00:00:00 --- log: started forth/19.11.04 00:00:00 right, there's no appetite for dealing with engineering tradeoffs 00:02:25 at least I wont have to worry about systemd being fitted to Forth ;-) 00:02:33 They want a tool were they can "express" the solution in some abstract way, rather than engineering a solution and deling with multi-level tradeoffs. 00:02:57 * dealing with 00:05:13 tp: Actually, I was planning to do just that 00:05:40 In Forth you choose the platform (i.e. what's underneath) and you are free to make whatever tradeoffs going up from there as long as you can communicate via the external standards that matter to your solution. 00:06:13 Jookia, include systemd in your Forth ? 00:06:49 tp: Support a few of its features 00:07:12 flamewargeneratord 00:07:18 There's a lot to be said about the good things in systemd 00:07:39 I like its service file format 00:07:43 but that's about it 00:08:08 jackdaniel, lol! 00:08:53 Yeah it's service file format is good, especially with how it handles setting up services without spawning a gazillion subshells because you're using shell scripts 00:09:05 Luckily, my Forth is *all* there is, there is no OS, nothing controlling it except me 00:10:11 (tp rubbing his hands together and laughing maniacally) 00:12:13 systemd also has ways to talk to the supervisor from a process and indicate status, such as 'started successfully' or 'still alive' 00:12:18 also structured logging is pretty good 00:12:56 the APIs for these are fairly clean 00:14:40 * jackdaniel puts his snarky remarks about systemd in his pocket and goes to make a tea. 00:17:45 yeah i have snarky remarks about it 00:18:08 but you have to appreciate good things when you see them. then steal them 00:18:32 who knows, maybe i'll write an init system in forth some day 00:19:13 I became a Linux refugee because of systemd, or at least thet was the catalyst, 3 years ago 00:19:25 Ive been happily using FreeBSD since then 00:28:41 i tried to use freebsd but it has poor driver support 00:28:57 there's also non-systemd distros 00:29:34 Devuan is actually pretty useable 00:29:45 being completely systemd free 00:30:28 though freebsd is probably better architected than linux :P 00:32:20 FreeBSD may have 'poor driver support' but it has great ZFS and FreeBSD Jail support :) 00:32:41 Im running a Nvidia GTX660 or something on this as I type 00:33:05 i wouldn't say great re ZFS- it only works on x86 00:33:16 I want to try NetBSD, because of the Rump kernel 00:35:05 i haven't looked at that yet, but i'm heavily interested in having my forth run on a microkernel. or even just do its own IO using message passing 00:35:44 --- join: TonySidaway joined #forth 00:35:56 https://en.wikipedia.org/wiki/Rump_kernel 00:36:26 weren't it able to run on any host (i.e on linux)? 00:36:36 --- quit: TonySidaway (Client Quit) 00:36:39 yup, that's it 00:37:17 rdrop-exit, Im not interested in non x86 on my desktop 00:37:20 oops 00:37:24 The rump kernels can also run without POSIX directly on top of the Xen hypervisor, an L4 microkernel using the Genode OS Framework[7] or even on *"OS-less" bare metal*. 00:37:28 Jookia, Im not interested in non x86 on my desktop 00:38:24 tp: yeah intel rules the desktop market, but it also means you can't use ZFS on ARM boards for things like NAS 00:38:32 The "OS-less" bare metal part is intriguing. 