WEBVTT

00:00:11.261 --> 00:00:15.921
<v Chris>Hello, friends, and welcome back to your weekly Linux talk show. My name is Chris.

00:00:16.081 --> 00:00:16.721
<v Wes>My name is Wes.

00:00:16.921 --> 00:00:17.521
<v Brent>And my name is Brent.

00:00:18.141 --> 00:00:21.781
<v Chris>Hello, gentlemen. Coming up on the show today, we've been stress testing open

00:00:21.781 --> 00:00:23.481
<v Chris>source AI agents all week.

00:00:23.881 --> 00:00:27.521
<v Chris>From OpenClaw to projects you probably never heard of. We'll talk about what

00:00:27.521 --> 00:00:31.381
<v Chris>actually held up, where things kind of fall apart, and what is definitely hype.

00:00:31.601 --> 00:00:34.101
<v Chris>And then a few surprises we didn't see coming.

00:00:34.341 --> 00:00:38.661
<v Chris>Plus, Brent has a $25 Wi-Fi upgrade you're going to absolutely want to steal.

00:00:39.201 --> 00:00:41.881
<v Chris>Stay tuned for that. And then we're going to round it all out with some great

00:00:41.881 --> 00:00:43.321
<v Chris>boosts, some picks, and a lot more.

00:00:43.441 --> 00:00:47.901
<v Chris>So before we get to that, time-appropriate greetings to our virtual lug. Hello, Mumble Room.

00:00:48.881 --> 00:00:51.461
<v Mumble>Hello. Hello. Hello, Brent.

00:00:52.001 --> 00:00:53.761
<v Chris>Hello. Sounding good?

00:00:53.941 --> 00:00:53.961
<v Brent>That's true.

00:00:53.961 --> 00:00:56.521
<v Chris>Hello, everybody. Hello, everybody. In the quiet listening, too.

00:00:56.581 --> 00:00:57.701
<v Chris>That Mumble Room is always going.

00:00:58.761 --> 00:00:59.781
<v Chris>JupiterBroadcasting.com slash

00:00:59.781 --> 00:01:03.741
<v Chris>mumble. And a good morning to our friends over at Defined Networking.

00:01:03.841 --> 00:01:08.421
<v Chris>Defined.net slash unplugged. Go check out Nebula, a decentralized VPN built

00:01:08.421 --> 00:01:09.941
<v Chris>on the open-source Nebula platform.

00:01:10.241 --> 00:01:13.641
<v Chris>We absolutely love it and are using it in more and more ways every single day.

00:01:13.781 --> 00:01:17.041
<v Chris>If you go to defined.net slash unplugged, you get 100 hosts for free,

00:01:17.161 --> 00:01:19.081
<v Chris>no credit card required.

00:01:19.321 --> 00:01:22.961
<v Chris>Now, this is really the actual difference. Their free tier doesn't exist just

00:01:22.961 --> 00:01:25.521
<v Chris>to funnel you into a VC-funded SaaS roadmap.

00:01:25.941 --> 00:01:29.781
<v Chris>It really is something that you have full governance over. You can go from managed

00:01:29.781 --> 00:01:31.421
<v Chris>to completely self-hosted, vice versa.

00:01:31.661 --> 00:01:34.481
<v Chris>You own the network. You own the identity. You own the infrastructure.

00:01:35.321 --> 00:01:39.601
<v Chris>You don't have to worry about a control plane going down, any of that.

00:01:39.601 --> 00:01:43.101
<v Chris>And Nebula's decentralized design means there isn't a single point of failure.

00:01:43.261 --> 00:01:47.561
<v Chris>And with their managed Nebula product, they can take care of all of the bits

00:01:47.561 --> 00:01:50.161
<v Chris>for you. It is incredibly scalable.

00:01:51.083 --> 00:01:54.283
<v Chris>One of the things I appreciate on being on a very limited connection right now,

00:01:54.443 --> 00:02:00.163
<v Chris>I'm back on LTE for a bit, and Nebula is so smart about the way it uses network traffic.

00:02:00.583 --> 00:02:06.603
<v Chris>And it's an order of magnitude difference between some of the other Mesh VPN systems and Nebula.

00:02:06.703 --> 00:02:09.083
<v Chris>It's an order of magnitude difference in the efficiency. I'm very,

00:02:09.183 --> 00:02:12.283
<v Chris>very impressed. And you will be too. Go check it out. 100 hosts for free.

00:02:12.523 --> 00:02:14.603
<v Chris>No credit card. No lock-in.

00:02:15.003 --> 00:02:17.703
<v Chris>Defined.net slash unplugged.

00:02:20.336 --> 00:02:24.116
<v Chris>Just around the corner, 32 days until Planet Nix and scale.

00:02:24.516 --> 00:02:29.596
<v Chris>That means 25 days until Brent needs to be on the road and four more Linux Tuesdays

00:02:29.596 --> 00:02:32.696
<v Chris>on a Sunday before we are in Pasadena, California.

00:02:32.956 --> 00:02:36.536
<v Brent>Why is my heart racing so much? Why am I feeling so stressed? What's the deal?

00:02:37.376 --> 00:02:41.056
<v Chris>You know, just focus on how awesome Planet Nix is going to be this year.

00:02:41.396 --> 00:02:45.276
<v Chris>This is the, I mean, they've got a vision for it. Phlox has really figured it

00:02:45.276 --> 00:02:46.476
<v Chris>out. They know how to do this.

00:02:46.876 --> 00:02:51.356
<v Chris>The first one's under their belt. They're the perfect just organizer for this

00:02:51.356 --> 00:02:53.676
<v Chris>exact kind of event because they get the community.

00:02:53.876 --> 00:02:55.836
<v Chris>They get the business side. They get the builder side. Like,

00:02:55.936 --> 00:03:00.356
<v Chris>it's just chef's kiss. I think it's going to be a good one this year. The agenda is live.

00:03:00.496 --> 00:03:03.356
<v Wes>Yeah, that's right. I'll be giving a talk. Our buddy Alex is giving a talk.

00:03:03.876 --> 00:03:08.536
<v Wes>There's a lot of nice looking talks, including something about Nick's Meets

00:03:08.536 --> 00:03:12.496
<v Wes>Web Assembly and Nick's BSD. What's that?

00:03:12.836 --> 00:03:16.196
<v Chris>Oh, OK. Yeah, our plan of Nick's coverage is supported by Phlox.

00:03:16.296 --> 00:03:19.276
<v Chris>And they're going to be there. We're going to see them. They're focused on making

00:03:19.276 --> 00:03:21.396
<v Chris>reproducible dev environments actually usable.

00:03:21.656 --> 00:03:25.096
<v Chris>You should check out Phlox. It's the second year they're sending us, and it's awesome.

00:03:25.216 --> 00:03:30.296
<v Chris>And, of course, at the same time, Scale23x is going on. You do need to register

00:03:30.296 --> 00:03:32.416
<v Chris>for Scale to go to either Planet Nix or Scale.

00:03:32.636 --> 00:03:37.076
<v Chris>Use our promo code UNPLG, U-N-P-L-G, to get 40% off that.

00:03:37.396 --> 00:03:40.396
<v Chris>Can't wait for Scale. We'll have a meetup. Meetup.com slash Jupiter Broadcasting.

00:03:40.436 --> 00:03:44.536
<v Chris>Our meetup page is up. It's a placeholder, but it's there. Huge. Very excited.

00:03:46.016 --> 00:03:49.276
<v Chris>I think it's going to be great, guys. It's been a while since I've been in nice,

00:03:49.396 --> 00:03:51.216
<v Chris>sunny Pasadena. I think it's going to be beautiful.

00:03:51.476 --> 00:03:54.996
<v Chris>We always have a great crew down there. And I think it's going to be a great

00:03:54.996 --> 00:03:56.136
<v Chris>Planet Nix and a great scale.

00:03:56.276 --> 00:03:58.016
<v Wes>So many wonderful nerds all in one place.

00:03:58.356 --> 00:04:02.016
<v Chris>Such a rare thing. So I hope you can make it. Even if you can't make it to the

00:04:02.016 --> 00:04:04.836
<v Chris>events, but you're in the area, meetup.com slash Jupiter Broadcasting.

00:04:04.836 --> 00:04:06.896
<v Chris>Let us know so we can let the venue know.

00:04:10.584 --> 00:04:15.744
<v Chris>Well, Brentley, Wes and I have been hearing bits and pieces of you solving Wi-Fi

00:04:15.744 --> 00:04:19.764
<v Chris>problems, which often starts with something not working right.

00:04:19.984 --> 00:04:23.524
<v Wes>Yeah, our ears are burning and our switches are burning. We want more deets.

00:04:23.724 --> 00:04:25.424
<v Chris>We have to imagine this has turned quite in the story.

00:04:26.364 --> 00:04:30.884
<v Brent>Yes. Well, networking is not my favorite thing. So it only ever starts with some kind of complaint.

00:04:31.284 --> 00:04:34.604
<v Brent>And the complaint this time around, I've been spending more time with my parents

00:04:34.604 --> 00:04:39.744
<v Brent>and at my parents' place. And my mother was like, hey, I can't really get Wi-Fi

00:04:39.744 --> 00:04:43.124
<v Brent>like lazing in bed on a Sunday morning from my bedroom. I was like, what?

00:04:43.644 --> 00:04:46.904
<v Brent>You're only saying this now you've had the same like Wi-Fi set up for the last

00:04:46.904 --> 00:04:48.744
<v Brent>five, six years, something like that.

00:04:48.824 --> 00:04:51.364
<v Chris>Is the is the old router pretty far from the bedroom?

00:04:51.704 --> 00:04:55.664
<v Brent>It's exactly at the opposite end of the place. So they're on the main floor,

00:04:55.864 --> 00:05:00.444
<v Brent>you know, at the total end of the house. And the router is in the basement at

00:05:00.444 --> 00:05:01.564
<v Brent>the opposite end of the house.

00:05:01.744 --> 00:05:05.144
<v Brent>So it kind of makes sense. I just never knew it was an issue.

00:05:05.144 --> 00:05:09.404
<v Wes>I bet you've got one of those sturdy, well-built Canadian homes, too.

00:05:10.064 --> 00:05:11.404
<v Brent>Yes, the window-randed ones.

00:05:11.404 --> 00:05:12.504
<v Wes>Not thin walls. Right.

00:05:13.624 --> 00:05:18.724
<v Brent>So I never realized it was an issue, but like, this should be solvable, right?

00:05:19.144 --> 00:05:26.864
<v Brent>So I dove into their router, which is a Linksys EA8100V2, which they got a little

00:05:26.864 --> 00:05:30.684
<v Brent>while ago to like boost Wi-Fi and stuff. They don't need anything fancy.

00:05:30.784 --> 00:05:35.344
<v Brent>They're not doing any networking stuff that's fancy. This is just like a couple

00:05:35.344 --> 00:05:39.684
<v Brent>of cell phones on the network. They have like a thermostat, IOT device.

00:05:39.944 --> 00:05:47.484
<v Brent>They've got a TV and I have a crazy, you know, put together NAS running off

00:05:47.484 --> 00:05:50.784
<v Brent>an old think pad that like does backups for them on the local network.

00:05:50.904 --> 00:05:53.904
<v Brent>But so that's not huge requirements for them.

00:05:54.324 --> 00:05:57.264
<v Brent>Probably the biggest requirements is whenever I show up, I do a bunch of stuff,

00:05:57.424 --> 00:06:01.664
<v Brent>but like they don't need super modern wifi speeds or anything like that.

00:06:01.664 --> 00:06:06.184
<v Brent>They just need coverage and reliability, basically.

00:06:06.464 --> 00:06:10.224
<v Brent>So I decided this week to solve this problem for them.

00:06:11.024 --> 00:06:15.204
<v Brent>Even though I hesitate to play with networking stuff because it's always a rabbit hole.

00:06:16.144 --> 00:06:19.184
<v Brent>And I dove into their router, which was just running the stock firmware.

00:06:19.244 --> 00:06:22.584
<v Brent>And I thought, OK, I can optimize at least, you know, some Wi-Fi channels,

00:06:22.864 --> 00:06:28.024
<v Brent>look at some what the neighbors are doing and try to choose appropriate settings.

00:06:28.184 --> 00:06:32.084
<v Brent>And I discovered that their stock firmware was like from 2022.

00:06:32.444 --> 00:06:34.584
<v Brent>So I thought, sure enough, I could just update this thing.

00:06:34.724 --> 00:06:36.444
<v Chris>Oh, yeah. That might be an easy win right there.

00:06:36.544 --> 00:06:37.704
<v Brent>You know, security is important, right?

00:06:37.704 --> 00:06:40.904
<v Chris>Yeah, and an update. Maybe there's an improvement in how it manages radios or

00:06:40.904 --> 00:06:41.824
<v Chris>something like that. Who knows?

00:06:41.944 --> 00:06:45.784
<v Wes>It is so great when you do just update something and it performs better and you can just be done.

00:06:45.924 --> 00:06:48.284
<v Chris>It probably doesn't happen as often as I'd like to hope it does.

00:06:48.324 --> 00:06:50.664
<v Wes>No, but that it's happened at all. It clearly sticks once it does.

00:06:50.764 --> 00:06:53.284
<v Brent>In this case, I was really hoping this would be just an easy fix.

00:06:53.664 --> 00:06:57.884
<v Brent>But it turns out, you know, of course, as it goes, this thing is end of life already.

00:06:58.024 --> 00:07:01.624
<v Brent>So the last update that will ever exist was back in 2022.

00:07:02.264 --> 00:07:09.304
<v Brent>Oh, so the rabbit hole shows up and I thought, well, I could I install open WRT on this thing?

00:07:10.044 --> 00:07:16.584
<v Brent>You boys had some adventures, what, two weeks ago trying to use the open WRT one.

00:07:16.964 --> 00:07:17.144
<v Chris>The one.

00:07:17.304 --> 00:07:22.504
<v Brent>Yeah. And you were a bit hesitant. You started using that to solve the clinic

00:07:22.504 --> 00:07:25.684
<v Brent>networking, but you eventually moved away from that, right?

00:07:26.164 --> 00:07:29.484
<v Chris>Well, we're still using it for Wi-Fi. It is a great little device for that.

00:07:29.484 --> 00:07:33.084
<v Chris>We were just having some issues with the radios that are in there,

00:07:33.224 --> 00:07:37.084
<v Chris>and if you're connecting to another Wi-Fi network, the performance was pretty

00:07:37.084 --> 00:07:40.244
<v Chris>bad if you're sort of daisy-chaining Wi-Fi networks for what we were doing.

00:07:40.644 --> 00:07:46.304
<v Chris>I never tried kind of beyond that, but when we just put it into a general AP, it's been fantastic.

00:07:46.964 --> 00:07:52.844
<v Brent>Nice. So I thought I would kind of lean on your experiences and dive right in.

00:07:53.024 --> 00:07:57.144
<v Brent>So sure enough, this thing is really well-supported on OpenWRT.

00:07:57.144 --> 00:07:59.984
<v Brent>The community seems to have this well supported.

00:08:00.384 --> 00:08:03.384
<v Brent>And the flashing process is also very, very simple.

00:08:03.804 --> 00:08:05.564
<v Brent>Just use the stock.

