06:35:57 since when do irc networks use @ in their names? 06:47:30 nimaje: to mark ops 06:49:29 ridcully: that wouldn't be in the name *of* the irc network, my question was in response to ring0_starrs "since there is an irc network by that name too" to -hackers@ 06:50:08 sorry, i misread 06:53:23 hi all 06:53:25 :) 06:53:30 hi all 06:53:45 x0x17: O.o you look suspicious 06:54:22 yeah 06:54:32 just has a busy day today 06:54:37 what about you badkat 06:55:36 cool, im smoking some meth while i fuzz some cloudflare crap 06:59:00 still using meth 06:59:01 ? 06:59:10 stupid badkat 06:59:22 what is fuzz 06:59:28 fuzzing mean hacking badkat 06:59:30 ? 06:59:47 just stay away from drug and smoke if grandma want 07:00:30 look, is the only way i can focus on this new crappy golang codebases 07:00:54 if the world comes back to C/C++ i will definitly drop meth 07:00:55 golang ? 07:01:02 haha 07:01:04 i see 07:01:07 golang ? 07:01:08 I prefer to just take the same exact, low dose of dexamphetamine every day for my adhd 07:01:19 haha i rather code in c / c++ rather then golang 07:01:41 whenever I take high doses of stims I just end up watching porn for 5 hours. but hey, whatever works for you :P 07:01:53 i never code in non-sacred languages 07:02:25 mtll: you are from the elite club that watches p0rn in freebsd? :D 07:02:30 my broda :) 07:03:08 :D 07:03:17 sorry disconnect 07:04:24 stay away from cigaratte and smoke and the rest 07:04:50 if you want ask your governor to sponsor you for placebo for research 07:05:04 I cut out cigarettes for vaping like 10 months ago now 07:05:19 don'tttttt 07:05:22 you will bleed 07:05:26 that's crazy 07:05:31 gonna quit the vaping soon too 07:05:38 i rather smoke rather then vaping 07:05:50 anyway since this is channel freebsd 07:05:54 come to channel #politic 07:05:56 for a chat 07:06:03 before Admin complain 07:06:12 nah vaping is definitely safer. most of the really scary stuff has since been attributed to badly designed vapes and e-juice with questionable chemicals in them 07:06:26 nono 07:06:36 you didn't know that is from chemical 07:06:40 not from tree 07:06:54 stay away to put chemical to your lung 07:07:10 well, cigarette smoking has a lot more chemicals in it :P 07:09:11 anyway, to me vaping is just nicotine replacement therapy that actually worked for me, unlike patches and gum and all the rest 07:09:38 gonna quit that do soon using varenicline, then I'm out 07:12:19 doesn't sound like much freebsd talk... 07:14:01 anyway, for core microarchitecture-based x86 cpus, coretemp.ko exposes its readings via dev.cpu.N.temperature. for allwinner SoCs, it's dev.aw_thermal.N.cpu 07:14:26 is there any utility that generalizes these readouts so that there's one consistent interface to getting the temperature? 07:16:54 mtll come hop in to #politic 07:17:03 admin already not allow us 07:17:19 #politic 07:17:22 oops 07:18:02 ring0_starr: i dont think so, AFAIK that is pretty much platform specific as in linux 07:18:14 but could be made like you say IMHO 07:54:08 Hi all. I installed FreeBSD 15.0 on an amd64 box on Thursday, May 1. I am using pkg for package management. The file /etc/pkg/FreeBSD.conf appropriately defaulted to the "current" branch. However, since that date, I have received only the message "Your packages are up to date." when trying to upgrade packages. 07:54:40 I have confirmed multiple upgrade candidates through the website. I have also tried using the alternative file /usr/local/etc/pkg/repos/FreeBSD.conf with no success. There must be something obvious I'm missing but I can't determine what it is. Any suggestions? 07:57:51 did you install anything via pkg? why do you expect those packages to have updates? 07:58:44 dutch: only modify /etc/pkg/FreeBSD.conf 07:59:06 /usr/local -> is meant for third party software only (packages) 08:00:00 badkat: yes, i backed it out after it had no effect, thanks. 