--- Log opened Sun Jun 14 00:00:49 2020
00:48 -!- Urchin[emacs] [~user@unaffiliated/urchin] has joined ##hplusroadmap
00:59 -!- darsie [~kvirc@84-114-73-160.cable.dynamic.surfer.at] has joined ##hplusroadmap
01:01 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
01:01 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 272 seconds]
01:23 < TMA> nmz787: twenty years ago, there was silence in the forest near where I live not far from the edge; car trafic can be heard in the very centre of it today. I do agree, that the preference is valid (I have a similar one). At the same time it is not necessary for survival -- you can survive even when very uncomfortable.
01:41 -!- wallet42 [sid154231@gateway/web/irccloud.com/x-qxkrpcnnakptdxdo] has joined ##hplusroadmap
01:42 -!- sivoais [~zaki@unaffiliated/sivoais] has quit [Ping timeout: 264 seconds]
01:44 -!- sivoais [~zaki@unaffiliated/sivoais] has joined ##hplusroadmap
01:50 -!- ffranr [ffranr@gateway/vpn/protonvpn/ffranr] has joined ##hplusroadmap
01:53 -!- ffranr_ [ffranr@gateway/vpn/protonvpn/ffranr] has joined ##hplusroadmap
01:54 -!- ffranr [ffranr@gateway/vpn/protonvpn/ffranr] has quit [Ping timeout: 240 seconds]
02:26 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 272 seconds]
03:06 -!- sivoais [~zaki@unaffiliated/sivoais] has quit [Ping timeout: 256 seconds]
03:06 -!- sivoais [~zaki@unaffiliated/sivoais] has joined ##hplusroadmap
03:17 < juri_> the SEM i used to hack on (which has been mal-treated since i left) is up for grabs.
03:17 < juri_> so, if you're near DC, and want a working turbo, turbo controller, vacuum sensor, and slightly leaky column, hit me up.
03:21 < juri_> (slightly leaky = pumps down to 0.3 torr.)
03:35 < darsie> I'm too far (Austria), but how much would you take for it?
03:49 < juri_> free to a good home.
04:05 -!- ffranr [ffranr@gateway/vpn/protonvpn/ffranr] has joined ##hplusroadmap
04:08 -!- ffranr_ [ffranr@gateway/vpn/protonvpn/ffranr] has quit [Ping timeout: 265 seconds]
06:05 < L29Ah> were you working on metalicarap?
06:06 < L29Ah> too bad shipping a vacuum turbopump to .ru would probably cost more than acquiring an oil vapor pump, and i can't even allocate funds for that
06:11 < juri_> L29Ah: yep.
06:13 < L29Ah> juri_: what do you think on the project's future?
06:19 < juri_> I prefer to work with things that are a little more accessible to others. FDM is the gateway to the diamond age, not metalicarap.
06:20 < juri_> fdm with plastic, under gravity and atmosphere, that is'
06:38 < L29Ah> juri_: and what about the technological standpoint? i'm confused that almost ten years have passed, rapatan have grabbed all the required expensive equipment long ago, and there's still no info on PoC operation or even a project log
06:51 < juri_> I went in different directions technologically long ago. the custom vacuum chamber requirements are just bonkers.
08:18 -!- mauz555 [~mauz555@2a01:e0a:56d:9090:11a8:67ae:75b1:26a5] has joined ##hplusroadmap
08:41 -!- yashgaroth [~ffffffff@172.58.22.164] has joined ##hplusroadmap
08:49 -!- mauz555 [~mauz555@2a01:e0a:56d:9090:11a8:67ae:75b1:26a5] has quit []
08:56 -!- ffranr [ffranr@gateway/vpn/protonvpn/ffranr] has quit [Ping timeout: 240 seconds]
08:57 -!- ffranr [~ffranr@213.167.139.227] has joined ##hplusroadmap
09:10 -!- CryptoDavid [uid14990@gateway/web/irccloud.com/x-szinfyirvfzjwghi] has joined ##hplusroadmap
09:13 < lsneff> juri_: I question if FDM is the gateway to APM, but I think the sentiment is very good.
09:14 -!- ffranr [~ffranr@213.167.139.227] has quit [Ping timeout: 260 seconds]
09:16 < srk> lsneff: APM?
09:17 < lsneff> srk: atomically precise manufacturing
09:17 < srk> thanks
09:17 < lsneff> Specifically with molecular assembly
09:17 < lsneff> There are others kinds of APM that are less useful
09:38 < juri_> APM is not as generally useful. I'm sure eventually it will be, but this is quite a developing field, even now that the 'buzz' is gone.
09:39 < lsneff> I'm not sure it's even a developing field. There has been barely any progress towards molecular assemblers in the last 20 years.
09:41 < L29Ah> how do i solar panel with FDM
09:41 < L29Ah> even PCBs are a challenge
09:43 < juri_> PCBs are a challenge because the bottom dropped out of the market. ordering PCBs is just so cheap.