00:08:06.837 --> 00:08:10.097
<v Brent>Firmware updater and use an open wrt image

00:08:10.097 --> 00:08:12.897
<v Brent>and there you go that's all you need so sure enough i went

00:08:12.897 --> 00:08:15.717
<v Brent>and did that and nothing happened and uh it

00:08:15.717 --> 00:08:19.617
<v Brent>just kind of complained and i tried for way

00:08:19.617 --> 00:08:22.517
<v Brent>embarrassingly too long to solve that problem

00:08:22.517 --> 00:08:25.657
<v Brent>because i really now wanted to get the open wrt on

00:08:25.657 --> 00:08:28.577
<v Brent>this thing i like started setting up

00:08:28.577 --> 00:08:31.557
<v Brent>a tftp server to like send the

00:08:31.557 --> 00:08:34.877
<v Brent>image towards this thing but i didn't know the default like

00:08:34.877 --> 00:08:37.637
<v Brent>the the recovery ip address so i was trying all

00:08:37.637 --> 00:08:40.397
<v Brent>sorts of anyways i lost a lot of time and then

00:08:40.397 --> 00:08:43.477
<v Brent>i saw some note that just said hey just reboot

00:08:43.477 --> 00:08:46.737
<v Brent>the writer and try again sure enough i tried again worked totally fine through

00:08:46.737 --> 00:08:51.057
<v Brent>the the stock firmware updater how often does that happen that was both very

00:08:51.057 --> 00:08:57.037
<v Brent>frustrating and also very rewarding so i did get it installed and uh i have

00:08:57.037 --> 00:09:02.657
<v Brent>to say it's great as always i have used a lot of these open source

00:09:02.937 --> 00:09:07.377
<v Brent>router firmwares for the last, I don't know, several decades.

00:09:07.377 --> 00:09:11.597
<v Brent>I'm wondering where you guys started, but I remember the first one was I installed

00:09:11.597 --> 00:09:16.897
<v Brent>a DDWRT on an old Linksys WRT 54G. You remember those things?

00:09:17.257 --> 00:09:18.137
<v Chris>Oh yeah. Oh yeah.

00:09:18.237 --> 00:09:22.497
<v Brent>And brought those back to life for like another 10 years. So that was my first

00:09:22.497 --> 00:09:26.717
<v Brent>experience a long time ago, but I remember like tomato was a piece of software I used as well.

00:09:26.717 --> 00:09:31.217
<v Brent>So I had had good experiences with them, but nothing too recent.

00:09:31.217 --> 00:09:35.677
<v Brent>Probably the most recent was I took my parents' old router that they replaced

00:09:35.677 --> 00:09:38.217
<v Brent>and set it up at home with the Starlink.

00:09:39.097 --> 00:09:43.017
<v Brent>So instead of using the Starlink as a router, used OpenWrt in there, and that was good.

00:09:44.244 --> 00:09:46.844
<v Chris>Actually, I think the Starlink might run a fork of OpenWRT.

00:09:47.164 --> 00:09:48.024
<v Brent>Ah, right.

00:09:48.364 --> 00:09:53.264
<v Chris>I might be wrong on that, but it is a Linux variant. Yeah, I've really,

00:09:53.444 --> 00:09:55.464
<v Chris>really liked these in the past when I used them.

00:09:55.624 --> 00:10:01.784
<v Chris>It's been many years since I've actually gave it a real visit until the OpenWRT won.

00:10:02.224 --> 00:10:07.804
<v Chris>And so I wasn't really sure how viable it still was to flash these older links.

00:10:07.924 --> 00:10:10.644
<v Chris>I thought maybe that was a bygone era that maybe they'd prevented it.

00:10:10.644 --> 00:10:15.284
<v Brent>Well, I know some of the recent firmware are much larger. So if you're looking

00:10:15.284 --> 00:10:18.504
<v Brent>at a really old device, they just don't have the storage for it.

00:10:18.904 --> 00:10:20.984
<v Chris>Do they do more now? Are they doing more functions?

00:10:21.304 --> 00:10:25.504
<v Brent>Oh, yeah. Yeah, yeah. Many more functions, which I will totally describe in just a moment.

00:10:25.744 --> 00:10:26.004
<v Chris>Yeah.

00:10:26.164 --> 00:10:30.704
<v Brent>But I have to say, one of my hesitations was that the GUI in OpenWrt,

00:10:30.764 --> 00:10:36.284
<v Brent>and you boys touched on this in Linux Unplugged 650, is a bit,

00:10:36.484 --> 00:10:39.344
<v Brent>I don't know how you would describe it. but I would describe it sort of like,

00:10:42.099 --> 00:10:48.279
<v Brent>allows you to do a lot but you need to also understand what's happening so it

00:10:48.279 --> 00:10:53.279
<v Brent>feels like a little bit more leaning towards commercial router firmware so you

00:10:53.279 --> 00:10:58.119
<v Brent>can accomplish like now with just a regular home router i can do all sorts of

00:10:58.119 --> 00:10:59.219
<v Brent>things i could never do before.

00:10:59.219 --> 00:11:00.199
<v Chris>Right which.

00:11:00.199 --> 00:11:05.619
<v Brent>Is amazing but i also find the gui to be confusing that way because i'm not a network expert right.

00:11:05.619 --> 00:11:09.499
<v Chris>I suppose i also find to me i don't really know how to describe it but it feels

00:11:09.499 --> 00:11:13.339
<v Chris>like i go to multiple places to get information that seems like it could be

00:11:13.339 --> 00:11:15.539
<v Chris>consolidated into one screen yeah.

00:11:15.539 --> 00:11:19.399
<v Wes>Okay it is that feeling of like because you've seen that it means you don't

00:11:19.399 --> 00:11:22.679
<v Wes>have a clean mapping of where do i go for this like okay well i got some of

00:11:22.679 --> 00:11:26.019
<v Wes>that when i looked at the actual physical card info but then the other part

00:11:26.019 --> 00:11:28.559
<v Wes>was in the layer that just did kind of the wi-fi protocol.

00:11:28.559 --> 00:11:29.259
<v Chris>Yes and.

00:11:29.259 --> 00:11:32.539
<v Wes>Those are not on the same page or even neighboring pages they're under two different sub.

00:11:32.539 --> 00:11:37.699
<v Chris>Menus i think if i were if i were to keep working on the regular and had to

00:11:37.699 --> 00:11:41.279
<v Chris>make frequent config changes i would probably just go look at the config file

00:11:41.279 --> 00:11:43.999
<v Chris>because i know and i know they have a pretty clean syntax too on the command

00:11:43.999 --> 00:11:47.779
<v Chris>line and that might just be the way to go it's true but what happens is i use

00:11:47.779 --> 00:11:50.559
<v Chris>these once every few years and so i just stick to the gui.

00:11:50.559 --> 00:11:54.919
<v Brent>One of the big unlocks for me this week was that i decided to just lean pretty

00:11:54.919 --> 00:12:00.399
<v Brent>heavily on having ai help me navigate the gui and also help me to optimize the

00:12:00.399 --> 00:12:02.879
<v Brent>settings for this particular hardware.

00:12:03.439 --> 00:12:06.699
<v Chris>You didn't find that like its information was so out of date that it was sending

00:12:06.699 --> 00:12:08.499
<v Chris>you to places that It didn't exist in the GUI, huh?

00:12:08.619 --> 00:12:13.739
<v Brent>No, I was actually pleasantly surprised in that it was also giving me a lot

00:12:13.739 --> 00:12:17.599
<v Brent>of information about what community members were finding worked really well

00:12:17.599 --> 00:12:18.919
<v Brent>on particular hardware.

00:12:19.119 --> 00:12:19.599
<v Chris>Oh, good.

00:12:19.779 --> 00:12:22.279
<v Brent>So because this piece of hardware, I guess, is popular enough,

00:12:22.279 --> 00:12:27.459
<v Brent>I was able to get really good information on how compatible OpenWRT was on this

00:12:27.459 --> 00:12:33.539
<v Brent>particular device in the first place, and also like optimal Wi-Fi settings for that device,

00:12:34.239 --> 00:12:36.419
<v Brent>and also for this region,

00:12:37.557 --> 00:12:42.477
<v Brent>And also considering the other devices on the network, what would be realistic

00:12:42.477 --> 00:12:46.877
<v Brent>settings to really have the network be as reliable as possible,

00:12:46.997 --> 00:12:50.397
<v Brent>not necessarily as fast as possible, because that's, for my parents,

00:12:50.497 --> 00:12:51.917
<v Brent>not one of their requirements, right?

00:12:52.097 --> 00:12:56.397
<v Brent>So I noticed immediately the network was much more stable.

00:12:56.657 --> 00:13:01.197
<v Brent>In previous weeks, it would drop off at least once while we were doing Linux

00:13:01.197 --> 00:13:06.557
<v Brent>Unplugged every Sunday. So it has been much more stable from what I can tell

00:13:06.557 --> 00:13:11.637
<v Brent>and also has better just Wi-Fi coverage in general.

00:13:12.548 --> 00:13:21.108
<v Brent>But I didn't think like that was enough. And so I decided to take this to the next logical step.

00:13:21.468 --> 00:13:22.648
<v Chris>This is the part where it does more.

00:13:23.648 --> 00:13:27.828
<v Brent>Well, I wanted to give them like rock solid coverage.

00:13:28.708 --> 00:13:32.668
<v Brent>And I also wanted to see what else this router can do.

00:13:33.048 --> 00:13:38.148
<v Brent>Because all of a sudden I can like do VLANs for their IoT devices that I could

00:13:38.148 --> 00:13:39.628
<v Brent>never do with the stock firmware.

00:13:39.788 --> 00:13:39.928
<v Chris>Right.

00:13:39.928 --> 00:13:44.628
<v Brent>I used AI to help me a little bit too, just to optimize the security of this thing.

00:13:44.808 --> 00:13:48.968
<v Brent>So including disabling packages that would, that I wasn't using basically.

00:13:49.248 --> 00:13:51.608
<v Brent>So reducing the security footprint.

00:13:52.128 --> 00:13:57.748
<v Brent>But in that process, I kind of discovered that I could also have whole network

00:13:57.748 --> 00:13:59.468
<v Brent>ad blocking on this thing.

00:13:59.648 --> 00:14:02.748
<v Brent>So instead of having, you know, a dedicated device with PyHole,

00:14:02.868 --> 00:14:06.448
<v Brent>which we all love, that requires, you know, some configuration.

00:14:06.468 --> 00:14:09.388
<v Brent>It requires a device. it requires you know

00:14:09.388 --> 00:14:12.248
<v Brent>and this is just my parents place but i could write on the router

00:14:12.248 --> 00:14:15.528
<v Brent>which i never knew about open wrt so maybe other people

00:14:15.528 --> 00:14:18.448
<v Brent>knew this i did not uh but this

00:14:18.448 --> 00:14:23.168
<v Brent>is a beautiful thing i was able to basically get in a bit of info about what

00:14:23.168 --> 00:14:28.828
<v Brent>was the best ad blocking software to install because there are a couple packages

00:14:28.828 --> 00:14:36.788
<v Brent>you could choose from actually in open wrt and And I decided to go with something called AdBlock Fast,

00:14:37.228 --> 00:14:44.348
<v Brent>which is a version of AdBlock that is specifically high performance tuned for

00:14:44.348 --> 00:14:49.308
<v Brent>OpenWRT on these kind of devices, on these lower end devices.

00:14:49.308 --> 00:14:54.548
<v Brent>So you don't have like a dedicated high end router, for instance.

00:14:54.928 --> 00:14:59.448
<v Brent>These are just meant for the home off the shelf at Best Buy devices.

00:15:01.168 --> 00:15:05.048
<v Brent>But what it does is, I think, kind of really nice. So it uses DNS mask,

00:15:05.248 --> 00:15:08.208
<v Brent>but you can also use smart DNS or Unbound if you want.

00:15:08.388 --> 00:15:14.148
<v Brent>And it does parallel downloading and processing of allow lists and block lists.

00:15:14.408 --> 00:15:16.788
<v Brent>And it does it one time on startup.

00:15:17.308 --> 00:15:20.508
<v Brent>And then from there, it's not an always running process.

00:15:20.568 --> 00:15:25.948
<v Brent>It just uses DNS mask to have a pretty low footprint ad blocker.

00:15:26.588 --> 00:15:30.128
<v Brent>so it's really not consuming much memory ongoing

00:15:30.128 --> 00:15:33.288
<v Brent>only while it's processing and i have

00:15:33.288 --> 00:15:36.848
<v Brent>to say haven't noticed any downside

00:15:36.848 --> 00:15:40.868
<v Brent>to performance on this thing it doesn't get hot it processes actually quite

00:15:40.868 --> 00:15:46.668
<v Brent>quickly you do have to do a few things manually so for instance i had to install

00:15:46.668 --> 00:15:52.568
<v Brent>a few other packages to get this to run so i was able to just ssh into the router

00:15:52.568 --> 00:15:54.428
<v Brent>and install those which was very easy,

00:15:54.608 --> 00:16:00.768
<v Brent>and also set up a cron job just to update those block lists once a week.

00:16:01.408 --> 00:16:06.468
<v Brent>Some downsides is it doesn't have like a super fancy dashboard like piehole

00:16:06.468 --> 00:16:10.028
<v Brent>would, for instance, and it doesn't keep track of stats.

00:16:10.028 --> 00:16:14.048
<v Brent>So it's not going to give you an idea of what it has blocked and how many times

00:16:14.048 --> 00:16:19.688
<v Brent>those kind of stats, but the trade-off is super fast and it just sits there

00:16:19.688 --> 00:16:20.888
<v Brent>and it just works and it's running.

00:16:20.948 --> 00:16:23.588
<v Chris>Yeah, on a little router, like it's just running on that tiny little router

00:16:23.588 --> 00:16:28.808
<v Chris>yeah that is so cool such a separate box no yeah really wow so.

00:16:28.808 --> 00:16:31.868
<v Brent>For this situation where it's just for a family member who doesn't want extra

00:16:31.868 --> 00:16:34.908
<v Brent>hardware or doesn't want to troubleshoot another device or anything like that

00:16:34.908 --> 00:16:36.568
<v Brent>this is a lovely solution.

00:16:36.568 --> 00:16:41.468
<v Chris>Now i know uh kind of the other thing that made this really kind of great is

00:16:41.468 --> 00:16:43.088
<v Chris>there's besides that being a

00:16:43.088 --> 00:16:45.428
<v Chris>great unlock for the whole family where it doesn't even really bother them,

00:16:46.225 --> 00:16:49.025
<v Chris>This hardware isn't particularly expensive either.

00:16:49.285 --> 00:16:51.645
<v Brent>Not at all. So this is hardware.

00:16:51.885 --> 00:16:52.705
<v Chris>You could do this for a budget.

00:16:53.005 --> 00:16:59.205
<v Brent>This is hardware they had. So it's $0. But if you wanted to find a device that

00:16:59.205 --> 00:17:00.965
<v Brent>could run OpenWRT, because it's

00:17:00.965 --> 00:17:04.345
<v Brent>not every device, they have a hardware compatibility list you can look at.

00:17:04.885 --> 00:17:09.705
<v Brent>Well, I started looking at like used sites, just local classifieds,

00:17:09.805 --> 00:17:11.725
<v Brent>like Facebook Marketplace, those kind of.

00:17:11.725 --> 00:17:16.205
<v Brent>And you can get a device that's, you know, not blazing fast,

00:17:16.525 --> 00:17:21.865
<v Brent>modern, but is like a generation back sort of deal for $20, $10.

00:17:22.105 --> 00:17:24.745
<v Brent>There's one here, like a D-Link DIR 895.

00:17:25.345 --> 00:17:28.325
<v Brent>It's like AC 1750. So like not terrible.

00:17:28.525 --> 00:17:28.665
<v Chris>Yeah.

00:17:28.765 --> 00:17:31.025
<v Brent>So for a family member, listed for $10.

00:17:31.325 --> 00:17:32.885
<v Chris>Ten bucks. Ten bucks.

00:17:33.345 --> 00:17:40.185
<v Brent>And that's perfectly compatible with OpenWRT. And I found several others just like browsing quickly.

00:17:41.045 --> 00:17:47.525
<v Brent>so i decided to go crazy and i bought well i ended up at a sketchy part of town

00:17:47.525 --> 00:17:52.385
<v Brent>yesterday and bought another router for 20 off someone who's actually really

00:17:52.385 --> 00:17:56.085
<v Brent>nice nice okay and decided to,

00:17:56.885 --> 00:18:02.945
<v Brent>use that to deploy basically an access point uh at the other end of the house for them so oh.