00:39:17 btrfs to the rescue I suppose, it got very stable over the years 00:39:52 everyone's kind of waiting on bcachefs to replace btrfs 00:39:59 since btrfs has some design flaws 00:40:54 Jookia, thats cool, my NAS is dual Intel Atom :) 00:42:01 jackdaniel, btrfs is coming along true, but it needs to be proven in the wild for a decade to be even where ZFS was 10 years ago 00:42:42 I've heard some praising words about zfs, so I'm not going to express doubt on that - mind though that Linux has much bigger volume of users 00:42:50 so the "wild testing" phase will take much less 00:42:58 (more eyeballs) 00:43:09 jackdaniel, absolutely, Linux is the 100 ton gorilla, no argument from me 00:43:40 so my point is that proving it will take less time than it did with zfs 00:43:52 that said, I'm satisfied with my .. (/me types mount) 00:44:12 ext4 filesystem 00:44:13 ;) 00:44:14 jackdaniel, I ran Linux full time 1997 to 2016, thats 19 years, I know it pretty well 00:45:30 jackdaniel, Linux definitely needs a "zfs" class of filesystem because EXTx is very basic in comparison and Linux users are really missing out 00:46:11 there is no way that I could ever go without ZFS now, it's the most critical part of this workstation 00:46:21 I'm not denying that, I've mentioned that I use ext4 to hint, that I'm not a good authority to discuss filesystems and all my remarks end with the abovementioned opinions 00:47:02 I have a friend who is a "true" btrfs believer though ,) 00:47:08 jackdaniel, no worries, I'm a forther before anything, a electronics tech by profession and I just use pc's and OS's to get my work done 00:47:36 Im happy using any Unix, just dont bring a windows box near me 00:47:39 btrfs is fairly good 00:48:25 --- quit: ryke (Ping timeout: 240 seconds) 00:48:25 Jookia, does btrfs run on arm ? 00:48:56 yeah, that's what i'm using atm 00:49:03 it does (both arm and aarch64) 00:49:05 jackdaniel, you can easily install zfs on Linux, it's just a tribulation to make a Linux box zfs raid bootable 00:49:29 sure, but I've heard that there is some licensing mess which can't be fixed 00:49:51 yeah, zfs is CDDL licensed and thats not GPL compatible 00:50:02 oracle being oracle isn't going to fix that 00:50:07 but then ZFS was made by SUN 00:50:19 that is quote off given how bsd's praise their liberal licensing 00:50:23 quite 00:50:26 sure, may oracle burn in the pits of whatever 00:51:19 well openbsd wont allow zfs due to the cddl, so not all *bsds are that liberal 00:51:40 but I'm not giving up ZFS 00:54:26 I mean it the other way around, openbsd seems to be more liberal (with things you may take from it) than others 00:56:09 I really like openbsd, but they insist on 100% open 00:56:23 and that denies me zfs unless I install it myself 00:57:16 but openbsd is so minimalist, clean, neat and tidy 00:58:24 I had some bad experience with the community, but it might be that I was asking the wrong questions 01:04:15 wuth the openbsd community ? 01:04:49 i remember going there to tell them I thought it was great and they nearly banned me, lol 01:05:00 yes, with them 01:05:37 openbsd makes openbsd for it's own developers, and you can use it, but if you dont like it, they dont care, if you criticise it, they ban you 01:05:56 actually, I like that atittude 01:06:00 zfs is 100% open 01:06:06 oh 01:06:09 well, it's copyleft 01:06:33 zfs is CDDL licensed 01:08:04 yeah, which is an open source license 01:08:22 agreed 01:08:31 it's just not compatible with the GPL 01:10:19 isn't zfs 100% open? 01:10:23 or am i misunderstanding 01:10:57 I'm not a lawyer, both the GPL and CDDL are open, but theyre not compatible 01:11:02 Jookia: it's released under CDDL, which is a weak copyleft license. The problem is that the terms of the CDDL and the GPL, while both "open" and "free" are incompatible with another, and the GPL doesn't like code to be mixed with licenses that have incompatible terms 