2026-04-10 11:29:43 tpbsd: Am I allowed to hate autotools? 2026-04-10 12:50:35 veltas, I've no idea Ive never used them myself 2026-04-10 12:53:48 veltas, Im a big Make fan, but only use the simpler parts of it 2026-04-10 12:54:28 veltas, and I dont write anything ib C code now, and havent for years 2026-04-10 12:54:34 Amen. Make was just fine. 2026-04-10 12:54:35 ib = in 2026-04-10 12:55:30 I think the later tools were created more in the interest of the project developers than the user community. 2026-04-10 12:57:20 Make was simple and intuitive and easy to understand. If something didn't make right away I could usually tweak the makefile to make it work. 2026-04-10 12:57:44 but I really have no idea what's going on with all the new stuff. 2026-04-10 12:58:07 It's my usual complaint - they just kept adding on more and more layers of complexity. 2026-04-10 12:58:29 Im far more likely to design in a LMV431BIM3/TR and hand solder it into a prototype then write code as Im a hardware guy 2026-04-10 12:59:15 Modern makefiles have become Greek to me. 2026-04-10 12:59:30 KipIngram, over the years Ive read a fair number of automake hate posts online 2026-04-10 13:00:15 I *love* Make because it has made all my projects so easy to tie everything together 2026-04-10 13:00:40 and I admit that some of my Makefiles are quite long 2026-04-10 13:01:00 Yes - plain vanilla make is simple and powerful. It's obvious what's going on. 2026-04-10 13:01:17 Long doesn't matter, as long as each part of it is clear. 2026-04-10 13:01:42 yes, it's the shorthand plagued makefiles I can't begin to understand 2026-04-10 13:01:49 It was based on one simple idea - learn that idea and you've got it. 2026-04-10 13:02:10 Sort of like Forth. 2026-04-10 13:03:03 very true KipIngram 2026-04-10 13:06:32 I remember when I was first exposed to the (original) make, my reaction was "Oh, sure - that makes total sense. Got it." Then when I started seeing later makefiles it was like... "Huh????" 2026-04-10 13:06:58 I felt "dis-empowered." 2026-04-10 13:07:36 If I had problems building an old project, I could roll up my sleeves and figure out what was going on. If a modern project won't build right away, I'm mostly just out of luck. 2026-04-10 13:08:00 KipIngram, in my FURS project, I include 'pichar' flowcharts for every Makefile rule so the function is clear 2026-04-10 13:08:03 Some open source priest somewhere knows what's going on, but I don't. 2026-04-10 13:08:28 The whole modern trend is the emergence of priesthoods. 2026-04-10 13:08:47 People who know the secret language that the rest of us don't. 2026-04-10 13:09:04 thats well put, Ive never looked at it that way 2026-04-10 13:09:12 So open source has become less accessible to the masses rather than more accessible. 2026-04-10 13:09:41 I just see their work as the code of some master programmer who operates at a level I can't 2026-04-10 13:10:02 tpbsd: I think this is why a lot of software academics shun Forth. It's TOO SIMPLE - there's no place to establish a knowledge dominance for yourself. 2026-04-10 13:10:55 are humans really that bad ? I've never looked at it that way either 2026-04-10 13:11:04 No place to plant yourself from where you can lord your expertise over everyone else. 2026-04-10 13:11:36 I don't think all of them are, but there are definitely some who are like that. Maybe subconsciously. 2026-04-10 13:12:01 Ive rarely come across people who 'protect' their jobs like that, but they do exist (and are very incompetent, in my experience) 2026-04-10 13:12:22 I run into it trying to read math papers too. Most math papers are communiques between club members, rather than efforts to distribute the ideas as widely as pssible. 2026-04-10 13:12:58 naturally I've got tales of having such people cause me grief and confusion over the years 2026-04-10 13:13:29 Several times when I have been trying to learn some particular bit of math I find papers, but my eyes glaze over because the proportion of terms I don't understand is too high. High enough that I can't figure them out from context. 2026-04-10 13:13:59 Then later I find a paper written in a more "teaching" way, and discover that it was never really that hard an idea at all - it was just hidden behind a wall of insider lingo. 2026-04-10 13:14:35 ahh, once again I always assume I lack the mental horsepower to comprehend such documents 2026-04-10 13:14:38 But it can sometimes take years to find that "teaching document." 2026-04-10 13:15:04 Well, SOMETIMES it's actually hard, but a lot of the time I've discovered it's not. 2026-04-10 13:15:39 That it can be explained in a "different way" that's a lot easier to grok. 2026-04-10 13:15:44 The problem with old school make is that it's simple and easy to understand and rarely solves any real problems 2026-04-10 13:15:46 it's refreshing to hear such an opinion, it means there may be hope for me yet! 2026-04-10 13:16:10 Because most things now don't have a small number of dependent files with slow build times, they tend to have thousands of files before it's slow 2026-04-10 13:16:21 Although I find make works well for latex because that's still slow 2026-04-10 13:16:27 I write a lot of Quora answers - somehow just fell into the habit. Mostly on physics and math stuff. I always try hard to write them in an accessible way. 2026-04-10 13:16:36 it's true that if we cant explain something to a 6 year old, perhaps we dont really understand it ourselves 2026-04-10 13:16:39 But I do like old-school make 2026-04-10 13:16:45 It seems like the way to help more people. 