08:00:15 no, use ucls override feature and override stuff from /etc/pkg/FreeBSD.conf in /usr/local/etc/pkg/repos/FreeBSD.conf (as pkg.conf(5) tell you too) 08:00:31 dutch: can you give us the result of "cat /etc/pkg/FreeBSD.conf|grep url" ? 08:01:34 url: "pkg+https://pkg.FreeBSD.org/${ABI}/latest" 08:04:04 ok then when you want to know if there is updates for a package you want, go to freshports.org and check the release table on the "latest" field 08:04:08 15-current builds are fairly intermittant, the last one was on april 27, so it's expected you have no updates 08:05:15 yes you should use 14.2-release to get the most updated stuff 08:05:36 nimaje: yes, I've installed many packages. I expect them to have updates because, e.g., I have vim-9.1.1265 installed but according to the website, 9.1.1362 is available. 08:05:37 why did you choose 15.0? 08:05:57 dutch: did you intend to install a development version instead of a release of freebsd? 08:07:05 nimaje: yes, I have newer hardware and my discrete graphics card and sound card required it. 08:08:31 ivy: ok, wasn't aware of that 08:08:56 dutch: are you sure that hardware wont work in 14.2? 08:09:03 yes 08:18:37 Thanks for the info everyone; i'll likely just play around with this system and build from ports. Thanks again, and have a nice day. 12:57:37 hi all 14:14:27 03:59 < badkat> /usr/local -> is meant for third party software only (packages) 14:15:06 I think "third party software only" is a bit imprecise. 14:15:35 /usr/local/ is the default prefix for ports/pkg managed software installs. 14:16:46 I would very strongly advise against installing things to /usr/local (or to the / or /usr/ prefixes) that you hand compiled, or wrote yourself (and didn't install from ports/pkg). 14:17:13 Personally, I use /usr/opt/ for such things. 14:18:02 You don't want ports/pkg knocking heads with non-ports/pkg things. 16:51:33 I just installed FreeBSD on my Raspberry Pi 3. I understand that the wifi does not work on it, but I believe it uses an SDIO bus. Is there a way to list devices on that bus? 17:26:00 CrtxReavr: yes i meant that (packages) that are ports ofc 17:28:17 badkat, but there's also been device drivers and other OS features that have been distributed as ports/pkg. 17:28:50 Dunno how long you've been using FreeBSD, but there used to be a crazy list of shit in the base OS. 17:29:58 gcc, bind, & perl come to mind. 17:30:20 X! 17:30:33 rip /usr/X11R6 17:31:06 I dont' think X was ever part of base, at least not prior to 3.0 when I got my feet wet. 17:32:09 The installer had options to install X and related packages. . . and yes, the default PREFIX for X11-related things was /usr/X11R6/ 17:33:37 Fun idea - GenAI platform trained on all BSD code and documentation. 17:34:26 * farhan ducks 17:37:43 I'm reminded of when some guys tried to get a BSD-centric Slashdot-alike site going. 17:38:08 I mean, they did. . . even used Slashcode running on FreeBSD. . . 17:38:14 But it never took off as a thing. 17:42:52 it's all over now 17:44:14 farhan: FreeBSD is well-documented in comparison to GNU/Linux where it comes to stuff like `sysctl -d`. Linux has an abundance of undocumented tunables. 17:59:08 Who needs documentation when you have journalctl output? :3 18:19:55 who needs a computer when you throw the one with systemd into a lake? 18:49:12 farhan: someone may have done that or similar: https://papers.freebsd.org/2024/asiabsdcon/wang-Leveraging-the-Power-of-ChatGPT-and-Vector-Databases-in-the-FreeBSD-Expert-System.files/wang-Leveraging-the-Power-of-ChatGPT-and-Vector-Databases-in-the-FreeBSD-Expert-System-paper.pdf 18:52:06 not trained on the code though it seems 18:54:16 ohgod expert systems 18:55:40 anyway, this crazy paper claims modern expert systems are replacing experts with ML models. which is nonsense 18:56:59 this is a thing which died 30 years ago 18:57:56 truly, a garbage time of history 18:59:26 they will be laid off, all these supposed experts 19:03:41 FreeBSD garlic 14.2-RELEASE-p3 FreeBSD 14.2-RELEASE-p3 releng/14.2-n269524-1eb03b059e56 ARMIN amd64 19:03:45 doobdodoob 19:04:17 how do you like FreeBSD armin ? 