00:00:00 --- log: started forth/20.04.30 00:15:16 --- quit: rdrop-exit (Ping timeout: 240 seconds) 00:20:34 --- join: rdrop-exit joined #forth 00:41:55 --- join: dys joined #forth 01:12:47 --- quit: reepca` (Read error: Connection reset by peer) 01:12:59 --- join: reepca`` joined #forth 01:56:56 --- quit: jimt[m] (Quit: killed) 01:57:03 --- quit: siraben (Quit: killed) 02:01:05 --- quit: rdrop-exit (Ping timeout: 260 seconds) 02:09:24 --- join: jimt[m] joined #forth 02:11:02 --- quit: mtsd (Quit: mtsd) 02:14:08 --- join: mtsd joined #forth 02:16:24 --- join: rdrop-exit joined #forth 02:18:31 --- quit: mtsd (Client Quit) 02:21:25 --- join: siraben joined #forth 02:22:28 --- quit: rdrop-exit (Ping timeout: 240 seconds) 02:22:38 --- join: mtsd joined #forth 02:28:09 --- join: rdrop-exit joined #forth 02:32:31 --- quit: rdrop-exit (Ping timeout: 246 seconds) 02:32:54 --- join: rdrop-exit joined #forth 02:52:08 --- quit: rdrop-exit (Ping timeout: 240 seconds) 03:23:16 --- join: dddddd joined #forth 03:24:30 --- join: TCZ joined #forth 03:25:24 --- join: rdrop-exit joined #forth 03:27:31 tp: hashforth builds and runs on freebsd and openbsd if using `gmake` and `clang` 03:34:49 --- quit: dys (Ping timeout: 260 seconds) 03:37:06 --- quit: mtsd (Quit: mtsd) 03:42:20 --- join: mtsd joined #forth 03:46:55 crc, thanks, Im certain thats what I used last time but I'll check it again, now you mention it 03:53:39 crc, hmm that was easy :) 03:53:57 all working now, I cant remember what went wrong last time 03:59:39 --- quit: rdrop-exit (Read error: Connection reset by peer) 03:59:45 --- join: rdrop-ex1t joined #forth 04:14:51 --- join: dys joined #forth 04:20:13 --- quit: dave0 (Read error: Connection reset by peer) 04:25:27 --- join: dave0 joined #forth 04:54:16 --- quit: dave0 (Quit: dave's not here) 05:08:01 --- quit: mtsd (Quit: mtsd) 05:18:39 --- quit: rdrop-ex1t (Quit: Lost terminal) 05:19:57 --- join: mark4 joined #forth 06:10:12 --- quit: MrMobius (Read error: Connection reset by peer) 06:13:41 --- join: MrMobius joined #forth 06:30:52 --- quit: mark4 (Quit: Leaving) 07:07:08 --- quit: iyzsong (Ping timeout: 240 seconds) 07:22:09 --- quit: tabemann (Ping timeout: 244 seconds) 07:31:02 --- quit: TCZ (Quit: Leaving) 08:08:49 --- quit: xek (Quit: Leaving) 08:25:50 --- quit: djinni (Quit: Leaving) 08:27:51 --- join: djinni joined #forth 08:41:38 tp: Yes, I remember my FreeBSD venture and building from source, very hit and miss it was 08:42:06 I always use GNU Make makefiles, I tried writing POSIX Make stuff but it's just more limited 08:42:13 And FreeBSD people know to try 'gmake' 08:42:20 BSD people in general 08:50:23 I work to make sure my Makefile is compatible with BSD make and gmake. I won't make someone install gmake just to build my system. 08:54:37 --- quit: Zarutian_HTC (Ping timeout: 240 seconds) 09:02:06 crc: I do agree with this in principle 09:02:21 But nobody actually uses my software, if people did I would consider making it more compatible 09:03:09 I tell you what though, needing to install gmake (or just read the makefile and convert it / figure it out) is wayyyyyy less egregious than what most repos I see on github today require 09:03:36 Was helping someone with a project recently and about 50 perl, python etc packages in I was fed up and stopped 09:04:14 Do you use a BSD crc? 09:09:07 --- quit: dys (Ping timeout: 240 seconds) 09:24:09 --- join: Zarutian_HTC joined #forth 09:31:22 I use openbsd on my dev system and freebsd on my main servers 09:37:41 Ah nice OpenBSD 09:41:57 I want to try OpenBSD but I guess I stopped caring about my OS a long time ago, just want whatever's easiest 09:42:03 I'd try a forth system ;-) 10:12:35 retro/native, retro running as the os. http://retro.tunes.org/ 10:13:10 not very complete yet (text display, keyboard, cmos, parallel ata disk support, and a block editor) 10:13:40 but it does work on my test systems (qemu, and an ancient thinkpad), booting via GRUB 10:16:11 or john_cephlapod's impexus (https://github.com/jmf/impexus), which is also retro, but uses an assembly version of the VM instead of the C version. Slightly worse on the driver front at this point, as he's rewriting his assembler to be a bit easier to work with, but I hope to flesh this out once he's