2026-04-10 13:17:16 KipIngram, Ive only ever answered one 2026-04-10 13:17:38 I usually lack the knowledge to answer them 2026-04-10 13:17:43 I like Stack Overflow and related sites 2026-04-10 13:18:38 veltas, same 2026-04-10 13:19:15 I gave up on trying to participate in Stack Overflow - I found that more often than not people just belittled the person asking a question, or told them more or less "You're stupid to want to do that." 2026-04-10 13:19:42 In fairness to SO, it may have been Stack Exchange that left me with that impression. It was one or the other and I now can't remember which. 2026-04-10 13:19:52 on a positive note, my small order of voltage regulator parts from LCSC.com (Hong Kong) arrived fine in Australia after a couple of weeks 2026-04-10 13:20:23 :-) Given the concerns we're having about supply chains at the moment, that's good. 2026-04-10 13:20:41 so the world supply chain isnt FUBAR yet it seems 2026-04-10 13:22:05 and I've a busy trucking depot opposite my shed, so I'll probably be able to tell from the sudden silence when diesel runs out here 2026-04-10 13:22:33 from lask of exhaust brakes waking me during the night 2026-04-10 13:24:20 on a unrelated note, it's getting darker here as winter approaches and every morning at 4am I look for satellites in the Starlink shells, but havent seen a single one, which shows they are really blacked out and having mininal effect on astronomy 2026-04-10 13:24:52 I know where and when they are as Im the one who set up my Starlink antenna 2026-04-10 13:25:44 Ive seen a couple of meteorites etc, but zero sattelites, and I've a great view of the sky on my morning walk 2026-04-10 13:30:07 I watched the ISS go over Houston one evening. It was quite interesting - it was bright as can be as it was cruising across the sky a little after sunset, and then suddenly it "winked out" - it has passed into Earth's shadow, though it was still completely in a visible part of my sky. It was like someone threw a switch. 2026-04-10 13:31:17 A passenger jet lane runs roughly over my house, southwest to northeast - I can see the same effect more gradually with them; at first they're really bright, but they dim gradually as the reflection angles don't work as well for getting the light to me. 2026-04-10 13:31:52 This all of course just "shortly after" sunset, when I'm no longer in the sun but the airplanes are. 2026-04-10 13:32:16 Later on in the evening you only see their own blinking lights. 2026-04-10 13:32:26 perhaps the sun position isnt ideal for me at this time of the night 2026-04-10 13:33:32 I did see some of the Starlinks when it first came out, before they were blacked out to lessen the complains from earthbound astronomers 2026-04-10 13:34:52 I think the worst blow we've struck to astronomy is just city lights. For the majority of the planet's population we've turned the night sky from an absolute marvel into something that can't really be appreciated at all. 2026-04-10 13:35:24 So, naturally, most people never give it a second thought, instead of being practically knocked off their feet by it. 2026-04-10 13:36:05 thats so true 2026-04-10 13:36:44 Here in Houston I can't even really make out the constellations. Orion is still fairly noticeable, but it has to be an exceptionally dark, clear night for me to even clearly make out the Big Dipper. 2026-04-10 13:37:03 I have to look very carefully to see the North Star. 2026-04-10 13:37:32 in my travels in the remote parts of australia, the night sky is like an assault of stars, they literally cover the entire sky like a silk sheet 2026-04-10 13:37:49 it's breathtaking 2026-04-10 13:37:57 Yes - it's a true wonder. And I've still never seen it in ALL its glory. 2026-04-10 13:38:17 I can drive for about four hours to get to Bortle class 2 skies. 2026-04-10 13:38:30 but that observation is 40 years old, and times have changed 2026-04-10 13:38:40 And plan to, at some point. The only problem is weather - if I get out there and it's cloudy, it's a wasted trip. 2026-04-10 13:39:16 People living more toward the Eastern seaboard of the US are just flat out of luck - they'd have to drive for a day or more. 2026-04-10 13:39:36 last time I rode a motorbike around Australia was 2006, and no where i rode was free of barb wire fences, all the land has been owned and fenced now 2026-04-10 13:40:21 but when I was the sky was full of stars, fences were rare 2026-04-10 13:41:59 at one point I noticed a massive bassalt 'highway' miles wide, either side of the road in the outback, and so I rode it at 60 mph for a long time, just weaving around