happen any day. I think it's a good prediction. I'm going to do it. I'm going to do it because Popey's egging me on, so I'm going to do it. Shoot for the stars, Chris. Do it. All right. I'm going to say that in 2021, it becomes public that a distribution has more users on WSL than physical installs in the wild. Okay, Wes, let's round out yours with your third and last prediction. Ooh, okay. My final prediction. Got to make it a good one, I suppose.

Yeah, make up for mine, would you? Because that was rough. All right, let's say that there's a Rust-based kernel module that is widely distributed, either included optionally in the main tree or shipped by Ubuntu, Fedora, or Arch. I'm sorry, what did you say there, Wes? Did you say...

Rust! there, Wes? Did you say... Rust! Every time I see this coming up, I know that it's Rust. Even when I haven't listened to the show. Just recognize the waveform at this point. Yeah. I love this idea, and it's a solid one.

And you know it'd get some hype, because, you know, Rust. So... i don't really have any problems with this because i feel like it would go public you know most modules nobody really talks about the news but a rust module i'm talking about it what do you think judge yeah i think lock it in i predict that in 2021 the first rust-based kernel module will be widely distributed either optionally in the main tree or shipped by Ubuntu, Fedora, or Arch. Man, you know, you didn't even have to lock it down. You could have left it open-ended for if it ships in a disk or not.

Okay, well, Drew, are you ready for your... So now you tell me. Sorry, there it was. You know, you always got to play the game. You always got to play the game. always got to play the game drew are you ready for your uh third and final prediction sir so another enterprisey prediction from me rancher ownership will change hands what no i don't see that one they just didn't susa just buy them this year they just got a new brand on them yeah and it's like by the way one

of my unofficial predictions was that susa would up their Kubernetes game and they did. They bought Rancher OS, dude. And it's like, that's like the big thing that they're like proud of when they're peacocking now. They're peacocking about Rancher OS. Does it include SUSE getting bought out by somebody else? It's very possible that somebody could buy SUSE for Rancher. Oh, so you're saying it changes ownership via SUSE changing ownership. Or Google just straight up buys Rancher. Who knows? I mean, there are lots of options here. But at the end of the year, I don't think that Rancher will have the same owner, be it whether SUSE gets bought or Rancher gets bought, one of the two will happen. It's about time for Sousa to get bought.

It really is, yeah. I really almost went with Sousa ownership will change hands, but I think the Rancher ownership one is a little spicier, so I like it better. I think that's pretty workable. What do you think, Judge? Yeah, I think go for it. All right. I predict in 2021 that Rancher ownership will change hands.

It's nice and tidy too. I like that. Okay, Alex, are you ready for a third and final prediction? I think so. I'm going to continue on the theme of hard drives like last year, because apparently that one came true, even though I didn't think it did. So the Easy Store line, I'm going to base my prediction off the easy store line that you can buy at Best Buy. The, um, prediction is that you'll be able to buy a 20 terabyte drive for under 250 bucks at Best Buy within 2021 at some point. I think this is too similar to last year's man, but it's price-based though. That's the, so instead of, so instead of saying it's, it's going to be X amount of storage, you're saying you're going to get X amount of storage for a price,

which is kind of a different take on it. Yeah, true. He also specifically said Best Buy, and I kind of partly want Best Buy to go out of business this year just to screw with it. I don't know. Do you think he needs to tweak it to make it work, Ref? Or do you think it's...

No, I think that is totally testable. So let's just do it. I predict slash hope that a Western Digital EasyStore 20 terabyte hard drive will cost $250 or less at some point in 2021. Remove the Best Buy mention.

Well done. Yeah. Oh, you can edit that in if you want. No, it's probably better. It's because, you know, because now it's like if it's on Newegg. The thing is, the easy stores are best by exclusives. Oh, so it's no choice.

So I guess it's in there anyway. Joe, we saved the last and the best right for now. And it's you and it is your crazy prediction of the year. Are you ready, sir? Fuchsia, I think you'll be able to download an image of that and get it working with a free desktop environment such as xfce or plasma or gnome or something and run a browser all on fuchsia not officially necessarily but someone will hack it together so you'll be able to boot a fuchsia system with a browser and a proper desktop environment. And I'm not saying

what the performance is going to be like, but you will be able to use a browser. Fascinating. And it's probably a BlinkBase browser. Probably, yes. Okay. So it's Fuchsia with all the way up to a browser is essentially the prediction. I think that's pretty straightforward. You want to lock it in? I predict that in 2021 you'll be able to use everything up to a browser in Fuchsia. That is going to make me nervous if that one happens. I don't want

that one to come true. Okay, so we're going to now go into the Mumble Room round. So good job, gentlemen. Mumble Room, tag me with mum in the chat room and I'll call on some of you to get some of those in. And gentlemen, gentlemen. Mumble Room, tag me with mum in the chat room, and I'll call on some of you to get some of those in. And gentlemen, congratulations. Your 2021 predictions are locked in. I want to also thank our Unplugged Core contributors at unpluggedcore.com. That's a membership program that helps keep this program independent and helps reduce the ad load needed

to make the show profitable. And you also, as a core contributor, get access to two feeds, a limited ad version of the show, same full production, just limited ads, and a second feed, which this week is going to be a massive, massive recording. That's the full live stream feed, all our screw-ups, the stuff that never makes it into the show, and a huge pre- and post-show that doesn't get recorded because the disk space in the main recorder would be crazy. It's basically an extra show that never gets released.