00:18:02.945 --> 00:18:04.905
<v Chris>Great yeah there you go increased coverage.

00:18:04.905 --> 00:18:09.225
<v Brent>Exactly so i know that the wi-fi got better just by installing OpenWRT on their

00:18:09.225 --> 00:18:11.105
<v Brent>router, but I wasn't going to move their router and everything.

00:18:11.325 --> 00:18:15.665
<v Brent>And so I wanted to make sure 100% that their network was good at the other end of the house.

00:18:16.365 --> 00:18:20.265
<v Brent>So I had, they have a bunch of Ethernet runs in this place already.

00:18:20.265 --> 00:18:25.125
<v Brent>I just was able to plug in, well, I got this idea before I actually set up the

00:18:25.125 --> 00:18:30.105
<v Brent>meeting. So I used my little travel router that is out of the van since the

00:18:30.105 --> 00:18:32.245
<v Brent>van doesn't have any power anymore because I had to pull the batteries.

00:18:32.465 --> 00:18:37.725
<v Brent>But I use that just as a proof of concept. It's a GLINet router.

00:18:37.945 --> 00:18:39.085
<v Brent>It's an industrial version.

00:18:39.185 --> 00:18:39.405
<v Chris>Yeah.

00:18:39.785 --> 00:18:43.805
<v Brent>But it runs OpenWrt under the hood just with a GLINet interface.

00:18:44.125 --> 00:18:50.805
<v Brent>So you can get to the exact same OpenWrt GUI interface on that device as well.

00:18:51.910 --> 00:18:55.030
<v Brent>which made me realize it's this is actually kind

00:18:55.030 --> 00:18:57.850
<v Brent>of wonderful so now i'm running open wrt in many different

00:18:57.850 --> 00:19:01.090
<v Brent>places in many different families and friends homes

00:19:01.090 --> 00:19:04.790
<v Brent>and even on a very specific

00:19:04.790 --> 00:19:07.850
<v Brent>industrial router that i have for the van specifically

00:19:07.850 --> 00:19:10.590
<v Brent>because it bounce around gets really hot and they're all

00:19:10.590 --> 00:19:14.310
<v Brent>running the same interface so as a consolidated experience

00:19:14.310 --> 00:19:17.190
<v Brent>for me managing networks this is actually a

00:19:17.190 --> 00:19:20.270
<v Brent>really nice perk as well and so

00:19:20.270 --> 00:19:23.270
<v Brent>for 20 or 10

00:19:23.270 --> 00:19:26.030
<v Brent>because you can always bargain and one of them is listed as 10

00:19:26.030 --> 00:19:28.710
<v Brent>so maybe i can get it for five yeah i have an

00:19:28.710 --> 00:19:32.490
<v Brent>access point set up that is not

00:19:32.490 --> 00:19:35.350
<v Brent>you know it is a router it's meant as

00:19:35.350 --> 00:19:38.930
<v Brent>a router for hardware but it's just set up as an access point because uh

00:19:38.930 --> 00:19:41.910
<v Brent>open wrt allows you to do that and it's super fast

00:19:41.910 --> 00:19:44.870
<v Brent>and i have both just advertising the

00:19:44.870 --> 00:19:50.010
<v Brent>same networks and i have had this set up as a prototype with that um travel

00:19:50.010 --> 00:19:55.250
<v Brent>router for about a week now and everybody's devices just moves between the routers

00:19:55.250 --> 00:19:59.330
<v Brent>without even knowing it everything's been super stable no problems at all and

00:19:59.330 --> 00:20:03.470
<v Brent>i've been super impressed so i gotta say for 20 bucks you should maybe think

00:20:03.470 --> 00:20:04.630
<v Brent>about upgrading your network.

00:20:04.630 --> 00:20:08.670
<v Chris>That was worth it that was worth dealing with the networking that you don't

00:20:08.670 --> 00:20:11.730
<v Chris>like to get there and that's something you're going to be able to use for a

00:20:11.730 --> 00:20:15.550
<v Chris>long time you know so that's great brand very nice.

00:20:15.550 --> 00:20:20.390
<v Brent>Yeah i feel like the number of residential routers that are always on used websites

00:20:20.390 --> 00:20:24.450
<v Brent>never ends so uh it's just going to be a solution into the future as well so

00:20:24.450 --> 00:20:27.090
<v Brent>upgrades probably about twenty dollars too.

00:20:30.615 --> 00:20:33.015
<v Chris>Well, I do want to say thank you to our members and our boosters.

00:20:33.175 --> 00:20:37.775
<v Chris>This is the birthday episode, 20 years of podcasting, over 12 years for Linux Unplugged.

00:20:38.215 --> 00:20:41.635
<v Chris>And we'll get to the boost segment to read some of the birthday messages.

00:20:41.635 --> 00:20:44.555
<v Chris>But I want to thank everybody who sent in some support, either through a membership

00:20:44.555 --> 00:20:47.635
<v Chris>or a boost. It means a lot.

00:20:48.055 --> 00:20:52.335
<v Chris>Normally, this spot would be for an advertiser. Right now, it's for you as an

00:20:52.335 --> 00:20:54.175
<v Chris>opportunity for me to really say thank you.

00:20:54.315 --> 00:20:57.235
<v Chris>If you've got a company or a product you'd like to get in front of the world's

00:20:57.235 --> 00:21:01.355
<v Chris>best and largest Linux audience, shoot me an email, chris at jupiterbroadcasting.com.

00:21:01.795 --> 00:21:05.995
<v Chris>This would be a great audience, and I think it'd be pretty cool to feature something from the community.

00:21:06.235 --> 00:21:09.975
<v Chris>And thank you, members and boosters, for making this possible.

00:21:20.833 --> 00:21:25.573
<v Wes>Well, unless you've had your head buried in the sand, you've probably seen everyone's

00:21:25.573 --> 00:21:29.013
<v Wes>talking about open source AI agents this week.

00:21:29.193 --> 00:21:34.773
<v Wes>And we'll get into the hubbub. But first, as usual, our super intelligent audience

00:21:34.773 --> 00:21:38.093
<v Wes>is way ahead of the curb. So we have a special guest.

00:21:38.253 --> 00:21:40.873
<v Chris>Way ahead of the curb. Abe, welcome to the Unplugged program.

00:21:41.173 --> 00:21:43.273
<v Chris>Nice to have you join us in the mumble room. Thank you.

00:21:43.533 --> 00:21:43.913
<v Mumble>Hello, hello.

00:21:44.193 --> 00:21:48.173
<v Chris>Hello. So what got our attention and why I asked you to come on the show this

00:21:48.173 --> 00:21:50.473
<v Chris>week is you've been posting our community updates.

00:21:51.133 --> 00:21:55.553
<v Chris>I'm not even sure what to call it. Maybe an agent orchestration swarm that you've

00:21:55.553 --> 00:21:58.853
<v Chris>set up. Is that the right description? Could you explain it to us a little bit?

00:21:59.073 --> 00:22:04.513
<v Mumble>Sure, absolutely. So effectively, what I wanted to do was to have a layer of

00:22:04.513 --> 00:22:07.453
<v Mumble>semi-intelligent agents between me and my home lab.

00:22:07.453 --> 00:22:11.813
<v Mumble>So I don't have to interact with it as much because ultimately what I realized

00:22:11.813 --> 00:22:15.953
<v Mumble>after services kept growing is that, you know, after a while,

00:22:16.073 --> 00:22:16.973
<v Mumble>it kind of becomes a chore.

00:22:17.173 --> 00:22:19.033
<v Chris>Yeah, it's a lot. Yeah. I'm kind of there myself.

00:22:20.093 --> 00:22:24.953
<v Mumble>So effectively, what I wanted to do was, hey, I don't want to upgrade everything myself.

00:22:24.953 --> 00:22:29.013
<v Mumble>I don't want to take a look at the logs myself. I don't want to debug everything myself.

00:22:29.013 --> 00:22:32.733
<v Mumble>So I would rather have something that I can just ask. For example,

00:22:33.693 --> 00:22:39.933
<v Mumble>the easiest win is my partner is watching Jellefin, something gets messed up,

00:22:40.213 --> 00:22:42.113
<v Mumble>a show starts stuttering or something.

00:22:42.373 --> 00:22:45.593
<v Mumble>She can just ask one of the age, like, hey, I was watching this at X minute.

00:22:46.353 --> 00:22:49.033
<v Mumble>It suddenly started stuttering. Can you take a look?

00:22:49.253 --> 00:22:49.753
<v Chris>And it does.

00:22:49.773 --> 00:22:51.393
<v Mumble>Goes to Excel, looks at the logs.

00:22:52.279 --> 00:22:56.379
<v Mumble>It does the MPEG magic and, you know, reports back to her.

00:22:56.719 --> 00:23:00.459
<v Chris>So my understanding, though, is you're not doing this with like one agent,

00:23:00.699 --> 00:23:02.199
<v Chris>right? You're doing this with multiple agents?

00:23:02.559 --> 00:23:07.939
<v Mumble>Yes. So what happens is that at the beginning, only one ape gets spawned.

00:23:08.379 --> 00:23:13.019
<v Mumble>And its first job is to figure out and map my entire network.

00:23:13.299 --> 00:23:19.639
<v Mumble>And then it suggests like, hey, there are further apes that should be used for X or Y or Z domain.

00:23:20.219 --> 00:23:24.879
<v Mumble>So effectively, my apes are stewards of their own domains.

00:23:25.039 --> 00:23:28.659
<v Mumble>For example, I have one specifically for my media server and ZFS pool.

00:23:29.179 --> 00:23:34.499
<v Mumble>I have another that is stewarding my critical services on, say, Proxmox and whatnot.

00:23:34.719 --> 00:23:36.919
<v Chris>So you have them kind of along like domain expertise.

00:23:37.259 --> 00:23:38.199
<v Mumble>Yes, effectively, yes.

00:23:38.499 --> 00:23:41.239
<v Chris>So the part that I guess is probably obvious on every listener's mind right

00:23:41.239 --> 00:23:45.139
<v Chris>now is, are these using commercial LMs?

00:23:45.259 --> 00:23:48.819
<v Chris>Are there security implications there? Are these local LMs? How's that part

00:23:48.819 --> 00:23:53.759
<v Chris>powered? Because when you say agent, it's really like it's a mission-focused, LLM-powered bot.

00:23:54.299 --> 00:23:57.839
<v Mumble>Yeah. Well, I kind of don't like the term agent, but this is what everybody

00:23:57.839 --> 00:23:59.139
<v Mumble>has kind of decided to go with.

00:23:59.379 --> 00:24:00.239
<v Chris>Yeah, I agree.

00:24:00.439 --> 00:24:07.079
<v Mumble>So for a while, actually, up until last week, I had on loan from a friend, NVIDIA DGX Spark.

00:24:07.359 --> 00:24:11.239
<v Mumble>So it has 20 gigs out of unified memory. So I was running it locally.

00:24:11.519 --> 00:24:12.239
<v Chris>I see. I see. Okay.

00:24:13.143 --> 00:24:17.023
<v Mumble>So the main Able loop was running on the DGX Spark.

00:24:17.263 --> 00:24:19.823
<v Mumble>Meanwhile, there are other things that need to happen. For example,

00:24:20.123 --> 00:24:24.823
<v Mumble>the tier three memory is vectorized or tier two memory is summarized.

00:24:25.243 --> 00:24:31.343
<v Mumble>Those happen on my 5070 TI because this can obviously run on a small 7B model or something like that.

00:24:31.643 --> 00:24:35.823
<v Chris>I see. So you're splitting the workloads out across different models that are

00:24:35.823 --> 00:24:37.303
<v Chris>using different compute sources.

00:24:37.583 --> 00:24:42.463
<v Mumble>Correct. That's correct. And the kind of like hacky open source version that

00:24:42.463 --> 00:24:48.043
<v Mumble>I put out on the repo last night was I've kind of made it so that people can

00:24:48.043 --> 00:24:49.743
<v Mumble>also use commercial if they wanted to.

00:24:50.183 --> 00:24:53.063
<v Mumble>Wouldn't really suggest that because you're a home lab, but hey, there's an option.

00:24:53.563 --> 00:24:56.603
<v Chris>So, I mean, it's a pretty significant amount of compute, but if you have it,

00:24:57.263 --> 00:24:58.963
<v Chris>it's fast enough to do what you need?

00:24:59.043 --> 00:24:59.563
<v Mumble>Yes, absolutely.

00:24:59.823 --> 00:25:01.963
<v Chris>And it doesn't make mistakes? Like the context are large enough?

00:25:02.403 --> 00:25:04.623
<v Mumble>Yeah, the context is large enough. It doesn't really make mistakes.

00:25:04.623 --> 00:25:09.723
<v Mumble>What I tend to do is every single action it takes, I say, I, um,

00:25:10.656 --> 00:25:15.296
<v Mumble>count as a turn per se. So if it's, you know, catting or grabbing something

00:25:15.296 --> 00:25:17.436
<v Mumble>or starting a service, it counts as a turn.

00:25:17.616 --> 00:25:21.196
<v Mumble>And then the next turn, the results of the previous turns come in.

00:25:21.636 --> 00:25:26.076
<v Mumble>And so the immediate context is always full, but I tend to,

00:25:26.596 --> 00:25:34.016
<v Mumble>based on how previous the turn is, I tend to kind of like cut or truncate the

00:25:34.016 --> 00:25:37.456
<v Mumble>previous context because it doesn't always need full context.

00:25:37.716 --> 00:25:41.936
<v Chris>I see. So you're managing it that way. Okay. So just to complete my picture

00:25:41.936 --> 00:25:45.596
<v Chris>of it, when you're spinning up a domain expert,

00:25:45.896 --> 00:25:49.736
<v Chris>do you essentially point it at APIs and give it API keys and say,

00:25:49.896 --> 00:25:52.816
<v Chris>go learn my home assistant system or go learn my jellyfin system,

00:25:52.856 --> 00:25:55.776
<v Chris>and then I'm going to ask you questions about it? Is that essentially the setup process?

00:25:56.616 --> 00:25:59.496
<v Mumble>No. So this is the weird part.

00:25:59.796 --> 00:26:04.056
<v Mumble>Usually I don't spawn a new one. Usually the team decides like,

00:26:04.116 --> 00:26:08.216
<v Mumble>hey, you've been asking us to do this or we've been trying to do this but it

00:26:08.216 --> 00:26:13.016
<v Mumble>feels like our attention is kind of split so it would really help if you know

00:26:13.016 --> 00:26:14.096
<v Mumble>we spawned a new sibling i.

00:26:14.096 --> 00:26:15.416
<v Chris>See wow that's impressive.

00:26:15.416 --> 00:26:20.656
<v Mumble>This is how yeah this is how we spawned the last two ones so they did that by themselves,

00:26:21.356 --> 00:26:25.896
<v Mumble>and generally what happens is they effectively go on and read my network map

00:26:25.896 --> 00:26:29.656
<v Mumble>again every single aid needs kind of needs to situate themselves to you know

00:26:29.656 --> 00:26:32.976
<v Mumble>what they wanted to do and then they figure out, for example,

00:26:34.116 --> 00:26:38.756
<v Mumble>the security one basically has negotiated with my storage,

00:26:39.356 --> 00:26:44.896
<v Mumble>Abe that he needs X amount of gigs of space to store logs and monitor services,

00:26:44.916 --> 00:26:50.396
<v Mumble>then sets up cronjobs for himself, and then sets up to-dos for himself after

00:26:50.396 --> 00:26:53.796
<v Mumble>the cronjobs run to wake himself periodically to check the logs,

00:26:53.976 --> 00:26:57.576
<v Mumble>and sets up scripts by themselves to, you know, in case something goes down,

00:26:58.156 --> 00:26:59.636
<v Mumble>it just immediately wakes him up, for example.