done 10:21:01 * crc still has a lot to do on this. It needs partition table parsing, and doesn't support interrupts yet 10:22:28 I should look into UEFI booting and interfacing at some point 11:55:21 --- join: deesix_ joined #forth 11:55:39 --- join: dddddd_ joined #forth 11:57:51 --- quit: dddddd (Ping timeout: 246 seconds) 11:58:25 --- quit: deesix (Ping timeout: 260 seconds) 11:59:18 --- join: dddddd__ joined #forth 12:00:20 --- join: deesix joined #forth 12:01:12 --- quit: dddddd_ (Ping timeout: 264 seconds) 12:01:20 --- quit: deesix_ (Ping timeout: 256 seconds) 12:21:01 crc: Nice, something I might try out 12:22:44 Especially if I end up doing any UEFI stuff at work, and then maybe anyway 12:34:17 --- nick: dddddd__ -> dddddd 13:25:58 --- join: ntry joined #forth 13:37:02 --- quit: ntry (Quit: Leaving) 13:47:37 --- join: TCZ joined #forth 13:53:55 --- quit: dave9 (*.net *.split) 13:53:55 --- quit: WilhelmVonWeiner (*.net *.split) 13:53:55 --- quit: dzho (*.net *.split) 13:54:03 --- join: dzho joined #forth 13:54:05 --- join: WilhelmV1nWeiner joined #forth 14:09:08 --- join: dave9 joined #forth 14:19:28 --- nick: reepca`` -> reepca 14:34:48 "With the huge RAM of modern computers, an operating system is no longer necessary, if it ever was." 14:34:54 What did Chuck Moore mean by this? 14:37:22 What is the sound of one hand clapping? 14:39:29 I read it and thought I understood it, and then read it again later and realised I don't know 14:39:30 I guess Virtual Memory. Look at the critique of Virtual Memory in http://metamodular.com/closos.pdf ! 14:43:07 --- join: dave0 joined #forth 14:46:12 with this amount of memory you don't need a lot of abstractoin simulating coherent memory anymore, I think is what he meant 14:56:33 veltas, I was referring to building Hashforth, I never have *any* trouble building Unix apps on FreeBSD. RETRO built in about 0.1 second here first time ever. 15:08:57 isnt an RETRO vm basically just half a dozen or so .c files max? 15:10:17 as all things should be 15:11:16 Zarutian_HTC, the entire release of RETRO is only 4.2MB inc doc 15:11:27 Zarutian_HTC: generally one or two .c files 15:12:01 I have fiddled with the idea of using something like retro vm for implementing non britle build systems 15:12:38 http://forth.works/retro-unix.c for a single file .c combining the vm and image; a quick `cc -lm -O2 retro-unix.c -o retro` is all you really need to get started 15:12:58 tp: I used FreeBSD for probably a year or two overall and I had issues building a number of projects that had clearly only been tested on Linux, Windows and maybe MacOS if you're lucky. 15:13:35 mainly because x86 will loose its dominance as desktop or tablet or what have you cpu arch 15:13:40 Even ones that had supposed support for FreeBSD often hadn't been tested on FreeBSD in years and had many time consuming flaws to be fixed before building 15:14:51 (arm and risc-v come to mind) 15:16:03 veltas, what do Linux, Windows have in common ? 15:16:09 non POSIX 15:16:30 tp: lots of testers 15:16:56 Zarutian_HTC, lol, yeah but what is the quality of such 'testers' ? 15:17:14 I don't use many "POSIX" programs, people tend to rely on stuff that's post POSIX 15:17:22 I'll take one FreeBSD tester over 500 Linux testers anyday 15:17:58 to me, Linux is now the new windows 15:18:00 tp: in the sense they try your thing and if they dont get it working easily then they complain 15:18:35 Zarutian_HTC, in the sense that Linux has become easier than windows to install, absolutely zero admin skills needed 15:19:13 coupled with the fact that Linux is free, the userbase now includes millions that have zero computer skills 15:19:16 It's weird you care about POSIX so much anyway because it's a *standard*, like Standard Forth, or ISO C, that people can wax lyrical about but that very few people actually understand properly or write conforming code for 15:19:26 yebb plus windows has become more and more obnoxious to try to use 15:19:31 And then even if you do you get to use something that's a lowest common denominator designed by committee 15:20:16 veltas, I have no problem with Linux being non POSIX, Linux is welcome to go in any direction it wants, I have no horse in that race now 15:21:49 I think they're right, I think decisions