And I suppose the other perk of it is it's released as fast as we can, so usually within an hour or two of us wrapping up recording, it goes out in the feed. The downside is if you're driving in the car or you want something that's a little less time-consuming, that's where the main feed's a little nicer because it's the full production, it's been tweaked EQ-wise, it sounds better, and it's shorter runtime.

So that's why we make two feeds available, shorter or longer, and that just is part of being a core contributor. So thank you very much to our unplugged core contributors. And we now transition to the mumble room, and we'll start with Riton. My prediction last year was that Canonical will IPO and Microsoft will buy some parts of it. That didn't happen. And my prediction this year is there will be a phone with a Linux kernel sold in stores that's not Android.

Will stores exist? Yeah. Maybe Amazon. Okay. Does Amazon count? I think Amazon should count. I like that. All right. stores that's not android will stores exist yeah maybe amazon okay does amazon count i think amazon should count i like that all right okay uh poppy i know you've got a few so uh what are yours sir don't be good the bad or the ugly oh uh all of them just give them all to me just rapid fire all right so in contrary to abin i predict in 2021, Canonical will neither IPO nor be bought by Microsoft. Okay. In 2021, Hector Martin and collaborators will get 90% of the hardware in the first generation M1 Macs booting on Linux.

Now, does the GPU count as 10% or does that count more? I would go by whatever Hector's gauge of what 90% is. Okay. Okay. That'd be great to see that. And then is this next one the good or the ugly? The ugly. Okay.

In 2021, Red Hat will do something disliked by the community, which intentionally makes life hard for the Rocks Linux developers. Oh, I think it's going to be the opposite but that would be very spicy all right thank you sir i like those uh bite i know you've got a few too you want to jump in i don't think google goes that big with fuchsia i think google coming here will take their next big step, but only into that. All right. I hope you're right about that.

I think that'd be really good. Now, did you have a couple other ones? I think 2021, there will be a credit card size computer with a full-fledged RISC-V CPU. Ooh, I like that one. That's a good one. All right.

I'm going to keep moving because we've got more, but we will have the post show as well. JJ, get yours in. I predict that in 2021, streaming DRM will restrict users on Linux from enjoying content. I hope not. That would be a fail. Jeff, you've got a Silverblue prediction.

Yeah, I predict that I will deploy Fedora Silverblue on mostly home-built ThinkPads to all my customers who want it. And I'll make a profit for the first time. I mean, it's impossible. Y'all know I'm not going to make a profit on the hardware. That's impossible. We all know that. But I actually believe I can actually support Silverblue all over the world if I need to. I can just FedEx on the new box when it breaks, just like the enterprise guys do and like you call the helpline.

So I need to go official, I guess. What do y'all think? Have y'all tried Silverblue? Yes. Well, Drew has talked about it a little bit on the show. I hope that works out for you. I think that's a good prediction. And then, Colonel, you have a PinePhone prediction.

Yep. So my prediction is that sometime in 2021, one of the software distros for the PinePhone will become daily driver ready. So for daily driver, what do we define that as? Like has that navigation and what? Like what's daily driver? Or could Chris stand to use it for a week? Let's not kid ourselves. Come on. Let's not. If the Pine phone is, you know, installed and set up, somebody who's not a complete and total nerd can use it day to day for phone calls, text messages, and some web browsing. All right. I like it.

If you're listening to this and thinking, gosh, I'd like to get a prediction and I've got a good one, let us know. Go to linuxunplugged.com slash contact. Over the next few episodes, we'll read a few more 2021 predictions. Because let's be honest, it's still pretty early in the game. So it'll still count if you get it in January via email.

Wouldn't that be neat? Maybe you got a good one in you. So that's linuxunplugged.com slash contact. You can find our sponsor at Cloud Cloud Guru, on social media. It's really simple. You just go to slashacloudguru at YouTube, Twitter, or Facebook. All of that is there. Links to subscribe or find.

Usually we have a bunch of links to stuff we talked about, but this is a special episode. But generally you can find our show note links and everything like that at that Linux Unplugged website. It's really quite simple. And then, of course, we'd love to have you join us live. We do this show on Tuesdays at noon Pacific, 3 p.m. Eastern. See you next week. Same bad time, same bad station. And there is a lot to it.

You can join our mumble room and hang out in our virtual lug. Of course, we also have the LUP lug on Sundays. So you set up the mumble room once, you get access to the show and the Lup Lug on Sundays. It's pretty great. All this stuff that we do live it's at jupiterbroadcasting.com slash calendar. But last but not least, let me just say here's to a great 2021.

Thank you for listening. We appreciate everyone who downloads or shares this show. All of our core contributors. All of you. Hope you have a great 2021 and we'll see you right back here. Not next Tuesday. No, my friends. If you can believe it, we'll see you right back here

next frickin' year! Come on, everybody, next year! Happy New Year! Happy New Year! Ha ha ha ha ha ha! Thank you.