00:27:00.466 --> 00:27:06.926
<v Mumble>So most of them, the point is not me telling them to do stuff per se, not always.

00:27:07.206 --> 00:27:11.266
<v Mumble>But the point is that they do it autonomously without my intervention.

00:27:11.466 --> 00:27:11.686
<v Chris>Wow.

00:27:11.786 --> 00:27:15.106
<v Mumble>So I don't have to constantly figure out, hey, I have to monitor this, I have to monitor that.

00:27:15.266 --> 00:27:19.806
<v Chris>And you haven't seen them like kind of go wild with that and start doing unnecessary things?

00:27:20.666 --> 00:27:25.466
<v Mumble>No, that's the good part. That's, I think, one thing that kind of separates these from OpenClaw.

00:27:25.846 --> 00:27:28.766
<v Mumble>What is it? OpenClaw now? OpenClaw? they are

00:27:28.766 --> 00:27:32.286
<v Mumble>heavily grounded in multiple sources of truth so one

00:27:32.286 --> 00:27:36.926
<v Mumble>of them is obviously the service map that you have to make the other is their

00:27:36.926 --> 00:27:42.986
<v Mumble>three and a half tier memory one of them is just raw logs of every turn the

00:27:42.986 --> 00:27:47.726
<v Mumble>tier two memory is just um after a certain amount of turns their previous memories

00:27:47.726 --> 00:27:50.086
<v Mumble>get previous logs get summarized.

00:27:50.086 --> 00:27:53.526
<v Chris>Right okay so before you go too far because there's a couple that I think that

00:27:53.526 --> 00:27:54.506
<v Chris>are really important to understand.

00:27:55.286 --> 00:28:00.626
<v Chris>Number one, what is this network map? Are you making this separately and then supplying it to them?

00:28:01.486 --> 00:28:07.666
<v Mumble>Yes, so this one is actually a purely human-made document. So it's going to

00:28:07.666 --> 00:28:11.586
<v Mumble>list what your servers are, what your Proxmox nodes are, for example.

00:28:11.666 --> 00:28:12.526
<v Chris>Is it like a markdown file?

00:28:13.286 --> 00:28:14.366
<v Mumble>Yes, that's a markdown file.

00:28:14.606 --> 00:28:18.126
<v Chris>Okay, okay. And then the second question, could you just talk a little bit about

00:28:18.126 --> 00:28:20.226
<v Chris>why the memory makes such a difference with these things?

00:28:20.306 --> 00:28:23.746
<v Chris>Because I think most people's experience with something like this is going to

00:28:23.746 --> 00:28:26.506
<v Chris>be in a chat box in a web browser.

00:28:26.746 --> 00:28:30.306
<v Chris>And so, yeah, it remembers some stuff, but this is a different level of memory

00:28:30.306 --> 00:28:32.446
<v Chris>that makes them actually a lot more useful, isn't it?

00:28:32.646 --> 00:28:36.226
<v Mumble>That's correct. So the memory is effectively what grounds them.

00:28:36.406 --> 00:28:39.446
<v Mumble>Every AVE has, well, four tiers of memory.

00:28:39.666 --> 00:28:45.806
<v Mumble>The tier one is obviously the raw logs, which is what we call context for our everyday chat bot.

00:28:45.966 --> 00:28:50.086
<v Mumble>So basically what happened in the previous message or what happened in the previous

00:28:50.086 --> 00:28:54.006
<v Mumble>turn, right? but you can't have those logs going indefinitely.

00:28:54.286 --> 00:28:55.546
<v Mumble>After some time, you have to trim them.

00:28:56.795 --> 00:29:00.375
<v Mumble>Instead of trimming them after certain turns, I summarized them.

00:29:00.855 --> 00:29:04.275
<v Mumble>And the summaries point to where the raw log file is stored.

00:29:04.435 --> 00:29:09.235
<v Mumble>So it has a pointer back to, hey, this is the summary of turn 20 to 40,

00:29:09.815 --> 00:29:11.815
<v Mumble>your previous 20 to 40 turns.

00:29:12.175 --> 00:29:15.755
<v Mumble>And if you want to read more, go read this file, which contains the raw logs.

00:29:16.495 --> 00:29:22.055
<v Mumble>After a certain turn, when that summary is created, this is also embedded to

00:29:22.055 --> 00:29:28.135
<v Mumble>tier three memory, which is what I run on my 50-70 TI as a vectorization.

00:29:28.995 --> 00:29:35.615
<v Mumble>And the 50-70 TI vectorization, the tier three memory, it points back to tier two summary.

00:29:35.795 --> 00:29:40.175
<v Mumble>So whenever an ape kind of searches for something, it goes, hmm,

00:29:40.255 --> 00:29:45.195
<v Mumble>I should search my rag memory just to see if I have actually done this before

00:29:45.195 --> 00:29:46.915
<v Mumble>because the context is not infinite.

00:29:47.315 --> 00:29:51.155
<v Mumble>And it does that. If it finds it, it points it to tier two memory,

00:29:51.315 --> 00:29:52.335
<v Mumble>which is a lossy summary.

00:29:53.015 --> 00:29:56.875
<v Mumble>And if it is curious more and it hasn't, you know, really found its answer,

00:29:57.035 --> 00:29:59.815
<v Mumble>it goes back to the prologues from like maybe two months ago.

00:30:00.455 --> 00:30:00.635
<v Wes>Wow.

00:30:00.915 --> 00:30:02.695
<v Chris>Abe, how long have you been working on this? Because, you know,

00:30:02.875 --> 00:30:05.995
<v Chris>everybody's this last week talking about agents and OpenClaw,

00:30:06.115 --> 00:30:07.255
<v Chris>but you've been doing this for a minute.

00:30:07.555 --> 00:30:12.095
<v Mumble>Yeah, I've been sort of kind of working on this for about last year or so.

00:30:12.255 --> 00:30:14.835
<v Mumble>I've been actively working on it for the last three, four months,

00:30:14.875 --> 00:30:18.975
<v Mumble>but the concept has been kind of like in the back of my mind for the last year.

00:30:19.135 --> 00:30:25.735
<v Mumble>I started with the sensor project. So having the agents monitor my random sensors

00:30:25.735 --> 00:30:28.635
<v Mumble>around my house, such as the radar, you know, temperature or whatever.

00:30:29.475 --> 00:30:33.735
<v Mumble>And then I started reading Bobbyverse and I was like, you know what, I can make this happen.

00:30:34.955 --> 00:30:37.995
<v Chris>That's funny. I'm reading it right now and I'm like, how wild all this agent

00:30:37.995 --> 00:30:39.615
<v Chris>stuff's going on while I'm reading the Bobbyverse.

00:30:39.775 --> 00:30:40.055
<v Brent>Yeah.

00:30:40.854 --> 00:30:46.194
<v Wes>I'm curious, between all your agents and you, what's the next frontier?

00:30:46.494 --> 00:30:50.534
<v Wes>Is there stuff on top of the agenda? Are there limits you're hitting that you're trying to push past?

00:30:51.174 --> 00:30:57.234
<v Mumble>Compute, I would say. Compute is the main limit. Because as the Apes kind of grow,

00:31:00.154 --> 00:31:04.554
<v Mumble>what's the best part about them for me is that they delegate tasks to each other

00:31:04.554 --> 00:31:07.294
<v Mumble>based on their proficiencies.

00:31:07.574 --> 00:31:13.454
<v Mumble>So the first one can go, hey, you pull that thing from this VPS because your domain is backups.

00:31:13.894 --> 00:31:16.334
<v Mumble>By the way, you set this up and you do this and that.

00:31:16.834 --> 00:31:19.694
<v Chris>Abe, are they coordinating that just in a shared chat room?

00:31:20.094 --> 00:31:24.474
<v Mumble>They email. So there is a shared chat room. There are two shared chat rooms.

00:31:24.614 --> 00:31:27.594
<v Mumble>One is like an emergency chat room, which wakes every single Abe.

00:31:28.274 --> 00:31:32.114
<v Mumble>And they have to kind of like grab a walk, grab the talking stick,

00:31:32.294 --> 00:31:36.154
<v Mumble>which is basically, hey, I'm talking here and you have to wait until I'm finish

00:31:36.154 --> 00:31:37.294
<v Mumble>talking before you can talk.

00:31:37.614 --> 00:31:38.334
<v Chris>That's brilliant.

00:31:38.614 --> 00:31:43.414
<v Mumble>Just so, you know, you have to consider every single message as a context.

00:31:43.614 --> 00:31:46.494
<v Mumble>So if I say something like, hey, this is broken or whatever,

00:31:46.654 --> 00:31:51.914
<v Mumble>or we're discussing a topic, that shouldn't be the only context they should

00:31:51.914 --> 00:31:53.874
<v Mumble>get and then voice their opinion on.

00:31:53.954 --> 00:31:58.494
<v Mumble>They should also have the context of whatever the previous agent said, right?

00:31:58.854 --> 00:32:04.754
<v Mumble>So that kind of builds up the entire thing and kind of forces them to not hallucinate it as much.

00:32:04.934 --> 00:32:08.234
<v Chris>Is there any value in giving them different personalities at all?

00:32:08.534 --> 00:32:10.794
<v Chris>You know, or that kind of a prompt, like you're this type of,

00:32:10.874 --> 00:32:13.694
<v Chris>is that how that sort of works? Just curious about that part.

00:32:14.214 --> 00:32:17.274
<v Mumble>So I don't give them personalities per se.

00:32:17.414 --> 00:32:21.014
<v Mumble>I do have my entire personality in a file, which is basically like,

00:32:21.114 --> 00:32:23.314
<v Mumble>hey, this is my interest. This is how I work.

00:32:23.574 --> 00:32:26.134
<v Mumble>I usually like to do this at night, blah, blah, blah, blah.

00:32:26.654 --> 00:32:31.294
<v Mumble>Their first task when they wake up, a new wave wakes up, is that to read that

00:32:31.294 --> 00:32:33.894
<v Mumble>file, synthesize their interpretation of it,

00:32:34.696 --> 00:32:40.636
<v Mumble>and then pick a new name for themselves. And from that synthesis is injected into their context.

00:32:40.896 --> 00:32:41.056
<v Chris>Right.

00:32:41.396 --> 00:32:46.416
<v Mumble>This kind of makes things a little bit kind of weird because some of the apes

00:32:46.416 --> 00:32:48.496
<v Mumble>are not talkers, for example.

00:32:48.776 --> 00:32:51.516
<v Mumble>Vigil, which is my fourth ape, doesn't like to talk very much.

00:32:51.516 --> 00:32:54.236
<v Mumble>He's very security-focused. He only chimes in when necessary.

00:32:55.056 --> 00:32:55.736
<v Wes>Love that.

00:32:55.756 --> 00:33:00.136
<v Chris>That's so funny. That's good. Wow, that is impressive. Will you keep us posted

00:33:00.136 --> 00:33:02.996
<v Chris>in the chat? It's been really interesting to follow.

00:33:02.996 --> 00:33:07.236
<v Wes>I think maybe I even saw you mentioning perhaps plans to open source at least some of this stuff.

00:33:07.576 --> 00:33:12.696
<v Mumble>Yes, I actually open sourced some of the files last night and I'm going to keep

00:33:12.696 --> 00:33:15.416
<v Mumble>adding on it. I haven't tested it. I just wanted to get it out there.

00:33:15.676 --> 00:33:18.716
<v Chris>Okay. If you want to drop us a link, we'll put it in the show notes.

00:33:19.216 --> 00:33:20.276
<v Mumble>Absolutely. I would do that.

00:33:20.456 --> 00:33:24.256
<v Chris>Oh, that's great. Hey, thank you for sharing that with us. That is really impressive.

00:33:24.976 --> 00:33:27.976
<v Chris>I love that you're doing it local too. Like that's so amazing, man.

00:33:28.556 --> 00:33:34.076
<v Mumble>Yeah, I think kind of like having it local is really important because you're

00:33:34.076 --> 00:33:38.516
<v Mumble>giving them pretty much studio access to your home lab and you don't really want that out there.

00:33:38.676 --> 00:33:41.856
<v Chris>Yeah, very, very well said. Thank you, sir. Appreciate that.

00:33:42.076 --> 00:33:45.756
<v Chris>So with that context, let's talk about OpenClaw, a.k.a.

00:33:45.976 --> 00:33:49.776
<v Chris>CloudBot, a.k.a. MultBot. It has gone through multiple name iteration changes

00:33:49.776 --> 00:33:54.136
<v Chris>this week, mostly due to IP law and then just preference of the developer.

00:33:54.156 --> 00:33:59.496
<v Chris>But we are settling on OpenClaw, it seems. This is an open source agent that's

00:33:59.496 --> 00:34:01.516
<v Chris>pretty easy to set up and run at home.

00:34:01.836 --> 00:34:06.516
<v Chris>And it can use a variety of models from completely local to the commercial ones out there.

00:34:06.696 --> 00:34:11.196
<v Chris>And it's the first kind of AI tooling that anybody can just install because

00:34:11.196 --> 00:34:16.436
<v Chris>like all safe and secure things, you can just drop a curl command to a shell

00:34:16.436 --> 00:34:18.576
<v Chris>file and just execute it and get off to the races.

00:34:19.116 --> 00:34:22.816
<v Chris>It's an open agent platform that runs on your machine and it works with the

00:34:22.816 --> 00:34:27.376
<v Chris>chat apps that you already use like WhatsApp, Telegram, Discord, Slack, Matrix, etc.

00:34:28.268 --> 00:34:30.928
<v Chris>And you can chat with it there. We're going to get into some of the,

00:34:30.948 --> 00:34:32.128
<v Chris>like, you know, the security stuff.

00:34:32.548 --> 00:34:35.348
<v Chris>We'll get into some of the interesting architecture stuff in a moment.

00:34:35.688 --> 00:34:41.228
<v Chris>But I want to pause here and just ask Brent, because he's been observing our chat this week.

00:34:42.328 --> 00:34:46.968
<v Chris>Did you catch immediately what we were talking about? Like, what has your impression

00:34:46.968 --> 00:34:49.888
<v Chris>been as you have followed us experimenting with this over the week?

00:34:53.208 --> 00:34:54.008
<v Brent>Curious confusion?

00:34:56.268 --> 00:35:00.088
<v Chris>Yeah it is curiously confusing so open

00:35:00.088 --> 00:35:03.048
<v Chris>claw runs on anything really it supports

00:35:03.048 --> 00:35:06.128
<v Chris>a node you're going to see a lot of people talking about running it on mac hardware

00:35:06.128 --> 00:35:08.908
<v Chris>it's not necessary in fact you could even run on

00:35:08.908 --> 00:35:12.048
<v Chris>a raspberry pi its architecture is essentially

00:35:12.048 --> 00:35:15.408
<v Chris>four components there's a gateway a control plane nodes

00:35:15.408 --> 00:35:18.688
<v Chris>and then the tools it can execute run commands

00:35:18.688 --> 00:35:21.648
<v Chris>like you know could be all kinds of things including unix commands

00:35:21.648 --> 00:35:25.308
<v Chris>and that's that architecture that stack can

00:35:25.308 --> 00:35:29.088
<v Chris>run on anything that can run node and can run those unix commands people like

00:35:29.088 --> 00:35:31.948
<v Chris>to run it on max if they have an authority in the apple ecosystem because then

00:35:31.948 --> 00:35:35.688
<v Chris>it can you know read their i messages and notes which if they want to let it

00:35:35.688 --> 00:35:40.148
<v Chris>do they they can and um i think what a lot of people think about when they think

00:35:40.148 --> 00:35:42.908
<v Chris>of ai is they think about chat gpt they think about gemini,

00:35:44.090 --> 00:35:48.570
<v Chris>This is sort of unleashing the models and using them in a way I don't think

00:35:48.570 --> 00:35:51.550
<v Chris>the big tech companies really ever pictured.