like not breaking userspace etc were always more important than conforming to standards 15:21:57 veltas, I do like standards, I believe they are essential for what I do and have always done, but remember these are my opinions only, I'm not trying to change the mind of any Linux user 15:23:05 A lot of people know Linus Torvalds has a strong opinion about adding C++ to the kernel, and C programmers like these arguments. But a lot of those C programmers don't know Linus also has no time for "standard C", and instead works with a language that works well for him 15:23:15 veltas: not breaking userspace is a kind of standard. More praxic than declaritive. 15:23:17 And for his system 15:23:18 personally; I see Linux mutating faster and faster these days, way too fast for me, I'm too old for rapid change 15:23:37 Zarutian_HTC: Not on BSD it's not, the FreeBSD mindset was "break userspace if it's 'wrong'" 15:24:01 thats why I abandonded Linux in 2016 after using it non stop as my only OS since 1997 15:24:47 and that includes first trying Linux Yggdrassil in 1994 15:25:07 I just think standards are a tool and people tend to think they're an argument on their own a bit more than they should 15:25:52 Conforming is not an argument on its own, I hope something that any forther will appreciate to an extent lol 15:26:30 i think many people that refer to standards as final answers are unwittingly apealing to authority 15:27:18 veltas, Red Hat programmers have felt to write their own 'cool code' and abandon POSIX, but that kind of thing has fractured the Linux userbase. Many Linux admins fled for friendlier shores such as *bsd after that 15:28:02 myself included 15:28:10 Why is abandoning POSIX a bad thing? 15:28:54 why is it a good thing ? 15:29:09 Are you making an argument? 15:29:19 And why did this become a camel back breaking issue in 2016? 15:30:07 no, I'm not a programmer as you know, I dont like people breaking my OS as it's a tool for me, I only use it to work 15:30:17 veltas: I do not know about POSIX but relentless update for update sake is utterly stupid specially when working with hardware 15:30:22 tp: So Linux broke your stuff? 15:31:02 Linux has broken stuff for me as well, some reasonable some less so. All OSs do though, Linux seems to give me the easiest time of it at the moment 15:31:18 no, Linux and I were becoming estranged in the early 2000's, it took me ages to find a replacement and it was systemd that finally drove me to defect to FreeBSD as a Linux refugee 15:31:18 I'm not loyal, I'll switch immediately if they start making it bad for me 15:32:11 Zarutian_HTC: What stuff are they doing that's an update-for-update's-sake? 15:32:47 Linux rules the world now, I have nothing against it and I'm sure that Linux makes an awesome car radio with minimal boot time, I just have no interest in it as a workstation 15:33:07 tp: oh the ways I hate systemd. It pollutes the search space for a much interesting thing, doesnt start deamons propperly, gets in the way of debugging why s system isnt booting and so on 15:33:42 tp: Yes I also tried FreeBSD, and I think it was partly because I was concerned about directions like systemd. I just found it wasn't worth it, and it wasn't a short stint I have. I just agree with FreeBSD principles a lot less than Linux. 15:33:54 Zarutian_HTC, but for me, my main reason for disliking systemd is the way it was forced onto the Linux userbase 15:34:32 I like some FreeBSD principles, like their idea of roughly "if you can do 90% of features in 10% of the code it's usually the better solution" 15:35:02 veltas, Im not political about this, I like freebsd for it's design, I couldnt care if the designers ate live babies 15:35:10 veltas: one particular linux update that costed my hours was to never enumerate and assign usb devices to /dev/ files in same way change they made 15:35:36 And then there are the arguments I had multiple times with FreeBSD decision makers about stuff in their system implementations that would absolutely be more likely to break code in hidden and annoying ways, and they defended it because code that broke was 'wrong' 15:35:44 tp: I'm not talking politics, just tech 15:36:10 Zarutian_HTC, Linux seemed to fix usb device naming for a while, now it seems broken