01:11:46 the CDDL doesn't mind as much, but such unions need to be consensual ;-) 01:12:40 thanks patrickg :) 01:14:30 Specifically, the CDDL has two clauses that are disliked on the GPL side: First, there's a patent grant thingy (that would be compatible with the GPLv3 because it grew the same clause, but not the GPLv2 which didn't account for patents being a big deal). Second, there's a "choice of venue" section which states where in the world a lawsuit over the code would have to happen. The GPL (any version) doesn't ha 01:14:30 ve that. 01:14:41 As soon as a license has terms that go beyond what the GPL says, the GPL doesn't allow mixing 01:36:47 i'm not sure if its gplv3 specific but the GPL is ok with additional permissions, so the patent grant shouldn't be an issue 01:41:49 --- join: dys joined #forth 02:05:19 Jookia, youll find that Linux has decided already, thats why theyre working on btrfs and why no Linux distro offers ZFS as a installation filesystem in any distro 02:06:13 sure you can install zfs after the Linux distro is already installed from a repo, but zfs isnt available as a alternative on the distro install 02:06:55 I don't think Linux decides things like that 02:07:41 sure they do, name one linus distro that includes ZFS as a installable option ? 02:07:45 Linux 02:09:09 if it wasn't a issue, they would ALL have it as a install-able option, they want it bad 02:09:22 name one distro that includes btrfs as an installable option 02:10:29 hahah, why are you comparing them ? ZFS has been proven in the field the last decade at least, btrfs is not ready for production yet 02:11:00 i think we should move on to another topic 02:11:02 no one wants to include a alpha file system, the cost is way too high as regards lost data 02:11:20 sure, this is a Forth channel anyway 02:11:35 apologies to forthers here for the noise 02:29:34 --- part: Jookia left #forth 02:39:06 --- join: iyzsong joined #forth 02:45:52 --- join: xek_ joined #forth 02:46:22 --- quit: xek_ (Remote host closed the connection) 02:48:31 --- join: xek joined #forth 02:50:22 tp: fwiw latest Ubuntu has zfs as installer option. But yeah, let's keep this channel on topic :-) 02:51:38 patrickg, I'll put my best foot forth! 02:58:25 random snippet of the day: 02:58:43 0 shadow Arithmetic - Logarithms 02:58:43 1 02:58:43 2 |log2 Floor of binary logarithm. *** N.B. |log2 of 0 is -1. *** 02:58:43 3 02:58:43 4 log2| Ceiling of binary logarithm. *** N.B. log2| of 0 is 0. *** 02:58:45 5 02:59:12 These are for a 64-bit Forth 02:59:23 : |log2 ( u -- n ) #lead-0s 63 \- ;inline 02:59:32 : log2| ( u -- u' ) dup 0<> + #lead-0s 64 \- ;inline 03:00:20 My contribution to on-topicness for the day 03:01:04 dinner time, catch you all later 03:01:29 cya! 03:02:02 cant be more on topic than that! 03:02:08 --- join: dave0 joined #forth 03:02:39 --- quit: proteus-guy (Ping timeout: 240 seconds) 03:36:15 --- quit: xek (Remote host closed the connection) 03:37:04 --- join: xek joined #forth 03:56:35 --- join: f-a joined #forth 04:22:42 --- quit: xek (Read error: Connection reset by peer) 04:25:49 --- join: xek joined #forth 04:29:21 --- quit: xek (Read error: Connection reset by peer) 04:29:46 --- join: xek joined #forth 06:05:24 --- quit: iyzsong (Quit: ZNC 1.7.1 - https://znc.in) 06:45:12 --- quit: f-a (Quit: leaving) 06:51:14 --- join: dddddd joined #forth 06:56:26 --- quit: xek (Remote host closed the connection) 06:58:41 --- join: xek joined #forth 07:21:18 --- join: proteus-guy joined #forth 07:29:14 --- quit: proteus-guy (Ping timeout: 264 seconds) 07:40:02 --- quit: tabemann (Ping timeout: 264 seconds) 07:47:00 --- quit: dave0 (Quit: dave's not here) 08:08:46 --- join: f-a joined #forth 08:15:16 --- quit: xek (Remote host closed the connection) 08:15:35 --- join: xek joined #forth 08:29:29 --- quit: cp (Quit: Disappeared in a puff of smoke) 08:30:20 --- join: cp joined #forth 09:06:46 --- join: ryke joined #forth 09:13:39 --- quit: ryke (Ping timeout: 240 seconds) 09:27:29 --- join: TonySidaway joined #forth 09:48:21 --- quit: f-a (Read error: Connection reset by peer) 09:52:02 --- quit: rdrop-exit (Quit: Lost terminal) 09:52:34 --- join: f-a joined #forth 10:16:25 --- join: WickedShell joined #forth 10:20:49 Really getting into tmux. tp, I do understand why you want your terminal mux to also be a serial terminal. I do think the tmux developer is being sensible in trying to avoid function creep. 10:22:16 tmux is really useful 10:22:53 There are plenty of serial terminal applications. They have to worry about stop bits and parity and a whole heap of serial cruft going back 50 years. 10:23:49 Can't blame a modern developer for thinking twice before diving into that swamp. 10:30:58 My development workflow cycle is that typically I'll peck away at stuff using a ssh client from my phone. If I need a rest (which is very often) it's great to be able to detach the tmux session and come back to it later when I've refreshed myself and thought through my immediate goals. 10:31:48 Or after I've spent an hour or two slacking on #forth :) 10:33:40 --- part: john_cephalopoda left #forth 10:34:26 When I started telling people I use my smartphone as my main computer terminal I was met with disbelief. Does work, though, if you're an ancient retired person with no deadlines. 10:34:43 well 10:34:52 I guess if you have 20/20 vision :P 10:34:59 I'll wait for a pyra dragonbox 10:35:04 can't stay w/o a keyboard 10:37:14 f-a, far from it! I'm extremely myopic. But that's an advantage. Without my lenses or varifocals my eyesight is excellent from 10cm to 20cm. I just hold the phone close to my face and I can see everything on the phone screen. 10:38:02 With my lenses or glasses I'm rather longsighted and I would have to squint a lot! 10:38:09 TonySidaway: how do you type 10:40:08 I use gboard for everything. My ssh client on Android handles a lot of the keyboard stuff. It does Ctrl/Alt/Esc and ask the usual PC keys like PgUp, PgDn, arrows, etc. 10:41:14 I can use all the Vi and Emacs key bindings easily. 10:47:17 A laptop is better but my trouser pockets are too small to hold one. 10:47:57 hehe 10:48:48 https://pyra-handheld.com/boards/pages/pyra/ 10:48:54 not out yet! 10:52:36 Those keys must be able the same size as a TI calculator, or smaller. My screen resolution is much better than the Pyra's, and I run a quite inexpensive mobile phone (Moto G5S, second hand). 10:54:19 --- quit: xek (Ping timeout: 240 seconds) 10:58:31 If that system ran Windows and supported lots of PC games, it might be potentially successful. If it had a faster processor and ten times more memory. 10:59:38 Otherwise I think they're aiming for a very small geek market, and missing by a mile. 11:04:23 Wikipedia talks about Pyra being up for preorders in May 2016. At this rate they'll be able to bundle Half-Life 3 with the first units delivered. 