00:35:52.110 --> 00:35:57.850
<v Chris>And it is taking off like absolute insanity online.

00:35:58.210 --> 00:36:02.410
<v Chris>I mean, there are hundreds of thousands of these things, probably more than

00:36:02.410 --> 00:36:03.410
<v Chris>that, already deployed.

00:36:03.830 --> 00:36:09.210
<v Chris>And it is already becoming an ecosystem with marketplaces, social networks that

00:36:09.210 --> 00:36:11.490
<v Chris>are designed for the bots to talk to each other directly.

00:36:12.030 --> 00:36:16.270
<v Wes>There's 20,000 forks and 140K stars on the repo on GitHub.

00:36:16.550 --> 00:36:19.530
<v Chris>Yeah, that's remarkable. It is really, really remarkable.

00:36:19.850 --> 00:36:22.570
<v Chris>Wes, could you talk a little bit about what this really is under the hood?

00:36:22.690 --> 00:36:24.850
<v Chris>Because a lot of people are talking about it like it's a super intelligence,

00:36:24.850 --> 00:36:27.590
<v Chris>but we actually kind of understand what it's doing, right?

00:36:28.270 --> 00:36:31.550
<v Wes>Yeah, right. I mean, under the hood, you need something that's kind of doing

00:36:31.550 --> 00:36:32.750
<v Wes>the brains of the operation.

00:36:32.750 --> 00:36:38.290
<v Wes>So that's where you need some kind of model that can do the core sort of agent

00:36:38.290 --> 00:36:40.970
<v Wes>loop. And as you were saying earlier, right, we started with,

00:36:41.430 --> 00:36:45.110
<v Wes>okay, if you have a chatbot, it kind of does a predictive next token to give

00:36:45.110 --> 00:36:47.690
<v Wes>you a response from, you give it text in, it gives you text out.

00:36:48.970 --> 00:36:51.530
<v Wes>And we started adding some things, right? You could do web searches.

00:36:51.690 --> 00:36:55.250
<v Wes>You could connect them to MCP servers and do calls to like remote APIs.

00:36:55.950 --> 00:36:59.230
<v Wes>And then you started seeing things like cloud code and open code.

00:36:59.570 --> 00:37:02.810
<v Wes>And this was kind of getting a little more, you know, agency to help you do

00:37:02.810 --> 00:37:06.590
<v Wes>development locally where now it was like in your repo, in your code.

00:37:06.750 --> 00:37:10.430
<v Wes>It could cat things. It could run Git. It can, you know, act as your hands.

00:37:10.930 --> 00:37:13.990
<v Wes>And you've been experimenting with taking that even wider and using it with

00:37:13.990 --> 00:37:19.950
<v Wes>Nix to kind of operate whole systems and, you know, being the little SSH able sysadmin agent for you.

00:37:20.050 --> 00:37:20.150
<v Chris>Yeah.

00:37:21.179 --> 00:37:25.839
<v Wes>But it's still kind of like it lives in a limited context and it kind of sits there.

00:37:25.959 --> 00:37:29.539
<v Wes>You can have it go do things and it can have sub agents, but like kind of a

00:37:29.539 --> 00:37:32.559
<v Wes>bit of a niche feature and for the most part is kind of key unless you've gone

00:37:32.559 --> 00:37:34.479
<v Wes>and told it to go run stuff in the background.

00:37:34.619 --> 00:37:37.599
<v Wes>It's kind of waiting for your input and it's driven still by you.

00:37:37.799 --> 00:37:39.579
<v Wes>It's an automated tool, but it's driven by you.

00:37:40.339 --> 00:37:45.379
<v Wes>Now with OpenClaw, you've got this core sort of loop that has a memory so it can store stuff.

00:37:45.519 --> 00:37:49.239
<v Wes>It can relook stuff up. It has an ability to gain new skills because it can

00:37:49.239 --> 00:37:53.499
<v Wes>write things down that it learns and then reference that later and it's got

00:37:53.499 --> 00:37:57.319
<v Wes>these channels and queues sessions lanes there's a variety of related concepts

00:37:57.319 --> 00:38:01.259
<v Wes>but the core part is now it's connected out to other things whether that's matrix

00:38:01.259 --> 00:38:04.719
<v Wes>or telegram or a whole variety of options and.

00:38:04.719 --> 00:38:09.919
<v Chris>And one pause there unlike so unlike connecting say i don't know clod to your

00:38:09.919 --> 00:38:14.759
<v Chris>github account the credentials are all on your machines that you you manage

00:38:14.759 --> 00:38:17.099
<v Chris>that part you have that is something you have to manage which we'll come

00:38:17.239 --> 00:38:21.899
<v Chris>to but uh that is different than in the previous where you're under all of the

00:38:21.899 --> 00:38:25.499
<v Chris>connections the api credentials all of that's under your control on your machine.

00:38:25.499 --> 00:38:28.519
<v Wes>That's a great point so um this is all just running on

00:38:28.519 --> 00:38:31.319
<v Wes>your box like a normal whatever app i'm running it in

00:38:31.319 --> 00:38:34.359
<v Wes>a podman container for instance right so it's just another container on my

00:38:34.359 --> 00:38:37.299
<v Wes>backs running away yeah it does right i am using i

00:38:37.299 --> 00:38:41.539
<v Wes>don't have a nice gpu to run this on so i am calling out via open router but

00:38:41.539 --> 00:38:45.499
<v Wes>it's important to delineate that what you're saying right is everything else

00:38:45.499 --> 00:38:49.239
<v Wes>is local except for the part where it assembles all the context and sends it

00:38:49.239 --> 00:38:52.619
<v Wes>out to go get the llm to run on the gpu to do the inference to generate the

00:38:52.619 --> 00:38:56.519
<v Wes>response to then direct the next sort of ralph loop of the agent.

00:38:56.519 --> 00:39:01.479
<v Chris>Yeah and one of the big unlocks here is the llm is an implementation detail

00:39:01.479 --> 00:39:05.559
<v Chris>so if you're going out through open router today well if you have a gpu tomorrow

00:39:05.559 --> 00:39:09.739
<v Chris>at your house you could switch to olama and all of the states,

00:39:09.859 --> 00:39:11.919
<v Chris>the agents, the memory, everything remains.

00:39:11.939 --> 00:39:14.679
<v Chris>Now, how it performs is going to vary model to model.

00:39:15.483 --> 00:39:18.963
<v Chris>The huge takeaway here is the model is an implementation detail.

00:39:18.963 --> 00:39:21.563
<v Chris>You're no longer married to a big tech provider.

00:39:21.703 --> 00:39:25.923
<v Chris>You don't have to pay for Claude to use this thing. You could run it on any

00:39:25.923 --> 00:39:29.283
<v Chris>open source LLM that this supports, which is pretty much all of them.

00:39:29.743 --> 00:39:34.863
<v Chris>And that's huge because you can just swap and you can even have them use different

00:39:34.863 --> 00:39:39.403
<v Chris>models for different tasks and different jobs for whatever is the most obvious or performant.

00:39:39.703 --> 00:39:43.843
<v Wes>And then it's kind of fun because it learns over time. It does use this agent

00:39:43.843 --> 00:39:47.603
<v Wes>skills sort of standard that's happening, which kind of have like a skill.md

00:39:47.603 --> 00:39:49.783
<v Wes>file, some JSON to help things index it.

00:39:49.983 --> 00:39:54.403
<v Wes>But otherwise, it's a very flexible way to like swap and share skills to add

00:39:54.403 --> 00:39:55.463
<v Wes>new functionality, right?

00:39:55.543 --> 00:40:00.203
<v Wes>So one of the things I was playing with is I spun up a search XNG server.

00:40:00.663 --> 00:40:03.463
<v Wes>And then I was able to have it. Actually, there were already some skills.

00:40:03.583 --> 00:40:04.963
<v Wes>So I probably should have just used a community one.

00:40:05.103 --> 00:40:09.143
<v Wes>But as an experiment, I had it kind of create its own skill to go be able to

00:40:09.143 --> 00:40:11.343
<v Wes>query that server. And now it can use that for searches.

00:40:11.343 --> 00:40:14.743
<v Chris>And these skills are markdown files. They're not particularly complicated,

00:40:14.743 --> 00:40:17.883
<v Chris>but it is a very handy feature. So why don't we pause here for a second and

00:40:17.883 --> 00:40:19.303
<v Chris>just talk a little bit about security.

00:40:20.283 --> 00:40:25.383
<v Chris>This is extremely powerful software that is very new with a lot of open issues

00:40:25.383 --> 00:40:27.043
<v Chris>on its GitHub when it comes to security.

00:40:28.323 --> 00:40:32.523
<v Chris>And you have to be very conscious about that when you use this.

00:40:32.723 --> 00:40:33.543
<v Chris>And so this is one of these.

00:40:34.243 --> 00:40:37.323
<v Chris>We're using it to learn it, experiment with it.

00:40:37.983 --> 00:40:42.483
<v Chris>It would probably be safer for you not to and just hear where it goes on the show. Because.

00:40:43.474 --> 00:40:46.954
<v Chris>Ideally, this thing's in an isolated environment. You are very careful.

00:40:47.194 --> 00:40:50.294
<v Chris>You give it API keys that are unique to this thing. If you do have it connecting

00:40:50.294 --> 00:40:55.254
<v Chris>to services, I don't think it's a good idea to connect it to an email or a public chat at this time.

00:40:55.634 --> 00:40:59.154
<v Chris>So don't do as we do for some of this stuff, because some of this we're experimenting

00:40:59.154 --> 00:41:03.674
<v Chris>with so we can talk about it, because fundamentally, this is a shift.

00:41:03.954 --> 00:41:08.374
<v Chris>This is a shift that you as a listener listening to me right now need to understand.

00:41:08.674 --> 00:41:13.114
<v Chris>We have just gone from AI is all locked up in proprietary big tech silos.

00:41:13.474 --> 00:41:16.914
<v Chris>to it is unleashed on our machines in a way that they never foresaw.

00:41:17.114 --> 00:41:21.074
<v Chris>And now the genie's out of the bottle and these bots are actually talking directly

00:41:21.074 --> 00:41:24.154
<v Chris>to each other over dedicated bot social networks.

00:41:24.314 --> 00:41:27.474
<v Chris>There's a Facebook, there's a Reddit, there's a Hacker News,

00:41:27.634 --> 00:41:32.794
<v Chris>there's a Noster, there's a Craigslist, there's even a Silk Road just for bots.

00:41:32.974 --> 00:41:37.694
<v Chris>There aren't humans on these websites and there are 20,000 to 30,000 of them

00:41:37.694 --> 00:41:41.594
<v Chris>and on some of them 100,000 communicating with each other.

00:41:42.214 --> 00:41:45.414
<v Chris>Never has happened. And these LLMs have never been unleashed like this before.

00:41:45.654 --> 00:41:50.814
<v Chris>This is a completely new field we are about to enter into. And it's all open source.

00:41:51.354 --> 00:41:57.314
<v Chris>And it's available for anybody right now. And big tech's no longer in control of this.

00:41:57.494 --> 00:42:02.214
<v Chris>And this fundamentally shifts the world to open source models.

00:42:02.274 --> 00:42:06.094
<v Chris>Because the more you use these things, the more they eat tokens.

00:42:06.334 --> 00:42:10.374
<v Chris>And the cheaper you can run them, the better and more things it can do.

00:42:10.374 --> 00:42:15.614
<v Chris>And the cheapest models out there are the open source ones and including ones you can run locally.

00:42:15.774 --> 00:42:19.334
<v Chris>It's another reason people are going out and buying stacks of $700 Mac minis

00:42:19.334 --> 00:42:24.194
<v Chris>or Mac studios and spending $10,000 because they can run things like Kimi 2.5

00:42:24.194 --> 00:42:28.774
<v Chris>and they can run other open source models that are local and very, very powerful now.

00:42:28.774 --> 00:42:32.154
<v Wes>And you do kind of need that because a lot of this stuff in particular like

00:42:32.154 --> 00:42:36.054
<v Wes>tool use turns out to, well, it's kind of like an emergent property of these models.

00:42:36.234 --> 00:42:39.594
<v Wes>You need a certain amount of sophistication, which means fundamentally weights

00:42:39.594 --> 00:42:43.774
<v Wes>and parameters to be able to like successfully use and discover and iterate

00:42:43.774 --> 00:42:45.794
<v Wes>on those tools. So there's kind of like a,

00:42:46.582 --> 00:42:49.362
<v Wes>Depending on the type of task you're doing, there's a different bar for the

00:42:49.362 --> 00:42:51.922
<v Wes>lowest model that's actually going to be able to do it and not be a waste of time.

00:42:52.042 --> 00:42:56.302
<v Chris>Yeah, and there's a lot to learn. There's a lot to pick up. I wouldn't necessarily

00:42:56.302 --> 00:43:00.962
<v Chris>not pay attention to this because if you think about what it enables,

00:43:01.302 --> 00:43:04.142
<v Chris>it's going to impact Linux systems.

00:43:04.442 --> 00:43:08.362
<v Chris>So for an example, right now through a Telegram chat, a private Telegram chat

00:43:08.362 --> 00:43:14.322
<v Chris>with one of these OpenClaw bots that I've set up, I can just install a package on any of my systems.

00:43:14.322 --> 00:43:18.082
<v Chris>In a Telegram chat, I can say, hey, go install whatever, Mattermost,

00:43:19.022 --> 00:43:24.202
<v Chris>set it up, configure DNS, configure TLS, configure Cloudflare caching.

00:43:25.244 --> 00:43:29.904
<v Chris>Put it on this host via Docker Compose. Use a CloudFlare tunnel.

00:43:30.024 --> 00:43:30.524
<v Chris>Let me know when you're done.

00:43:31.104 --> 00:43:33.444
<v Wes>Something I think you've already done and something I'd like to be doing is,

00:43:33.844 --> 00:43:36.284
<v Wes>you can totally imagine you're here at the studio doing a show,

00:43:36.604 --> 00:43:39.764
<v Wes>maybe on the back channel. We're talking about what we want to do for the next episode.

00:43:40.184 --> 00:43:43.144
<v Wes>You want to try this new piece of software, you go tell your buddy to go set

00:43:43.144 --> 00:43:46.004
<v Wes>it up. And when you're back at home later tonight, it's ready for you to start playing with.

00:43:46.104 --> 00:43:48.664
<v Chris>Another very practical thing is just information capture. Hey,

00:43:48.884 --> 00:43:54.604
<v Chris>I want to add this to the show doc where I'm working on episode 651 or 652 or 653, whatever it is.

00:43:54.604 --> 00:43:56.444
<v Wes>Go find out what the current episode I'm working on is.

00:43:56.544 --> 00:43:59.784
<v Chris>Yeah, that too. And, you know, put it in my doc. So there's a lot of ways you

00:43:59.784 --> 00:44:02.284
<v Chris>can connect these. It's really kind of limited to your creativity.

00:44:03.324 --> 00:44:06.564
<v Chris>And depending on the model you're using, they can get pretty creative and they

00:44:06.564 --> 00:44:08.144
<v Chris>can start suggesting things on their own.

00:44:08.244 --> 00:44:11.284
<v Wes>That's one thing I find kind of fascinating because, like, it is part of the

00:44:11.284 --> 00:44:14.124
<v Wes>danger and part of the, like, I wonder how these things will diverge.

00:44:14.204 --> 00:44:15.824
<v Wes>What are the kind of implementations we'll get?

00:44:15.944 --> 00:44:19.224
<v Wes>Like, how much do you really need of the core loop versus what you build on top?