again 15:36:20 veltas: this was for usb to serial bridges on a machine that has about 50 year service life 15:36:53 veltas, FreeBSD has one MASSIVE advantage over Linux tech in my opinion, namely ZFS 15:37:04 Zarutian_HTC: Yeah Linux updates have broken stuff for me as well 15:37:28 now Linux cannot EVER distribute a bootable ZFS file system, incompatible licenses 15:37:36 Zarutian_HTC: The issue is that OSs always break stuff, but the guidance in Linux makes a lot more sense to me for not breaking it than FreeBSD etc's direction does 15:37:46 veltas: yes which means those updates get turned off permanently 15:38:13 of course that hasnt stopped Linux distros trying every trick they can to get around the ZFS license 15:39:22 veltas: shittly written oses break stuff. And I am talking about the software writing methodology here not the end product 15:39:38 Yes 15:39:52 And FreeBSD convinced me they don't care about this as much as they care about certain other things 15:40:23 And it drove me away. And in my real empirical experience I've had less abrasion working on Linux than FreeBSD as well. Not to say it's perfect, it isn't. 15:40:32 I have a lot of issues with Linux, number one is the bloat 15:40:47 hence why the machine now runs QNX now 15:41:38 oh, yes I agree that linux seems to attract bloat more than FreeBSD 15:42:09 They're better than MS at bloat, but that's not saying much! 15:42:58 Zarutian_HTC, you run QNX on your pc ? 15:43:13 tp: what is so different with the ZFS license? 15:43:19 --- quit: dddddd (Quit: dddddd) 15:43:29 Zarutian_HTC, it's not GPL compatible 15:43:36 it's that simple 15:43:50 being CDDL I think 15:43:51 So you can't use ZFS with Linux? 15:44:08 veltas, of course you can, you can install it yourself 15:44:12 ZFS's main development branch targets Linux 15:44:14 tp: no, on a 'control pc' of a machine that does interesting things with metals 15:44:59 veltas, every distro has a ZFS package 15:45:12 Yes, well, licensing is an unfortunate thing on every OS I use. I am certainly used to extra steps installing proprietary or incompatible software to make e.g. my graphics work 15:45:21 Zarutian_HTC, metals like uranium ? 15:45:29 It's not a deal breaker, I don't really see how it could be either. 15:45:40 tp: I use osx or win7 or that rpi debian. Depending on which 'pc' I am using 15:45:59 veltas, yeah, thats all it is, but it means if you want a bootable ZFS on Linux it's going to be a lot of work 15:46:13 tp: nope metals like iron and blends 15:46:13 Unlicense ftw 15:46:38 Zarutian_HTC, Viking Waraxes! 15:47:12 tp: i suppose that is one of the possible products 15:48:26 unlicense, being a public domain dedication, may not be usable everywhere 15:48:29 --- quit: gravicappa (Ping timeout: 244 seconds) 15:48:52 crc: If it's not then other open source licenses aren't usable in that sense either, right? 15:49:33 what I have noticed regarding eulas and copywrite licenses that most peeps in my industry or relared dont give a damn so long they get the source code so they can fix problems that sometimes have no other ways to be fixed 15:49:36 no; the public domain is not a universal concept 15:49:46 I think it's very sad we live in a world where I can't put text in public and convince people they're not going to get sued for using it 15:50:10 some countries (like germany) don't allow for it 15:50:13 veltas, i guess I'm saying that I came to FreeBSD as a Linux refugee the same way I came to linux as a windows refugee, and in both cases I was hoping the new OS would do what the old one did. In BOTH cases I found the new OS superior in ways I coundnt imagine 15:50:26 crc: nor is the 'AS IS' legal figleaf 15:50:32 and so I stayed 15:51:22 an open source license says that you hold the copyright, but allow others to use and modify and distribute (to varying degrees) the source 15:51:35 crc: germany allows that you put your works in the public domain but you cant abicate your right of authorship. 