bushes that had grown in cracks in it 2026-04-10 13:42:45 it was amazing, and that huge area is now marked on maps of that area. It large enough to drive on it for days on a motorbike 2026-04-10 13:43:39 it's the remenants of some ancient volcano lava eruption I believe 2026-04-10 13:43:57 but it's smooth like the highway 2026-04-10 14:13:54 I'll try to make my earlier complaints more general. We complain about Windows, and talk about preferring Linux, because Windows is a closed system. Windows is closed through overt secrecy. But Linux is becoming closed too, through these layers of complexity I referred to earlier. The fact that it's "nominally open" isn't worth much if it's become complex to the point where most of the user 2026-04-10 14:13:57 base can't understand it and work with it. I think of myself as a pretty capable person, so if I can't penetrate the complexity then I feel like the bulk of the user community can't. I'm sure I COULD penetrate any part of it, if it were my burning passion to do so, but... I have other priorities. It shouldn't be that hard - it should be built in a way that makes it EASY for as many people as 2026-04-10 14:13:59 possible to get "hands on" with it. 2026-04-10 14:14:40 In my mind that should be the guiding principle of open source - "How do we do this so as many people as possible 'get it' and find it clear and straightforward?" 2026-04-10 14:15:06 But that's clearly NOT the mind set that has guided the last couple of decades of work. 2026-04-10 14:15:56 KipIngram, I remember building a CLI Linux back in the days of the 4.x kernel iirc, and it was only 80MB, everything was simpler back then 2026-04-10 14:16:09 Yes. 2026-04-10 14:16:53 amd of course, no USB, no Nvidia back then 2026-04-10 14:18:00 A black box is a black box - regardless of whether it's because of proprietary secrecy or difficulty of understanding. 2026-04-10 14:18:42 The whole point of open source was that you wanted millions of eyballs on the code making sure it was right. But if most of those eyeballs can't possibly follow it, you've lost that benefit. 2026-04-10 14:19:23 :-) So there's my theory of "populist software." 2026-04-10 14:21:18 But you know, it's happened all over. You used to be able to take your broken TV to Pop's TV Repair down the street, and he'd fix it for you. You used to be able to grab a wrench from your trunk and get your stalled car back on the road. But everything has gotten completely opaque for most people these days. 2026-04-10 14:21:38 We tend to blame it on people "not learning," but there are other factors involved too. 2026-04-10 14:23:16 progress! 2026-04-10 14:23:24 Oh, and when Pop fixed your TV he'd likely replace one tube, or one transistor or something. These days it's more likely that if it's worth repairing at all it requires a $400 PCB. More often you just chuck the whole thing and buy a new one. 2026-04-10 14:24:11 yeah, the parts are mostly SMD and there are no parts available 2026-04-10 14:24:51 they tend to make 'white goods' in batches and when the batch is done, parts arent made 2026-04-10 14:25:07 take hewlett packard for instance 2026-04-10 14:25:26 KipIngram: Have you tried 'penetrating' the kernel? 2026-04-10 14:25:40 I purchased a new digital storage scope in 1994, and paid $4600 aud for it 2026-04-10 14:25:59 And every step of it has moved us away from "independence" and toward reliance on central points of control. 2026-04-10 14:26:57 Another open source argument was "if something isn't working right you can fix it yourself." But more often we just have to cross our fingers and hope that "it gets fixed upstream at some point." 2026-04-10 14:27:04 6 years later the scope developed a problem, I phoned HP Australia and they told me "sorry, we only keep parts for 5 years now" so I couldnt even buy parts 2026-04-10 14:27:08 I bought a RISC-V SBC that runs Linux. I reckon would be capable of most modern computer tasks and was pretty cheap, and you could probably decipher how most of it worked 2026-04-10 14:27:52 KipIngram I am wondering though did you actually tru reading Linux kernel? 2026-04-10 14:27:56 try* 2026-04-10 14:28:32 No - I haven't. Like I said above, I'm pretty sure I could figure out any of it if I took the time, but there are other things I want to do with my time. 2026-04-10 14:28:45 so I ran a search on ebay for a motherboard for the scope, and 6 months later got a hit, a new motherboard for $25! I purchased it and it fixed the problem, been working ever since! 2026-04-10 14:28:53 Back to make - it took two minutes to understand how make worked. 2026-04-10 14:29:02 That's the level of "obviousness" I'd like in all of it. 2026-04-10 14:29:36 I want "Forth simplicity" EVERYWHERE. 2026-04-10 14:30:35 KipIngram, I asked my AI to explain the terms used in a makefile, hes the one liner "if you want to use the exact terminology from the GNU Make manual, you are writing a **rule**, which consists of a **target**, **prerequisites**, and a **recipe**." 2026-04-10 14:31:29 That's nice - I'll just say that each of those starred parts was entirely obvious in the old original makefiles. 