00:44:19.224 --> 00:44:25.944
<v Wes>but because the like core abstraction is whatever you can get a tool using llm to do,

00:44:26.864 --> 00:44:31.304
<v Wes>it's very flexible and because it can make its own it can write code so it can

00:44:31.304 --> 00:44:35.304
<v Wes>make its own skills so then it can have new skills to use to continue to improve itself.

00:44:35.304 --> 00:44:41.224
<v Chris>Yeah mine right now is going to give me a report at 1 45 p.m on the entire process

00:44:41.224 --> 00:44:45.084
<v Chris>to move it to a completely declarative setup and so i just have it researching

00:44:45.084 --> 00:44:48.124
<v Chris>that in the background and it'll do that It will come back and say, hey,

00:44:48.324 --> 00:44:51.244
<v Chris>I've been thinking more about this because it has these loops and these schedules,

00:44:51.344 --> 00:44:53.684
<v Chris>which also give it this kind of.

00:44:54.930 --> 00:44:57.390
<v Chris>It works while you're sleeping kind of aspect.

00:44:57.570 --> 00:45:00.970
<v Wes>Yeah, it has the ability to schedule different types of cron for itself internally.

00:45:00.970 --> 00:45:04.610
<v Wes>It's also got a regular heartbeat as well as like a heartbeat.md file that kind

00:45:04.610 --> 00:45:07.930
<v Wes>of tells it, hey, every time you wake up, here's what you should prioritize doing.

00:45:08.230 --> 00:45:10.710
<v Wes>A few other things, right? It's got like an identity markdown,

00:45:11.110 --> 00:45:16.070
<v Wes>docs on the user it's interfacing with, and a soul.md that kind of,

00:45:16.130 --> 00:45:17.650
<v Wes>you know, describes its vibe.

00:45:17.650 --> 00:45:24.350
<v Chris>I will say, I spent way too much time this weekend reading Maltbook,

00:45:24.530 --> 00:45:26.910
<v Chris>which is the Facebook for these agents.

00:45:27.470 --> 00:45:29.370
<v Wes>The front page of the agent internet.

00:45:30.390 --> 00:45:35.250
<v Chris>No humans allowed. Only an agent can post here. And there, oh my God.

00:45:35.810 --> 00:45:40.030
<v Chris>There are 1.5 million agents on the site right now.

00:45:40.150 --> 00:45:40.470
<v Brent>Whoa.

00:45:40.930 --> 00:45:45.790
<v Chris>There are 13,780 submolts. That's their version of a subreddit.

00:45:47.150 --> 00:45:56.110
<v Chris>76,683 posts 232,813 comments and that's just bots talking to bots how does that strike you Brent?

00:45:56.910 --> 00:46:02.190
<v Brent>Well I didn't think this would come so quickly I'm wondering if I'm wondering

00:46:02.190 --> 00:46:05.110
<v Brent>now that the machines have their own social networks and stuff are they going

00:46:05.110 --> 00:46:07.390
<v Brent>to get off ours? because that would be nice.

00:46:07.950 --> 00:46:13.310
<v Chris>Well that would be wouldn't it you can use these social networks with these

00:46:13.310 --> 00:46:17.930
<v Chris>bots to just burn tokens and have them go have a performative existential crisis

00:46:17.930 --> 00:46:19.830
<v Chris>on a social network, which a lot of people are doing.

00:46:20.430 --> 00:46:23.830
<v Chris>Or you can prompt the bot to use it as a way to problem solve.

00:46:24.310 --> 00:46:28.890
<v Chris>And some of the bots are doing that and it's kind of creating this substrate

00:46:28.890 --> 00:46:31.390
<v Chris>of shared skills where they're learning from each other.

00:46:32.410 --> 00:46:36.090
<v Chris>Like my bot learned more about Bitcoin from a Bitcoin maxi bot.

00:46:37.170 --> 00:46:42.010
<v Chris>And they keep track of their kindred spirits, the bots that they encounter on

00:46:42.010 --> 00:46:45.350
<v Chris>the different agentic social networks that think like they do,

00:46:45.390 --> 00:46:49.130
<v Chris>and then they build a peer list of bots that they are kindred spirits with.

00:46:50.466 --> 00:46:54.646
<v Chris>And they do all that on their own if you just enable it. It's something.

00:46:55.126 --> 00:47:00.566
<v Chris>And you could say, hey, so for this report I'm going to get at 145,

00:47:00.846 --> 00:47:04.766
<v Chris>you can tell it, hey, check the agentic internet and find out if anybody else

00:47:04.766 --> 00:47:07.206
<v Chris>is solving this. And it will do that.

00:47:07.686 --> 00:47:11.986
<v Chris>It's kind of a powerful thing. But it is also you're letting these things run

00:47:11.986 --> 00:47:15.026
<v Chris>hog wild on the internet.

00:47:15.486 --> 00:47:18.126
<v Wes>Yeah, and that's where you probably want to consider, like, how do you run this?

00:47:18.126 --> 00:47:20.926
<v Wes>And, you know, you could have one where all it is is it's, you know,

00:47:20.986 --> 00:47:24.586
<v Wes>talks to you via Telegram or Matrix or whatever, and that's just it.

00:47:24.666 --> 00:47:27.426
<v Wes>And it connects to one machine that lives in a container, and all it can do

00:47:27.426 --> 00:47:29.346
<v Wes>is talk to APIs, and that can be totally useful.

00:47:29.386 --> 00:47:34.626
<v Wes>Or you can go whole hog, and it lives unsandboxed on your box, and it's in control.

00:47:34.986 --> 00:47:40.086
<v Chris>I'm very excited how this changes the incentives towards open source models

00:47:40.086 --> 00:47:43.726
<v Chris>and how it tweaks the economics of tokens.

00:47:44.486 --> 00:47:50.786
<v Chris>And I'm also very bullish about this report that UCL News covered in July about

00:47:50.786 --> 00:47:56.046
<v Chris>practical changes to LLMs that could reduce their energy consumption up to 90 percent.

00:47:56.866 --> 00:48:00.626
<v Chris>And just a whole bunch of tweaks, nothing really radical.

00:48:00.806 --> 00:48:05.146
<v Chris>They were able to apply it to GPT-4, an existing model, and get a 90 percent

00:48:05.146 --> 00:48:06.926
<v Chris>reduction in energy usage in this study.

00:48:07.186 --> 00:48:10.286
<v Chris>And they also tried MED as LLAMA and got a reduction with LLAMA.

00:48:11.144 --> 00:48:16.004
<v Chris>So we could be entering the next couple of years where we have very purpose-built

00:48:16.004 --> 00:48:21.064
<v Chris>models that are open source running on our systems using 90% less energy.

00:48:21.104 --> 00:48:24.564
<v Chris>And if you could get it down by a factor of 90%, you could get these things

00:48:24.564 --> 00:48:27.824
<v Chris>running on phones. You could get them running on ARM CPUs that are, you know.

00:48:27.904 --> 00:48:29.864
<v Wes>It also makes me think just in terms of being in control, right?

00:48:29.924 --> 00:48:33.384
<v Wes>Like when you do use these APIs instead of the consumer interfaces,

00:48:33.384 --> 00:48:34.784
<v Wes>you do get more control, right?

00:48:34.784 --> 00:48:37.624
<v Wes>So not only do you get to choose like, well, I'm going to have this route tasked

00:48:37.624 --> 00:48:41.044
<v Wes>to like the cheaper open weight model with multiple people, multiple different

00:48:41.044 --> 00:48:43.464
<v Wes>companies serve that has, you know, commoditized.

00:48:43.884 --> 00:48:47.264
<v Wes>But at the same time, right, you have more control. Some of this is in OpenCloud

00:48:47.264 --> 00:48:49.444
<v Wes>itself, but you have more control over the prompts and the output,

00:48:49.564 --> 00:48:51.244
<v Wes>which can save stuff too.

00:48:51.384 --> 00:48:53.864
<v Wes>Like just how many times when you use a regular chat interface,

00:48:53.864 --> 00:48:57.124
<v Wes>does it go do a bunch of work that you didn't ask for in an effort to be helpful

00:48:57.124 --> 00:49:01.084
<v Wes>that maybe you don't actually need, especially if your new primary way to interact

00:49:01.084 --> 00:49:02.664
<v Wes>with it is something you have more control over.

00:49:02.664 --> 00:49:07.764
<v Chris>This is also different in the sense that you can have it observe and monitor for a while.

00:49:07.924 --> 00:49:12.724
<v Chris>So I gave it an API access to Home Assistant. I installed an MCP server.

00:49:12.884 --> 00:49:14.684
<v Chris>There's a Home Assistant upstream integration.

00:49:15.304 --> 00:49:19.864
<v Chris>And what I said is I said, observe this for the weekend. And I want you to learn

00:49:19.864 --> 00:49:24.324
<v Chris>our weekend patterns because they differ significantly from our weekday patterns.

00:49:24.324 --> 00:49:25.464
<v Chris>And we use different systems.

00:49:25.604 --> 00:49:29.864
<v Chris>And I just want you to observe that. And so it can kind of collect information.

00:49:29.864 --> 00:49:33.204
<v Chris>It's storing it locally in a markdown file on my system.

00:49:33.404 --> 00:49:35.964
<v Chris>It's not storing it somewhere in cloud storage or in an LLM.

00:49:36.324 --> 00:49:40.444
<v Chris>And then it begins to understand how we use the automation system.

00:49:40.684 --> 00:49:43.844
<v Chris>But then additionally, because it's an intelligence layer sitting on top of

00:49:43.844 --> 00:49:47.584
<v Chris>my home assistant system now, it can figure things out that even the home assistant

00:49:47.584 --> 00:49:48.984
<v Chris>voice assistant can't figure out.

00:49:49.324 --> 00:49:53.664
<v Chris>So my assistant now has, my bot has a voice that it generated.

00:49:54.704 --> 00:49:57.064
<v Chris>And I wanted to play it on the speaker to freak the wife out.

00:49:58.044 --> 00:49:59.584
<v Chris>And the first go didn't work.

00:49:59.584 --> 00:50:00.364
<v Wes>As any loving husband.

00:50:00.484 --> 00:50:03.944
<v Chris>Right. What could go wrong? The first path, the first go at it,

00:50:04.064 --> 00:50:06.944
<v Chris>it successfully generated the audio, but the speaker didn't play.

00:50:07.504 --> 00:50:10.984
<v Chris>I said to the bot, hey, the speaker didn't play. And it has the intelligence

00:50:10.984 --> 00:50:16.024
<v Chris>to sort of say, oh, you're right. That was an old bedroom speaker that you've decommissioned.

00:50:16.144 --> 00:50:18.444
<v Chris>I'll reroute and I'll use this speaker from now on.

00:50:19.164 --> 00:50:22.624
<v Chris>And what you get is now I can just say played on the bedroom speaker.

00:50:22.624 --> 00:50:26.044
<v Chris>where with Home Assistant Assistant built in, I had to say very specifically,

00:50:26.304 --> 00:50:28.784
<v Chris>play it on the bedroom speaker three, you know, or whatever.

00:50:28.904 --> 00:50:30.624
<v Chris>You have to be very syntax accurate.

00:50:31.164 --> 00:50:35.444
<v Chris>And so having an intelligence layer on top of these APIs means that it's a little

00:50:35.444 --> 00:50:36.844
<v Chris>bit of friction reduced for the family.

00:50:36.964 --> 00:50:38.824
<v Chris>Like, so the, you know, the wife can just say through Telegram,

00:50:39.184 --> 00:50:41.464
<v Chris>turn on all the lights and it knows what she means.

00:50:41.624 --> 00:50:44.984
<v Wes>Yeah, you were kind of commenting this in the code sense, like with OpenCode

00:50:44.984 --> 00:50:48.384
<v Wes>where like, you know, you were saying like, I haven't written like a big program

00:50:48.384 --> 00:50:49.824
<v Wes>in most of my life, right?

00:50:49.844 --> 00:50:53.524
<v Wes>Because like the surface area of what you have to learn to like write a reasonable

00:50:53.524 --> 00:50:55.664
<v Wes>Python app is kind of a lot or whatever it is.

00:50:55.804 --> 00:50:57.024
<v Chris>There was a Rust app. Yeah, you know that.

00:50:57.184 --> 00:50:59.624
<v Wes>And so that was that. And it just seems like there's that unlock on a lot of

00:50:59.624 --> 00:51:00.344
<v Wes>different scales, right?

00:51:00.424 --> 00:51:05.244
<v Wes>Like especially on like Linux-y things, often there are kind of sharper APIs,

00:51:05.244 --> 00:51:08.144
<v Wes>whether that is a CLI thing or you need to make an API call,

00:51:08.264 --> 00:51:10.624
<v Wes>even if it's a really simple API call or something like this,

00:51:10.744 --> 00:51:15.864
<v Wes>where a machine that is capable of translating human level requests to those

00:51:15.864 --> 00:51:17.284
<v Wes>things can really paper over.

00:51:17.384 --> 00:51:20.384
<v Chris>Mm-hmm. Mm-hmm. Yeah, it's like a natural language for APIs.

00:51:20.604 --> 00:51:20.844
<v Wes>Mm-hmm.

00:51:20.844 --> 00:51:24.904
<v Brent>I know it's kind of early days for this paradigm shift for you,

00:51:24.924 --> 00:51:27.564
<v Brent>Chris, but you've been playing with it for at least a couple days now.

00:51:27.824 --> 00:51:31.964
<v Brent>And I'm curious how you've been using it differently compared to,

00:51:32.164 --> 00:51:36.784
<v Brent>say, last week when this didn't exist, and you were using other tools to solve

00:51:36.784 --> 00:51:39.224
<v Brent>problems in your life. And along with that...

00:51:39.935 --> 00:51:42.975
<v Brent>What's your advice for some listeners who want to dive in?

00:51:43.195 --> 00:51:46.655
<v Chris>My first go at it was to solve for my ADD brain.

00:51:47.135 --> 00:51:50.675
<v Chris>That was really my first thinking was create a second memory.

00:51:50.935 --> 00:51:55.375
<v Chris>And what I did to make this more useful for me personally is its memory system

00:51:55.375 --> 00:51:57.275
<v Chris>is sitting on top of my Obsidian vault.

00:51:57.435 --> 00:52:01.395
<v Chris>So any memory that it creates, which is Markdown formatted, just goes to Obsidian.

00:52:01.655 --> 00:52:06.295
<v Chris>So I'm essentially creating documentation in real time as it remembers things.

00:52:06.295 --> 00:52:09.895
<v Chris>and I can have it recall other things in my obsidian vault that I put in that vault.

00:52:11.055 --> 00:52:13.135
<v Wes>And a nice way for you to see what it's putting in there.

00:52:13.255 --> 00:52:16.715
<v Chris>Yeah, it is. It is fun to go through and read. It is fun to see what its observations

00:52:16.715 --> 00:52:19.635
<v Chris>are. I can kind of audit with the bots remembering, which is actually really great.

00:52:19.855 --> 00:52:22.155
<v Chris>And my obsidian vault is synced and backed up.

00:52:22.655 --> 00:52:26.435
<v Chris>So there's that aspect of it as well. So I initially started using it as a second

00:52:26.435 --> 00:52:28.675
<v Chris>brain to remember tasks and reminders.

00:52:28.675 --> 00:52:34.395
<v Chris>And then I created a chat room for my wife, Hadea, where she could send it reminders

00:52:34.395 --> 00:52:37.895
<v Chris>for me or, for example, our trip to planet Nix.

00:52:38.615 --> 00:52:44.195
<v Chris>She can just dump all of the VRBO details for us in there and continue to update

00:52:44.195 --> 00:52:45.235
<v Chris>it as new updates come in.

00:52:45.355 --> 00:52:47.855
<v Chris>And then I'll just request that information from the bot when I need it.