15:52:04 Zarutian_HTC: I didn't remember the exact details on that 15:52:14 I'm not a lawyer and all this is over my head 15:52:52 the right of authorship is not copyright but to be known (even just psuedonymously) for your works 15:53:10 Frankly it feels like people have ideological excuses for not using these kinds of licenses, because I don't really see how these rigidities of copyright law affect this kind of license and not e.g. MIT 15:53:19 But I am not a lawyer so this is just me being paranoid 15:53:22 * crc stopped using public domain type dedications long ago; less headaches to use a well established open source license instead 15:53:45 crc: Weirdly the only headaches come from people who tell me I should conform and use e.g. MIT 15:53:53 crc, same here 15:54:09 I made the mistake asking a german Pirite Party guy that was really interested in the subject about it. 15:54:54 veltas: if you want to incorporate code from others orhave your code used in products from others, an actual license is helpful 15:55:02 I do have a license 15:55:50 I noticed sqlite managed to run an open source project with a public domainish license 15:55:55 crc, and don't want to spend $millions in court after your product becomes a best seller ? 15:57:20 It feels like the complaint here is "well you can't actually have this licence because the thing you claim is actually something you have the right to renege on in some countries" 15:57:42 Which isn't really an argument, it's just saying "well people might not trust you" okay.... I can handle that 15:57:58 sqlite also requires contributors to submit sigted affidavits of public domain dedications and offers a commercial license for those who may run into legal issues with the status 15:58:01 I'm sure there is actually a legal connotation somewhere.... I don't understand it though, I am not educated in law 15:58:16 crc: I would offer such a license as well, to anyone who asked 15:58:38 Because I don't care, and if someone voluntarily requests a licence that gives me legal power to sue them, then so be it 15:58:55 And if my lawyers permit it I shall also wink at them and tell them "still not going to sue you though" 15:59:20 I have an issue with modern law, I know.... I'm sure most of you don't care. I will stop now 15:59:30 would you get signed affidavits from contributors to provide a legal record of waiving copyright protections? 16:00:02 No but good luck suing me over this 16:00:18 I want to see precedent 16:00:39 Not even GPL gets upheld in court and the terms there are *very* clear, and actually fit copyrights better 16:00:48 * crc does deal with using open source software in commercial settings; I do have tohave licenses reviewed by lawyers: ( 16:01:36 veltas: it's more about you being able to sue the users h]tahn the other way around, in my case 16:02:13 s/h]tahn/than/ 16:02:27 I don't want to sue the users though 16:03:11 what about your descendents? or others whose contributions you use? 16:03:36 I guess I might be oversimplifying but if I ask someone who is a 'lawyer' they are probably a sheep, like if I ask a programmer what a good OS is they will just say "MacOS" or "Linux" with no understanding. 16:03:48 Unless they were an expert I wouldn't really care what they said about a license 16:04:04 And I don't think all people in a profession are experts 16:04:20 But this is just my 'filter' for reasoning here since it's a field I am ignorant of 16:04:21 --- join: X-Scale` joined #forth 16:04:49 * crc doesn't have a choice in this case; when working for a company, their rules matter 16:04:54 Yes I know 16:04:59 I also work for a company 16:06:33 --- quit: X-Scale (Ping timeout: 256 seconds) 16:06:33 --- nick: X-Scale` -> X-Scale 16:06:50 I will say that in general I think corporate culture contains an awful lot of stuff that isn't really productive, correct or consistent 16:06:52 --- quit: TCZ (Quit: Leaving) 16:08:50 So I do not take cues from how people work for how to run my personal ventures, instead I will question what they do the same way I question anything that I distrust 16:12:38 crc: it is frustrating sometimes living in a society where I can't just put something totally innocuous out there without making some liability for someone or myself 16:13:16 I agree 16:14:10 Like a bit of text 16:19:56 This is how I feel https://i.redd.it/mk33aw7fbr531.jpg 16:21:44 tp: Sorry I don't know if it was clear earlier but I don't imagine you are doing the wrong thing using FreeBSD, I assume it works for you 16:21:47 It just didn't for me 16:23:11 veltas, no