2026-04-10 14:31:37 I've moved to GLM-5.1 also, because it has a 'chain of reasoning' I prefer to see 2026-04-10 14:31:41 They're not any more, even if the terms still nominally apply. 2026-04-10 14:32:22 And also make no longer does the whole job - you have all those other layers of crud there too. 2026-04-10 14:33:57 ./configure ; make ; sudo make install was a beautifully simple recipe. And you didn't even HAVE to "sudo make install" if you didn't want to - you could test it locally there in that build directory before committing to system-wide changes. 2026-04-10 14:34:18 That's all we needed. 2026-04-10 14:35:45 I reckon you'd have a good chance of understanding stuff if you needed to, Linux kernel has done a good job managing their complexity 2026-04-10 14:36:12 I'm not telling you to go do it, just I don't think it's fair to assume it's unapproachable without approaching 2026-04-10 14:36:29 Linux is a pretty rare kind of C project though 2026-04-10 14:36:39 thats true 2026-04-10 14:37:33 I don't doubt that - but I'm a fairly capable person. I'd like for a majority of users not just "be able to understand it," but be able to easily enough that they actually DO. 2026-04-10 14:37:39 Im not a programmer, but I succesfully developed a parallel port controlled 'universal programmer' that I released under the GPL in 2000, called 'burn' writen in Gcc 2026-04-10 14:38:03 What all would it program? 2026-04-10 14:38:22 I designed the hardware and software, and one person I know off actually built one by following my project 2026-04-10 14:38:39 I spent the best part of my career at a programmer company - back in the late 90's when there was still a need for specialized equipment to do that job. 2026-04-10 14:39:11 anything really, I designed the code to be easily changed to suit any mcu or flash of that era 2026-04-10 14:39:14 Programming at that time was a delicate endeavor - some part sneeded things like precise timing, precise current ramp rates, etc. 2026-04-10 14:39:29 I.e., not things you could do over a parallel port - specialized hardware was required. 2026-04-10 14:39:48 But that model was doomed - the parts just kept getting easier and easier to program. 2026-04-10 14:40:03 We made a lot of money there in the late 90's, though. 2026-04-10 14:40:06 it was all parallel programming back then 2026-04-10 14:40:16 50% revenue growth year on year for like 7-8 years. 2026-04-10 14:40:22 nice 2026-04-10 14:40:36 Then the 2001 tech slump hit and those revenues just died on the vine. 2026-04-10 14:40:59 The nature of our business model was that revenues were proportional to GROWTH of the market - all it took was for growth to go flat, and... outch. 2026-04-10 14:41:22 s/outch/ouch/ 2026-04-10 14:41:50 2000 revneue was $50 million plus; 2001 revenue was like $12 million. 2026-04-10 14:41:52 I have the schematic of that project, you'll laugh, I have a habit of non-typical designs ... I used a 6820 PIO chip and relays so I could parallel program in-circuit in my design 2026-04-10 14:42:52 The desktop programmers we sold to engineers kept right on selling, but the big automated systems built around PnP machines, where the bulk of the revenue came from, just stalled totally - the ones we'd sold to our customers last year were still adequate to do their work. 2026-04-10 14:45:14 I resigned in 2002 and started consulting. I think if I'd stayed I'd have been laid off before too long - the one guy in the company with a higher salary than mine was fired in early 2002, and by the end of 2002 the entire top management layer was gone. 2026-04-10 14:45:30 https://sourceforge.net/projects/mecrisp-stellaris-folkdoc/files/burn-proj-schematic.jpeg/download 2026-04-10 14:45:37 I just decided to show myself the door before it was shown to me. Plus it wasn't as fun anymore - the downturn really changed the atmosphere of the place. 2026-04-10 14:46:01 thats the schematic, all opensource and done with gEDA gSchem schematic capture 2026-04-10 14:46:27 yeah, downturns are hell 2026-04-10 14:46:27 It's not surprising though - the owner had started the business right out of college, and had never known anything other than success. He was unprepared emotionally to cope with "economic struggle." 2026-04-10 14:47:15 the 1987 downturn in Australia killed the best job I ever had 2026-04-10 14:47:39 But - that is still the job I look back on as the best one of my career. I had all kinds of fun, and I was good at running that engineering team. 2026-04-10 14:47:53 which was technical manager of a SMT assembly plant 2026-04-10 14:49:42 And those years of fast growth did wonderful things to my salary - and once you've earned a salary at one company you're a lot more likely to get something like it at other companies later. It's almost like getting into a club. 2026-04-10 14:50:14 Even after my consulting period I was able to go back into the job market at more or less the same level. 2026-04-10 16:07:35 KipIngram: Yeah I'm not having a lot of fun in aftermath of recent downturn