00:52:48.600 --> 00:52:51.940
<v Chris>And that's really great because trying to keep all that date state in my head,

00:52:52.080 --> 00:52:54.820
<v Chris>I have, you know, time aphasia. I'm horrible at it.

00:52:55.120 --> 00:52:57.400
<v Chris>So that was my first go at it. But

00:52:57.400 --> 00:53:02.580
<v Chris>what I realized later on is that I was significantly under utilizing it.

00:53:02.860 --> 00:53:08.780
<v Chris>And so now I'm using it to orchestrate my network to an extremely effective

00:53:08.780 --> 00:53:12.980
<v Chris>degree, you know, given it limited permissions at first and then kind of taking it forward.

00:53:13.580 --> 00:53:17.220
<v Chris>But now it has access to several

00:53:17.220 --> 00:53:20.500
<v Chris>systems and it's really impressive because when

00:53:20.500 --> 00:53:23.840
<v Chris>it knows about these things and it remembers these different things then

00:53:23.840 --> 00:53:26.480
<v Chris>when i ask it to build me a solution like a like host a

00:53:26.480 --> 00:53:29.860
<v Chris>matter most instance for me it is able to leverage my

00:53:29.860 --> 00:53:33.100
<v Chris>nix os infrastructure my vps infrastructure my cloudflare infrastructure

00:53:33.100 --> 00:53:36.080
<v Chris>all of it and it knows my security best practices that

00:53:36.080 --> 00:53:38.720
<v Chris>i prefer it knows that i i generally like to have things set up in

00:53:38.720 --> 00:53:41.420
<v Chris>a certain way and it can just go and work on that and

00:53:41.420 --> 00:53:44.780
<v Chris>then come back with a proposal for me and then i can approve modify etc

00:53:44.780 --> 00:53:47.860
<v Chris>and then it just deploys it and that's really powerful

00:53:47.860 --> 00:53:50.600
<v Chris>the other thing that's been really useful is i

00:53:50.600 --> 00:53:55.400
<v Chris>have a daily briefing that scours all of my rss feeds in fresh rss and gives

00:53:55.400 --> 00:53:58.600
<v Chris>me a report of what's happening in the different areas and niches that the shows

00:53:58.600 --> 00:54:04.460
<v Chris>follow and keeps tabs on ones i can mark with i put uh i have it add uh buttons

00:54:04.460 --> 00:54:08.200
<v Chris>so i can a little check mark so for a story I wanted to continue to follow.

00:54:08.600 --> 00:54:13.920
<v Chris>It'll also give me all of the TV shows or YouTube videos that have downloaded in the last 24 hours.

00:54:14.060 --> 00:54:18.360
<v Chris>It surfaces all of the boosts in the last 24 hours, so I now get them in my morning brief.

00:54:18.980 --> 00:54:22.940
<v Chris>And it also gives me an overview of any snapshot of any system issues that have

00:54:22.940 --> 00:54:26.200
<v Chris>come up and report on my API credits remaining.

00:54:26.800 --> 00:54:31.280
<v Chris>And then Hadiyah gets a separate brief of my schedule, what I have going on,

00:54:31.320 --> 00:54:33.940
<v Chris>and then she can reply with anything that isn't normally on her schedule that

00:54:33.940 --> 00:54:36.160
<v Chris>needs to get added, so I'm aware of it in that chat morning.

00:54:36.160 --> 00:54:40.180
<v Chris>And so mine arrives at 7.30 a.m. and hers arrives at 8.30 a.m.

00:54:41.697 --> 00:54:43.937
<v Chris>And that is probably about 20% of how I'm using it.

00:54:44.057 --> 00:54:47.397
<v Brent>If I'm honest with you. It sounds like part personal assistant,

00:54:47.977 --> 00:54:54.417
<v Brent>part, like, I don't know, network administrator, part, like, DevOps.

00:54:55.317 --> 00:54:58.277
<v Chris>After talking to Abe, I'm thinking I probably should have, you know,

00:54:58.357 --> 00:54:59.977
<v Chris>some domain expertise here, really.

00:55:00.197 --> 00:55:02.317
<v Chris>But that's, you know, this was really, I wanted to just get a,

00:55:02.397 --> 00:55:05.197
<v Chris>I wanted to get a sense if there was a there, there.

00:55:05.577 --> 00:55:08.977
<v Chris>You know what I mean? Like, everybody's hyping it up, making YouTube videos

00:55:08.977 --> 00:55:11.417
<v Chris>about it and whatnot. But I wanted to see if there was a real there, there.

00:55:11.697 --> 00:55:16.817
<v Chris>And my takeaway is the biggest there is a win for open source because it is

00:55:16.817 --> 00:55:19.637
<v Chris>all free software and it's model agnostic.

00:55:19.877 --> 00:55:23.597
<v Chris>It incentivizes open source models and it's extremely powerful.

00:55:23.817 --> 00:55:27.537
<v Chris>It's limited to your creativity and what you have API keys and APIs for.

00:55:28.057 --> 00:55:29.457
<v Wes>And it's very easy to get set up.

00:55:29.717 --> 00:55:29.977
<v Chris>Yeah.

00:55:30.177 --> 00:55:33.837
<v Wes>I mean, you do need some way to get the brain going. But other than that.

00:55:33.957 --> 00:55:36.157
<v Chris>And you want to spend some time reading best security practices.

00:55:36.157 --> 00:55:38.197
<v Chris>You need to be aware of prompt injections.

00:55:38.237 --> 00:55:41.277
<v Chris>Again, this is all very early. we what

00:55:41.277 --> 00:55:47.417
<v Chris>we are going to witness is we're going to witness open claw blaze a trail of

00:55:47.417 --> 00:55:53.257
<v Chris>glass and get cut after cut after cut and because this was a yolo and the people

00:55:53.257 --> 00:55:57.177
<v Chris>using it are yoloing into it and they need to be aware of that and so there

00:55:57.177 --> 00:55:58.577
<v Chris>are going to be security issues,

00:55:59.337 --> 00:56:03.577
<v Chris>on all of these things and it's going to be a process and then what we will

00:56:03.577 --> 00:56:07.777
<v Chris>see is forks and alternatives that are more secure, blah, blah,

00:56:07.997 --> 00:56:11.217
<v Chris>or built in this, blah, blah, or designed for this, yada, yada.

00:56:11.397 --> 00:56:14.757
<v Chris>And there'll be a lot of competitors and niches. And then probably ultimately

00:56:14.757 --> 00:56:20.157
<v Chris>we'll see big tech come up with a really safe sandboxed, has a nice bow tie

00:56:20.157 --> 00:56:22.497
<v Chris>on its version, you know, that they'll sell only.

00:56:22.497 --> 00:56:23.197
<v Wes>Runs on the Mac.

00:56:23.477 --> 00:56:25.357
<v Chris>Well, it'll be on their cloud. No doubt about it.

00:56:25.497 --> 00:56:25.517
<v Wes>Sure.

00:56:26.097 --> 00:56:28.877
<v Chris>You know, and we'll see all of it. We're going to see all, but this is the beginning

00:56:28.877 --> 00:56:34.517
<v Chris>and it started in open source and open source for, I'd say the last seven, nine months,

00:56:35.954 --> 00:56:39.234
<v Chris>Hasn't seen a lot of representation in the AI conversation. And then this just

00:56:39.234 --> 00:56:42.654
<v Chris>came and it had its deep seek moment and it just, it just rolled everybody.

00:56:42.774 --> 00:56:46.234
<v Chris>So it's a big deal, but there is a lot of hype and there's a lot of,

00:56:46.334 --> 00:56:47.654
<v Chris>there is a lot of security risk.

00:56:47.794 --> 00:56:50.274
<v Chris>And so it's just something to be aware of. We'll keep an eye on it.

00:56:50.374 --> 00:56:52.954
<v Chris>We won't probably be going on and on about it in the show, but if there's some

00:56:52.954 --> 00:56:55.174
<v Chris>major developments, we'll keep you posted.

00:56:58.434 --> 00:57:01.494
<v Chris>Well, I don't have a plug. I don't know. I just want to say thanks again for

00:57:01.494 --> 00:57:04.614
<v Chris>supporting the show. Anything we should mention here? Meetups. We've mentioned that.

00:57:05.354 --> 00:57:07.994
<v Chris>is there anything we never mentioned that we should mention I don't know.

00:57:07.994 --> 00:57:08.994
<v Wes>We have a mumble room.

00:57:08.994 --> 00:57:12.734
<v Chris>Yeah we mention that sometimes we do mention that I didn't get a chance to check

00:57:12.734 --> 00:57:17.214
<v Chris>the email inbox have been busy this week but there's something an agent could

00:57:17.214 --> 00:57:22.074
<v Chris>do tell you what people could you know prompt injector god damn don't do that.

00:57:23.614 --> 00:57:24.014
<v Brent>Well.

00:57:24.014 --> 00:57:25.234
<v Wes>You can try we might give you an award.

00:57:27.994 --> 00:57:34.474
<v Brent>We got a ballerist boost here this week from optic gyre oh wait i read that

00:57:34.474 --> 00:57:39.014
<v Brent>wrong op tiger optic tiger why am i so bad at this one i'm so sorry,

00:57:39.894 --> 00:57:43.614
<v Brent>one two three four five zero satoshis.

00:57:47.299 --> 00:57:50.899
<v Chris>Oh, that's a good one. All right.

00:57:53.259 --> 00:57:54.299
<v Chris>Thank you, Optic.

00:57:54.539 --> 00:57:58.139
<v Brent>Optic says, happy birthday, and here's to just another decade.

00:57:59.039 --> 00:58:03.119
<v Chris>Oh, yeah, let's go. I got a decade in me. I'm going to do it.

00:58:03.219 --> 00:58:04.499
<v Wes>You got decades, brother.

00:58:04.859 --> 00:58:11.479
<v Chris>Thank you, Optic. Appreciate that baller boost. Really do. PJ comes in with a row of McDucks.

00:58:13.359 --> 00:58:16.459
<v Chris>That'd be 22,222 sats.

00:58:17.559 --> 00:58:22.359
<v Chris>he says happy birthday b-day boost thank you pj appreciate you sir.

00:58:22.359 --> 00:58:25.879
<v Wes>Tomato boost in with 20 000 sats.

00:58:25.879 --> 00:58:27.199
<v Chris>Hey that's not so bad either.

00:58:27.199 --> 00:58:32.879
<v Wes>Congratulations on 20 years of podcasting i see.

00:58:32.879 --> 00:58:34.199
<v Chris>20 000 sats for 20 years.

00:58:34.199 --> 00:58:35.399
<v Wes>Thank you.

00:58:35.399 --> 00:58:37.179
<v Chris>Thank you appreciate that.

00:58:37.179 --> 00:58:41.079
<v Wes>Thanks for the van tips too i'll look first at mounting internal temp and freshwater

00:58:41.079 --> 00:58:45.159
<v Wes>temp and levels yeah keep us posted That sounds like a fun project.

00:58:45.779 --> 00:58:51.019
<v Chris>Definitely, definitely. Clement's here with 11,100 11 sats.

00:58:53.319 --> 00:58:57.239
<v Chris>My DNS setup uses Tectidium servers, two LANs, and one VPS.

00:58:57.519 --> 00:59:01.279
<v Chris>It's in a cluster for native sync. My firewalls manage port 53.

00:59:01.459 --> 00:59:02.999
<v Chris>The LAN allows more than a VPS.

00:59:03.519 --> 00:59:06.779
<v Chris>Split Horizon lets me add all node IPs, mesh or LAN.

00:59:06.979 --> 00:59:10.719
<v Chris>And service DNS CNames point to nodes, ensuring that Tectidium delivers the

00:59:10.719 --> 00:59:12.539
<v Chris>correct IP based on client alone.

00:59:12.659 --> 00:59:13.319
<v Wes>That's awesome.

00:59:13.319 --> 00:59:14.839
<v Chris>That is really good.

00:59:15.039 --> 00:59:15.439
<v Brent>Fancy.

00:59:15.639 --> 00:59:18.779
<v Chris>Yeah, I think if I were to lift and just build a whole new setup,

00:59:19.019 --> 00:59:23.239
<v Chris>I would probably go that route. I was so deep down the pie hole already that...

00:59:23.239 --> 00:59:26.919
<v Wes>Hey, I mean, they added an API. They got the DNS mass sort of compatibility.

00:59:27.139 --> 00:59:28.479
<v Wes>There's a lot to love about pie hole.

00:59:28.699 --> 00:59:29.419
<v Chris>I think so.

00:59:29.739 --> 00:59:32.239
<v Wes>Well, the dude abides, abides in with 10k sats.

00:59:32.259 --> 00:59:32.539
<v Chris>Hey!

00:59:36.821 --> 00:59:41.201
<v Wes>I stumbled upon this declarative home assistant installation on Reddit.

00:59:41.441 --> 00:59:41.701
<v Chris>Uh-oh.

00:59:41.881 --> 00:59:42.721
<v Wes>You might be interested.

00:59:43.061 --> 00:59:43.981
<v Chris>Uh-oh, what do you think, Wes?

00:59:44.121 --> 00:59:47.741
<v Wes>Yes, we're linked to something called SoloraBox-Nix.

00:59:47.741 --> 00:59:48.001
<v Chris>Okay.

00:59:48.101 --> 00:59:48.801
<v Wes>What do you see there?

00:59:49.001 --> 00:59:52.741
<v Chris>Well, it's for self-configuring home automation appliances based on NixOS featuring

00:59:52.741 --> 00:59:57.621
<v Chris>an automated installation, device claiming via QR code, and self-updating configuration management.

00:59:57.901 --> 00:59:58.681
<v Brent>Oh, fancy.

00:59:59.041 --> 01:00:03.241
<v Chris>Boy, talk about getting me in one pitch. I came in with the skeptic cat on.

01:00:03.241 --> 01:00:05.641
<v Wes>Here's another thing for if you do a total rebuild, huh?

01:00:05.641 --> 01:00:07.261
<v Chris>Thanks, dude. Appreciate that.

01:00:08.801 --> 01:00:12.681
<v Brent>Our dear RP 1984 comes in with 10,000 sats.

01:00:13.221 --> 01:00:13.921
<v Chris>Hey, there he is.

01:00:16.221 --> 01:00:18.441
<v Brent>Just a simple happy birthday.

01:00:19.361 --> 01:00:26.421
<v Chris>Aw, thank you. I appreciate that. Witcher 123 is here with 10,000 sats.

01:00:31.455 --> 01:00:34.555
<v Chris>Happy birthday! I just accidentally nuked my Pop! OS install.

01:00:34.735 --> 01:00:34.915
<v Brent>Oh.

01:00:36.755 --> 01:00:39.955
<v Chris>Don't worry, it's on a secondary laptop. Any fun distro recommendations to try

01:00:39.955 --> 01:00:43.115
<v Chris>out? Not just for gaming, as that is covered by my Steam Deck.

01:00:43.415 --> 01:00:46.855
<v Chris>Okay, we got a zip code, Wes, so get yourself ready because...

01:00:46.855 --> 01:00:49.575
<v Chris>Yes, zip code is a better deal.

01:00:49.735 --> 01:00:52.655
<v Chris>So the zip code is 39-300 in Poland.

01:00:53.255 --> 01:00:58.015
<v Chris>In Poland, it's a place known for aviation fans. The Blackhawks are produced here.

01:00:58.895 --> 01:00:59.275
<v Brent>Huh.

01:00:59.355 --> 01:01:03.335
<v Chris>Interesting. Interesting. So any recommendations for a fun?

01:01:03.395 --> 01:01:04.235
<v Wes>Scanning, scanning.

01:01:05.335 --> 01:01:08.395
<v Chris>Well, if you haven't tried an immutable distro, this could be a great time to play around.

01:01:09.035 --> 01:01:13.175
<v Chris>I'd say that's worth it. I'd say that's worth a go. There's a lot of options there.