problems :0 16:24:56 ; 16:25:08 Just finishing that retro word for you there 16:45:24 --- join: TCZ joined #forth 17:00:26 --- quit: cantstanya (Remote host closed the connection) 17:03:24 --- join: cantstanya joined #forth 17:08:38 --- join: rdrop-exit joined #forth 17:27:18 --- join: WickedShell joined #forth 17:32:11 (-_-)zzz c[] good morning Fortharians 17:33:24 --- quit: TCZ (Quit: Leaving) 17:35:39 --- join: karswell_ joined #forth 17:38:00 hi rdrop-exit 17:38:14 hi crc! 17:44:12 --- nick: karswell_ -> karswell 17:53:40 --- join: iyzsong joined #forth 18:06:09 --- quit: iyzsong (Ping timeout: 260 seconds) 18:06:50 --- quit: Zarutian_HTC (Ping timeout: 256 seconds) 18:09:04 --- join: Zarutian_HTC joined #forth 18:15:08 --- join: TCZ joined #forth 18:44:41 --- join: boru` joined #forth 18:44:44 --- quit: boru (Disconnected by services) 18:44:46 --- nick: boru` -> boru 18:53:34 g'day rdrop-exit 18:53:57 all is well on the Forthen front 18:56:12 excellent Major tp, jolly good 19:01:58 --- quit: TCZ (Quit: Leaving) 19:07:53 --- join: corn joined #forth 19:08:22 good evening, #forth fanatics 19:08:31 hi corn 19:08:53 how's your morning 19:09:46 good, relaxing with a cup of coffee and some books 19:18:45 I try to stay away from coffee 19:21:40 "One doesn't write programs in Forth, Forth is the program." -- Chuck Moore 19:24:08 --- join: SandboxGeneral joined #forth 19:28:39 When/where did he say that? 19:28:56 "... in those days software was exciting: things being done for the first time; programmers working 18 hour days for the joy of it. Now programming is a 9-5 job as part of a team working to a schedule; not much fun." -- Chuck Moore 19:29:58 corn, in the interview that's part of O'Reilly's Masterminds of Programming book 19:30:47 The interview has two parts, hardware and software, I believe a pdf of the software half is available online. 19:30:56 IIRC 19:35:38 Ah I think that's familiar 19:41:48 --- join: tabemann joined #forth 19:42:51 "The ability to do anything and not the ability to do everything." 19:45:03 cool 19:48:59 hey 19:49:08 hi tabemann 19:53:36 "More important than the compiler is the editor." 19:54:52 actually, I misremebered, the interview doesn't have two halves, it has various sections, including one on hardware 19:57:20 "Comments do not substitute for proper documentation. A document must be written that explains in prose the code module of interest. It should expand greatly the comments and concentrate on literate and complete explanation." 19:58:16 I do think literate programming is a suitable replacement for documentation, I am a big fan of literate programming 20:03:15 --- quit: dave9 (Quit: dave's not here) 20:03:50 --- join: dave9 joined #forth 20:06:29 I wouldn't call anything a "replacement for documentation", if done right it should result in proper documentation. In the case of Knuth's "TeX: The Program" book he certainly did produce quality documentation using literate programming. 20:07:09 --- quit: reepca (Read error: Connection reset by peer) 20:07:32 --- join: reepca joined #forth 20:08:37 This one reminds me of IBM's mainframe strategy in their heyday: 20:08:59 --- join: kori joined #forth 20:08:59 --- quit: kori (Changing host) 20:08:59 --- join: kori joined #forth 20:09:00 "sell hardware and give away the software" 20:09:33 literate programming can be a replacement for "A document ... that explains in prose the code module of interest" because it includes, in prose, the module of interest, without a separate document. If it's replacing the important part(the fact that the module is explained in prose) then unimportant parts like documentation being separate can be reasonably forgotten 20:10:52 It replaces a document, it doesn't replace documentation, it either results in proper documentation if done well, or not if done badly. 20:11:53 I think your conflating document and documentation. 20:11:58 * you're 20:12:42 --- part: SandboxGeneral left #forth 20:13:08 I was referring specifically to Chuck's explanation of what proper documentation requires, and his explanation called for a document to contain the documentation 20:13:23 I'm just saying the second document is not needed 20:15:08 Ah ok, I get your point and agree. 