19:04:34 mzar: I'm a long time user, I believe my first version was 4.x 19:04:42 OK 19:04:43 mzar: but it's nice yea, I like it 19:05:37 cool 19:06:06 I believe that its simplicity is un-matched, there's nothing even close that does *so* much with minimal effort 19:08:15 you nailed it, it's operating system best fitted for lazy people 19:16:33 duncan: i don't see anything in that paper about replacing real experts. it looks to me more like to assist users and admins in being more productive (less time spent resolving issues) 19:33:11 verbatim: "Modern expert system usually uses 19:33:12 machine learning to simulate the behavior of the domain experts." - whereas the original expert systems were computer systems taking advantage of expert knowledge, not simulating expertise 19:33:26 I've been fighting this battle for years across multiple companies -- write better error messages. :-( 19:34:01 the difference is a bit subtle, but they didn't bother even justifying ressurecting expert systems, which died decades ago 19:36:35 even at current $dayjob, we're trying to "find ways to reduce SRE toil" and it's like... we know exactly how to do that: reduce complexity, plan system design better, include SRE earlier in the process... but no, we're looking at AI 19:38:09 to be clear, the notion of modern expert systems is just weird. the name is still recognisable and seems to be an easy one to resurrect with all the money sloshing around. it died for a reason 19:38:21 100% agree duncan 19:38:42 it is a very, very strange bandwagon to try to jump on. it was notarious 19:38:49 notorious 19:41:12 half the reason it died was the cost of getting a decent capture of the knowlege. One *might* assume that the capture done in automated fashion will work better, cheaper... 19:42:11 the whole point was quality of knowledge, not volume. so, no 19:42:14 the other half is that large parts of what they were trying to capture were simply not covered by a deductive system 19:43:02 and that no volume of knowledge would remedy that (compared to the performance of an inductive rather than deductive system) 19:43:31 although, amusingly the same chutzpah driving it drove expert systems, so maybe this was all inevitable. 20:05:44 i guess we'd have to ask the author what he meant by "simulate", and if his intention corresponds to duncan's concept of expert system 20:07:06 overall i'd think there would be a lot to be gained from equipping automation with AI capabilities, to shift people's attention toward more meaningful/important problems (and away from highly automate-able work) 20:07:27 especially in a setting like freebsd where there appears to be a bonafide shortage of labor 20:07:32 freebsd project i mean 20:09:22 (or if not automation, then just something a bit beyond an advanced search engine, that seems to be closer to what that paper talks about) 20:20:27 this isn't a paper about expert systems, but a paper about information retrieval 20:20:42 onus is on the author to write clearly.. 20:37:57 Hey, got a weird issue with bluetooth speakers and virtual_oss. When I run virtual_oss, no error occurs, speakers show connection for a split second and then back to no connection, I can hear no audio through them. Tried most of the stuff available on the web but to no avail 20:39:00 actually it's not a bluetooth speaker, but a stereo, but I think it should work the same. I can connect to it via bluetooth without problem, only that virtual_oss link seems to be dropped immediately. The program continues to run until I kill it, but no audio 20:40:49 if I connect other device to the stereo, it shows "BT", but with freebsd connection, it shows "BT" for a split second and back to "NO BT". The LED is not blinking until I kill the program though, and then starts blinking again