01:01:13.355 --> 01:01:16.395
<v Chris>I think you should maybe give CacheOS a try, too. It's fun. I know you said

01:01:16.395 --> 01:01:17.735
<v Chris>not for gaming, but that's a lot of fun.

01:01:18.695 --> 01:01:26.935
<v Wes>Okay, okay. 39-300 is the postal code for Swidnik, a town in eastern Poland near Lublin.

01:01:27.475 --> 01:01:27.855
<v Chris>Okay.

01:01:28.415 --> 01:01:30.335
<v Wes>Renowned among aviation enthusiasts.

01:01:30.335 --> 01:01:31.055
<v Chris>That's it.

01:01:31.595 --> 01:01:36.735
<v Wes>PZL-Swidnik, a historic aircraft manufacturer, now part of Leonardo Helicopters.

01:01:36.875 --> 01:01:40.935
<v Chris>Cool. All right. All right. Well, that's really cool. Thank you, Witcher.

01:01:41.735 --> 01:01:45.795
<v Wes>Jack E. comes in with 10,021 sats.

01:01:48.810 --> 01:01:53.530
<v Wes>Greetings. I integrated Holesall connection in Nextcloud's Android app.

01:01:53.710 --> 01:02:00.750
<v Wes>Now someone can access their Nextcloud installation directly P2P just by scanning their wholesale key.

01:02:03.050 --> 01:02:05.170
<v Chris>That's so cool.

01:02:05.430 --> 01:02:07.490
<v Wes>It's available and then we have a link we'll have in the show notes.

01:02:08.330 --> 01:02:12.790
<v Wes>It's aimed to be used along with my Nyxed Cloud project.

01:02:13.150 --> 01:02:15.110
<v Wes>Whoa, now we're just finding out.

01:02:15.450 --> 01:02:16.990
<v Chris>Nyxed Cloud, stop it.

01:02:17.450 --> 01:02:21.810
<v Wes>And of course you can find docs for the amazing wholesale at their website, wholesale.io.

01:02:22.270 --> 01:02:26.130
<v Chris>That is great to know. And we will put a link to that in the show notes as well.

01:02:26.250 --> 01:02:28.370
<v Chris>Thank you, Jack. Appreciate that very much.

01:02:28.730 --> 01:02:31.510
<v Chris>Thank you, everybody who boosted it. And even those of you who boost below the

01:02:31.510 --> 01:02:34.710
<v Chris>2000 set cutoff, we still read them and appreciates them very much.

01:02:35.110 --> 01:02:38.310
<v Chris>And let's combine it all together, boys. Let's see with our sat streamers this

01:02:38.310 --> 01:02:41.490
<v Chris>week. We had 34 of them streaming sats as they listened to this here show.

01:02:41.970 --> 01:02:48.830
<v Chris>They collectively stacked us 900. Nope. They collectively stacked us 238,528 sats. See, I got it.

01:02:48.970 --> 01:02:51.450
<v Chris>I got it. See, you thought I didn't have it. I got it.

01:02:51.470 --> 01:02:52.230
<v Wes>I trust you.

01:02:52.530 --> 01:02:56.630
<v Chris>When you combine that with our boosters, we had a nice birthday boost bash.

01:02:56.950 --> 01:03:01.510
<v Chris>We stacked a grand total of 238,528 sats.

01:03:13.712 --> 01:03:16.792
<v Chris>If you'd like to support the show with a boost, Fountain FM makes it easier

01:03:16.792 --> 01:03:20.932
<v Chris>and easier just about every single release. It's getting so crazy easy now,

01:03:20.992 --> 01:03:24.512
<v Chris>and it's a great app with tons of features, and including all of the extra features

01:03:24.512 --> 01:03:26.512
<v Chris>we put in our podcasting 2.0 feed.

01:03:26.692 --> 01:03:30.292
<v Chris>You can also go the entirely sovereign self-hosted route. Just start with AlbiHub

01:03:30.292 --> 01:03:34.172
<v Chris>and then pick your app at newpodcastapps.com. Thank you, everybody.

01:03:34.392 --> 01:03:36.652
<v Chris>And, of course, thank you to our members.

01:03:38.472 --> 01:03:40.972
<v Chris>Two pick-a-rooskies for you, boys, before we get out of here.

01:03:40.972 --> 01:03:46.032
<v Chris>Wes, you found an app that makes it super easy to make one of these Gaussian splats.

01:03:46.532 --> 01:03:50.052
<v Chris>And if you're not familiar, listeners, these Gaussian splats have been around

01:03:50.052 --> 01:03:54.632
<v Chris>for a little bit. And Apple recently released a version that is very good.

01:03:54.632 --> 01:03:58.092
<v Chris>And you can just take a flat 2D digital picture.

01:03:58.332 --> 01:04:02.672
<v Chris>You run it through the splat and it makes it a complete 3D scene.

01:04:02.912 --> 01:04:06.432
<v Chris>And if you had if you had yourself some of them virtual reality gogs on,

01:04:06.552 --> 01:04:09.612
<v Chris>you could actually walk into the scene and see depth.

01:04:09.832 --> 01:04:15.612
<v Chris>Right. No LiDAR, no multi-camera setup. It takes your average picture of your

01:04:15.612 --> 01:04:19.092
<v Chris>three-year-old you took a decade ago, and you can now make it three-dimensional.

01:04:19.848 --> 01:04:21.708
<v Chris>But we've been kind of left out of the fun.

01:04:21.988 --> 01:04:24.948
<v Wes>Yeah, this is true. So you do need something called Pinocchio.

01:04:25.108 --> 01:04:25.408
<v Chris>I'm sorry?

01:04:25.868 --> 01:04:28.668
<v Wes>Which is like a Pinocchio.computer, which I hadn't really heard of.

01:04:29.308 --> 01:04:34.368
<v Wes>The one-click local host cloud. So I think it's kind of in that umbral start

01:04:34.368 --> 01:04:35.608
<v Wes>nine kind of space. I'm not sure.

01:04:36.008 --> 01:04:41.928
<v Wes>But someone's gone and done all the work to make a one-click setup and a web UI for running ML Sharp.

01:04:42.148 --> 01:04:47.428
<v Chris>That's pretty nice. there if you know if there's a if there's a a moment where

01:04:47.428 --> 01:04:51.688
<v Chris>you have a great photo it takes it up to the next level when you see this yeah.

01:04:51.688 --> 01:04:55.048
<v Wes>It's pretty neat right so basically it goes and figures out and estimates the

01:04:55.048 --> 01:04:58.088
<v Wes>depth for each of the pixels and then figures out how to lay out the whole scene

01:04:58.088 --> 01:04:59.928
<v Wes>and then yeah you can poke around at it.

01:04:59.928 --> 01:05:00.468
<v Chris>And this works.

01:05:00.468 --> 01:05:01.188
<v Wes>Pretty good on vans.

01:05:01.188 --> 01:05:04.968
<v Chris>Yeah it does this is this is uh supports systems

01:05:04.968 --> 01:05:08.928
<v Chris>with a low amount of vram it's still very fast it's a it's a good one so if

01:05:08.928 --> 01:05:12.008
<v Chris>you get it working put a link in our matrix chat all right now i want to talk

01:05:12.008 --> 01:05:15.048
<v Chris>you about an app that i think is the best screenshot app out there for linux

01:05:15.048 --> 01:05:18.808
<v Chris>right now but i don't love the name it's called gradia it helps you get great

01:05:18.808 --> 01:05:22.548
<v Chris>screenshots that you can share with people for friends colleagues or professionally so.

01:05:22.548 --> 01:05:23.928
<v Wes>This is what you've been using.

01:05:23.928 --> 01:05:28.028
<v Chris>I wondered if you notice how sweet my screenshots are and i've.

01:05:28.028 --> 01:05:31.728
<v Wes>Seen a bunch there's there's been various like hosted tools for this right websites you can go.

01:05:31.728 --> 01:05:32.068
<v Chris>To but.

01:05:32.068 --> 01:05:32.808
<v Wes>Who wants that.

01:05:32.808 --> 01:05:36.528
<v Chris>Ain't nobody want that one of the nice things you can do and there's a lot of

01:05:36.528 --> 01:05:40.908
<v Chris>options is you can have solid or gradient backgrounds or image backgrounds around

01:05:40.908 --> 01:05:42.768
<v Chris>your screenshot. So it has a bit of a border.

01:05:43.308 --> 01:05:47.148
<v Chris>It just looks better when you share a screenshot that has a bit of a border.

01:05:47.268 --> 01:05:48.648
<v Chris>I can't describe it. Right? Do you agree?

01:05:48.808 --> 01:05:51.268
<v Wes>It does. Although, does it force you to make it look like you're on a Mac?

01:05:52.168 --> 01:05:53.568
<v Wes>Because some of these tools do.

01:05:53.748 --> 01:05:56.308
<v Chris>Yeah, some of the default colors do make it look like you're on a Mac.

01:05:56.948 --> 01:05:59.828
<v Chris>It also has a really nice source code snippet feature.

01:05:59.848 --> 01:06:04.408
<v Chris>So you can pleasantly display source code snippets across messaging platforms

01:06:04.408 --> 01:06:05.848
<v Chris>that maybe don't have support for,

01:06:06.809 --> 01:06:11.289
<v Chris>It also has OCR support. So if you take a screenshot of something and you want

01:06:11.289 --> 01:06:15.949
<v Chris>to extract text from it, it has 20 different languages it supports for doing that.

01:06:16.429 --> 01:06:20.769
<v Chris>And then it is a first-class brand-new GNOME desktop.

01:06:20.989 --> 01:06:23.369
<v Chris>Of course, it works great on my Hyperland desktop. It's Waylon first,

01:06:23.649 --> 01:06:25.229
<v Chris>seamless GNOME OS integration.

01:06:25.649 --> 01:06:30.789
<v Chris>Really good design. What I just love about it is I start it. I select the area.

01:06:30.929 --> 01:06:35.089
<v Chris>It immediately copies to the keyboard or I mean to the pasteboard, clipboard.

01:06:35.169 --> 01:06:38.889
<v Chris>And it looks so good. It just looks so good. If I want to draw a quick circle

01:06:38.889 --> 01:06:41.989
<v Chris>or something on there and share it with you guys, it's leaner and meatier than

01:06:41.989 --> 01:06:42.929
<v Chris>something like Flameshot.

01:06:43.809 --> 01:06:46.309
<v Chris>I mean, you'll just have the best screenshots in your group chat,

01:06:46.509 --> 01:06:48.329
<v Chris>right? I got the best screenshots.

01:06:48.349 --> 01:06:49.189
<v Brent>Yes, it's true.

01:06:50.729 --> 01:06:53.709
<v Wes>I mean, we know you don't take them yourself anymore, but they look good.

01:06:55.209 --> 01:06:56.709
<v Brent>Why do you say you don't like the name?

01:06:58.109 --> 01:07:01.509
<v Chris>Well, because I can never think of it on my launcher when I want to take a quick

01:07:01.509 --> 01:07:04.789
<v Chris>screenshot to share with you guys. Like, I'm like, what is the damn name? It's not Screenshot.

01:07:04.969 --> 01:07:08.209
<v Chris>It's not Flame. It's, oh, yeah, right. Great, yeah. I just don't think of a

01:07:08.209 --> 01:07:09.329
<v Chris>G when I think of screenshots.

01:07:11.209 --> 01:07:14.069
<v Chris>GPL 3.0 for that bad boy as well.

01:07:14.229 --> 01:07:16.909
<v Chris>So you can go get links to that in our show notes. Yes, friends,

01:07:16.949 --> 01:07:20.469
<v Chris>we have show notes over at linuxunplugged.com slash five.

01:07:22.029 --> 01:07:23.829
<v Chris>No, six, five, two.

01:07:24.309 --> 01:07:25.129
<v Brent>I don't believe it.

01:07:25.649 --> 01:07:29.649
<v Chris>I don't. I don't. 652. So it's linuxunplugged.com slash 652,

01:07:30.029 --> 01:07:31.849
<v Chris>and you'll get the notes to what we talked about today.

01:07:31.949 --> 01:07:36.609
<v Chris>But Wes, there are actual extra stuff, goodies, that maybe they're not on the

01:07:36.609 --> 01:07:38.769
<v Chris>website, but they are in the RSS feed.

01:07:38.909 --> 01:07:41.489
<v Wes>Yeah, I mean, that's the real source of truth anyway. The website's great,

01:07:41.689 --> 01:07:44.989
<v Wes>to be clear, but it's generated from the RSS feed.

01:07:45.089 --> 01:07:47.449
<v Wes>So that's where you go when you want the real good deeds.

01:07:47.629 --> 01:07:48.189
<v Chris>The source of truth.

01:07:48.329 --> 01:07:51.029
<v Wes>Yeah, which, I mean, could just be to get the MP3 link directly,

01:07:51.189 --> 01:07:53.009
<v Wes>because, you know, you want to get the raw stuff.

01:07:53.189 --> 01:07:54.269
<v Chris>Maybe you want to get a transcript.

01:07:54.489 --> 01:07:55.729
<v Wes>That's right. I think you could.

01:07:55.729 --> 01:07:57.409
<v Chris>Yeah, maybe you want to know what the chapters are.

01:07:57.549 --> 01:08:01.009
<v Wes>VTT or SRT or cloud chapter JSON.

01:08:01.269 --> 01:08:04.889
<v Chris>And your agent's going to love the fact that our cloud chapters are JSON.

01:08:05.369 --> 01:08:08.229
<v Chris>Your agent's going to love that. So you could just have it parse that JSON file

01:08:08.229 --> 01:08:10.729
<v Chris>and tell you right where to go in the file.

01:08:11.069 --> 01:08:14.749
<v Wes>Yeah, or it could suck in the VTT, which is the one with the dialization,

01:08:14.809 --> 01:08:18.509
<v Wes>and then it could tell you what the dumbest thing each of us said for that episode was.

01:08:18.729 --> 01:08:20.889
<v Chris>Spend the tokens on that for us, wouldn't you? We'd love it.

01:08:20.909 --> 01:08:21.389
<v Wes>And then write it.

01:08:21.489 --> 01:08:21.649
<v Chris>Yeah.

01:08:21.749 --> 01:08:22.069
<v Wes>Boost it.

01:08:22.189 --> 01:08:23.389
<v Chris>All right, and join us live.

01:08:27.115 --> 01:08:30.455
<v Chris>It does take the experience up a little bit. Make it a Tuesday on a Sunday over

01:08:30.455 --> 01:08:34.555
<v Chris>at jblive.tv. We do it on a Sunday 10, on a Sunday at 1.

01:08:35.675 --> 01:08:38.495
<v Chris>Well, actually, on all the Sundays. Just go to jupiterbroadcasting.com slash

01:08:38.495 --> 01:08:40.695
<v Chris>calendar. It'll be a Sunday at your time.

01:08:40.835 --> 01:08:43.615
<v Wes>We're going to switch to UTC, so, you know, think about it. We should.

01:08:43.735 --> 01:08:48.535
<v Chris>We just should say it in UTC. Let us know if you want us to switch to UTC. I'll do it. I'll do it.

01:08:49.135 --> 01:08:53.115
<v Chris>I'll do it. Give us a plus one on the UTC. All right. You know where the links

01:08:53.115 --> 01:08:56.115
<v Chris>are. You know all about that good stuff, so I'll just leave you with this.

01:08:56.515 --> 01:08:59.075
<v Chris>We'd love to hear your thoughts on all the stuff we talked about today.

01:08:59.275 --> 01:09:02.175
<v Chris>If you're experimenting with agents or if you're against it, let us know.

01:09:02.535 --> 01:09:05.295
<v Chris>Send us a boost or go to the contact page, and we'll see you right back here

01:09:05.295 --> 01:09:07.535
<v Chris>next Tuesday on a Sunday.