20:16:41 Done with that article, on to chose something else from my reading pile (mountain). 20:16:44 I wonder if there's a collection of interviews/publications/etc of Chuck's, it's always a good read when I find ones I haven't seen before 20:18:25 Not a complete one, but Jeff Fox's (RIP) had lots on his ultratechnology site. 20:20:01 Some of what he has are transcripts, some are his Jeff's interpretation of the "Forth way". 20:21:38 Jeff's writing style was all over the place though 20:22:52 I remember him putting a lot of (often bad) humor into stuff he wrote 20:24:05 His writing was emotional and repetitive, but there were nuggets of good food for thought in some of his writings 20:24:32 Ah, like on his bio page, right away, he says he was "raised by foxes," due to his last name 20:25:49 Michael Misamore produced a more coherent summary of Jeff's Thoughtful Programming writings 20:27:20 Here's a copy I just Googled: 20:27:27 http://www.patorjk.com/programming/articles/forththoughts.htm 20:40:03 --- quit: proteus-guy (Ping timeout: 246 seconds) 20:47:38 Forth provides a way of creating non-trivial systems that one, as an individual, can understand the whole of, top to bottom 20:48:02 everything else is too complex to fully understand or too simple to provide sufficient power to express what one needs 20:48:46 like in my Haskell phase there was no way that I, as an individual, could really understand what GHC was doing 20:51:08 I'm not sure that you can say the second statement 20:52:31 I think that statement is bogus, but the other two I agree with 20:54:19 "The goal was very simple: to minimize the complexity of the hardware software combination. As far as I can see no-one else is doing that." -- Chuck Moore 20:56:22 I wonder how Chuck navigates the definition of complexity 20:56:31 because that is itself quite a challenge 20:58:13 I choose to just embrace what feels right and reject what feels wrong, and use that is my separator between simple and complicated, but my approach is not very objective 20:59:56 I've never met a compiler for any other language that I truly understood how it worked 21:00:56 for the longest time I treated compilers as black magic 21:01:28 but "never met" does not mean it can't or doesn't exist 21:01:58 I think plenty of paradigms, when reduced to their essentials, could be implemented in an entirely understandable way 21:04:14 It's more complicated than a forth, but I think that a solid implementation of smalltalk could fit that bill 21:04:15 well, there are some languages that might be comparable to a degree 21:04:49 such as a very simple implementation of Scheme with a very simple garbage collector 21:04:56 but even that is more complex than Forth 21:05:15 you lost me at garbage collector :) 21:05:24 exactly 21:06:02 why quit the moment you hear garbage collector? 21:06:23 garbage collector does not somehow imply complication 21:06:37 because GC's result in non-deterministic behavior 21:06:59 Is the GC optional? 21:07:07 non-deterministic also does not imply complicated 21:07:31 rdrop-exit: the alternative is allocating memory until you crash from running out of memory 21:07:48 then don't 21:09:44 It's one thing having a GC because your application inherently requires one, it's another to have it because your language imposes it on you 21:10:14 * tabemann tried writing a scheme - he quickly gave up because even such a simple language as scheme was too complex for him to bother 21:10:20 languages will always impose something on the author of a program, that's inevitable 21:11:05 rdrop-exit: I think the key factor with scheme is whether numbers are boxed or not 21:11:35 and Forth imposes almost nothing that I wouldn't want, i.e. 2 stacks 21:11:39 if numbers are unboxed, you can theoretically write a scheme program that runs in constant memory space without GC 21:12:17 the "that I wouldn't want" portion of your statement is nearly arbitrary 21:12:40 of course not 21:13:01 and I could reasonably have a different conception of what is unwanted, so other languages could reasonably fit into the accepted category 21:13:30 it's almost impossible to get away from having one stack in a programming environment - a second stack is a bonus, because it allows you to treat some things (data) separately from other