11:04:34 --- join: ryke joined #forth 11:12:49 --- join: ryke1 joined #forth 11:12:58 --- quit: ryke (Remote host closed the connection) 11:12:58 --- nick: ryke1 -> ryke 11:26:08 --- quit: dys (Ping timeout: 250 seconds) 11:46:29 --- join: jedb_ joined #forth 11:48:40 --- quit: jedb (Ping timeout: 240 seconds) 12:12:21 --- quit: gravicappa (Ping timeout: 245 seconds) 12:43:30 --- quit: f-a (Quit: leaving) 14:00:21 TonySidaway I looked at TMUX, sadly there is no way it can do what I need. Tmux may improve some screen features but it is missing many others that I must have 14:01:22 --- quit: TonySidaway (Quit: TonySidaway) 14:01:39 --- join: TonySidaway joined #forth 14:01:39 --- quit: TonySidaway (Client Quit) 14:01:53 --- join: TonySidaway joined #forth 14:02:07 --- quit: TonySidaway (Client Quit) 14:02:15 fortunately screen does what I need for a forth serial terminal so I don't need to use anything else 14:03:04 --- join: dys joined #forth 14:06:45 --- join: TonySidaway joined #forth 14:06:45 --- quit: TonySidaway (Client Quit) 14:06:59 --- join: TonySidaway joined #forth 14:07:13 --- quit: TonySidaway (Client Quit) 14:29:27 --- join: TonySidaway joined #forth 14:29:27 --- quit: TonySidaway (Client Quit) 14:29:41 --- join: TonySidaway joined #forth 14:29:55 --- quit: TonySidaway (Client Quit) 15:04:44 --- join: TonySidaway joined #forth 15:16:24 I think this is about it for TMUX versus Screen: https://www.techrepublic.com/blog/linux-and-open-source/is-tmux-the-gnu-screen-killer/ 15:16:59 And yeah, it's missing some features you may require. 15:17:17 tmux is a subset of screen overall, that's why I cant use it 15:19:22 i definitely don't see tmux as any kind of 'screen killer' no matter the hype 15:28:25 --- quit: ryke (Ping timeout: 240 seconds) 15:32:58 --- quit: TonySidaway (Quit: TonySidaway) 15:40:47 --- join: dave0 joined #forth 16:01:43 --- quit: remexre (Quit: WeeChat 2.5) 16:02:03 --- join: remexre joined #forth 16:19:24 --- quit: pointfree (*.net *.split) 16:24:39 --- join: pointfree joined #forth 16:39:27 --- join: Jookia joined #forth 16:49:16 --- quit: pointfree (*.net *.split) 16:54:43 --- join: pointfree joined #forth 17:13:16 --- join: tabemann joined #forth 17:18:31 --- quit: dddddd (Remote host closed the connection) 17:36:04 --- quit: dave0 (Quit: dave's not here) 17:49:07 --- quit: pointfree (*.net *.split) 17:54:51 --- join: pointfree joined #forth 18:13:22 --- quit: tabemann (Ping timeout: 246 seconds) 18:17:03 --- join: karswell joined #forth 18:42:40 --- join: f-a joined #forth 18:49:12 --- quit: pointfree (*.net *.split) 18:50:36 --- join: ryke joined #forth 18:54:18 --- join: pointfree joined #forth 19:02:01 --- quit: proteusguy (Ping timeout: 240 seconds) 19:15:19 --- quit: cartwright (Ping timeout: 260 seconds) 19:19:46 --- join: cartwright joined #forth 19:57:22 --- join: tabemann joined #forth 20:04:46 --- part: tabemann left #forth 20:11:53 --- join: tabemann joined #forth 20:40:52 --- join: jedb__ joined #forth 20:43:35 --- quit: jedb_ (Ping timeout: 268 seconds) 20:52:36 --- quit: cartwright (Remote host closed the connection) 20:55:01 --- join: cartwright joined #forth 21:05:10 --- quit: dys (Ping timeout: 268 seconds) 21:25:25 --- join: gravicappa joined #forth 21:45:33 --- join: dys joined #forth 21:49:07 --- quit: pointfree (*.net *.split) 21:49:55 --- join: pointfree joined #forth 22:27:39 --- quit: f-a (Quit: leaving) 22:31:38 --- quit: jackdaniel (Remote host closed the connection) 22:38:25 --- join: [2]MrMobius joined #forth 22:38:28 --- quit: MrMobius (Ping timeout: 252 seconds) 22:55:16 --- quit: karswell (Remote host closed the connection) 22:55:49 --- join: X-Scale` joined #forth 22:56:34 --- join: karswell joined #forth 22:57:31 --- quit: X-Scale (Ping timeout: 276 seconds) 22:57:32 --- nick: X-Scale` -> X-Scale 23:41:59 --- quit: dys (Ping timeout: 240 seconds) 23:58:54 --- join: X-Scale` joined #forth 23:59:30 --- quit: X-Scale (Ping timeout: 252 seconds) 23:59:43 --- nick: X-Scale` -> X-Scale 23:59:59 --- log: ended forth/19.11.04