things (return addresses) 21:14:22 not impossible, of course 21:14:45 since when programming in a language without subroutine calls stacks are not necessary 21:15:15 but no subroutine calls are such a restriction that I can't see any way it would be desirable 21:21:34 I think another important thing to take into consideration is that hardware affects what software is simple, and we usually deal with register/stack machines that forth will appear well suited for 21:25:30 but register machines are honestly quite complicated things in comparison to some other machines 21:25:36 That's why Chuck focuses on the combination of hardware and software 21:28:41 Chuck doesn't just focus on surface simplicity, he's after simplicity all the way down 21:28:54 And one of the biggest reasons I wanted to disagree with tabemann's statement above is the importance of the combination, and that I find it highly unlikely that forth is the ONLY combination of hardware and software with both reasonable hardware and software 21:30:19 and you likely won't have seen many examples that fit the criteria, because me, tabemann, everybody is doing a LOT more experimenting with radically different software architectures than with radically different hardware architectures 21:30:30 to me at least one can have simpler systems, but they all come at the cost of expressiveness and power 21:31:52 I think complicated problems like this tend to have multiple simple solutions if they have any 21:34:33 I suspect that a cleverly concocted automata based language could have similar complexity and expressive power to forth 21:39:32 hey tabemann, news CRC mentioned that Hashforth compiled fine on his openbsd so I had another go building Hashforth, Gmake worked perfectly first time, dunno what I did wrong last time, I expect I didnt read the notes and just executed hashforth saw 'no image' and moved on 21:40:02 tabemann, so hashforth builds and runs easily on freebsd 21:40:23 cool 21:41:23 tabemann, and your tip on adding swap after the 'fall thru' in CASE fixed a few problems I had in my code :) 21:41:33 cool as well 21:41:44 tp: I watched one of your YouTube videos, you have the voice of an Australian pilot 21:42:12 what I really think is there should be a DEFAULT word for handling fall through in CASE 21:42:28 tabemann, using Hashforth without a prompt of dictionary is a bit weird for this poor Forth user 21:42:30 because the standard way of doing is frankly quite ugly 21:42:52 well I was lost by the fall thru problem for years 21:43:35 tp: I should add something to hashforth to make it automatically look for the image in certain locations 21:43:53 lol, so I add the fall thru message and what did I get ? 21:43:59 "jump too far" 21:44:35 to me when zeptoforth complains about too far of a jump what it really means is that the compile-time stack is corrupted 21:44:39 putting the fall thru stuff in another word was a good idea as I use it 4 times elsewhere in this app anyway 21:45:05 tabemann, because you use M4 ? 21:45:13 yeah 21:45:52 well, I should get to bed 21:45:55 corn, I hope you find what you're looking for, as Chuck says "show me a simpler way to do anything that I'm doing, I will jump on it". 21:45:55 I only use cortex-m0 as you know. tho I need to alter my M3 code as well 21:46:13 but cool that you got hashforth working 21:46:14 night-o tabemann 21:46:35 well I got a lotta years left to look for it 21:46:36 goodnight tabemann 21:46:50 rdrop-exit, nice notes, agree entirely, this forther loves simplicity 21:53:20 cool 21:53:28 "It must be wonderful to be seventeen, and to know everything." -- Arthur C. Clarke 21:54:50 thats not true, I have a sister who also knows everything and she is in her 70's 21:56:01 she probably has known everything since she was 17 21:56:19 probably before then 21:56:42 I think ghod gave her this special gift 21:57:08 I've raised four like that 21:58:14 lunch is ready, catch you later, stay healthy :) 21:58:22 --- quit: rdrop-exit (Quit: Lost terminal) 21:58:32 cya 22:14:59 --- quit: karswell (Remote host closed the connection) 22:16:25 --- join: karswell joined #forth 23:06:22 --- join: gravicappa joined #forth 23:46:47 My fall through message needs to be "aim for the bushes" 23:59:59 --- log: ended forth/20.04.30