10:18 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
10:21 < fenn> open source development is dead. that's why no new reprap developments in the past 10 years. it doesn't benefit anyone's bottom line
10:21 < fenn> i disagree that the easiest possible thing is the way forward
10:22 < fenn> usually the reason nobody has done it yet, is because it's hard
10:22 < fenn> even with FDM there is no five axis path planning
10:22 < fenn> and that's "just a simple matter of programming"
10:33 < srk> fenn: we're trying to improve the situation - https://distrap.org/
10:34 < srk> slowly but steady, reprap firmwares are hitting limits in terms of flash/complexity/scalability/maintenance so improving that is quite hard
10:35 < fenn> srk this is completely unrelated but a problem i run into constantly and can't figure out. could you try to view distrap.org in elinks or dillo and tell me if it works for you?
10:36 < srk> sure
10:36 < srk> oh, getting ssl error indeed, guess I need to upgrade/migrate the host
10:37 < fenn> could you tell me what went wrong?
10:37 < srk> sec
10:38 < srk> openssl s_client looks good tho
10:39 < srk> lol, it rotated logs like a minute ago
10:39 < jrayhawk> it's a tls version problem, probably: openssl s_client -connect distrap.org:443 -tls1_3
10:40 < jrayhawk> tls1.3 notably doesn't support and sort of renegotiation, which can result in a lot of broken configurations
10:40 < jrayhawk> er, any sort
10:41 < srk> I see, the machine needs migration from RHEL7 to NixOS anyway, want to move it to pi4 from the current VPS
10:41 < srk> fenn: tldr we're trying to build distributed control system utilizing CAN(Open) and closed loop control
10:42 < fenn> right. i was working on a similar project in 2007 ish
10:42 < srk> oss? any links?
10:43 < fenn> http://fennetic.net/machines/iskewb didn't really get to a publishable state but i was controlling a servo with an AVR
10:47 < fenn> i wanted to print my own indexed optical encoders on transparency sheets and use standard brushed dc motors
10:48 < fenn> there was also this idea to use a pseudorandom code embedded in the encoder to get absolute position after a few  degrees of movement
10:49 < srk> cool, we were talking about DIY encoders today. my controllers are stm32 based with off the shelf optical or hall effect SPI encoders now tho
10:50 < srk> and BLDC motors :)
10:51 < jrayhawk> huh, funny. i used to live with an ivory author.
10:51 < srk> Lee Pike? :D
10:52 < fenn> is it common for people to force https now?
10:52 < fenn> people with web servers i mean
10:52 < srk> yes, even with HSTS
10:53 < jrayhawk> it's pretty common to 301 the http namespace over ever since google publicly disclosed the SEO benefit
10:53 < srk> hehe, it's pretty easy to set-up and forget thanks to letsencrypt
10:54 < srk> jrayhawk: sadly they've kind-of abandoned the whole project as the grant expired but I try to keep it alive and add support for more devices
10:54 < fenn> what is ivory?
10:54 < srk> https://ivorylang.org/
10:54 < jrayhawk> another systems haskell derivative
10:56 < fenn> ah i see. "i used to live with a dumpster author"
10:56 < srk> in my case I'm using it to generate C code which is then linked with FreeRTOS via Tower library that adds safe concurrency for ivory (typed channels basically)
10:56 < fenn> no offense intended. but dumpsters were big in 2007
10:56 < srk> hehe, the name is not the best but I naming is the hardest ..
10:56 < srk> s/I//
10:57 < fenn> second only to off by one errors
10:57 < srk> it's a pun explained in one of the papers, meaning to bring embedded programming down from ivory tower
10:59 < srk> what I like about it is that it feels like using Haskell natively despite going thru C - if it compiles it usually works :)
10:59 < fenn> what
10:59 < fenn> you mean bring haskell down from the ivory tower, surely?
10:59 < fenn> embedded is like the most practical shit possible
10:59 < jrayhawk> well, type-safety constructs embodied by haskell
11:00 < srk> well :) it's arguably easier than plain C for embedded but it depends on the use-case, for complex systems it definitely is
11:01 < srk> complex systems.. like anything with concurrency
11:01 < jrayhawk> from my understanding, galois has mostly moved onto developing the rust ecosystem for practical work and other, even stronger type systems for research and proving
11:02 < srk> well they're working on what's paying for the research via grants. recent work I saw was voting machines and symbolic execution for program analysis
11:03 < srk> and betaflight on risc-v :D
11:06 < srk> hah even https://github.com/GaloisInc/RustyFlight
11:27 < L29Ah> 20:21:23]<fenn> open source development is dead. that's why no new reprap developments in the past 10 years. it doesn't benefit anyone's bottom line
11:27 < L29Ah> i'm loving my shiny new dual drive extruder, and i'm pretty sure it wasn't around 10 years ago
11:28 < L29Ah> 20:34:19]<srk> slowly but steady, reprap firmwares are hitting limits in terms of flash/complexity/scalability/maintenance so improving that is quite hard
11:28 < L29Ah> lold @ flash being a limit
11:29 < L29Ah> 20:33:06]<srk> fenn: we're trying to improve the situation - https://distrap.org/
11:29 < L29Ah> yay cool! how's closed loop BLDC-driven going g-code interpreter going?
11:29 < L29Ah> s/going//
11:32 < L29Ah> the wiki suggest it's mostly about infra and hardware interfaces though, and those are okay as is as for me :/
11:33 < srk> L29Ah: interpreter is part of https://github.com/distrap/gcodehs currently working on it to be able to vis the gcode for juri_
11:33 < jrayhawk> fenn: sketchpad and the apple lisa also didn't get widespread use and could be characterized as "dumpster" inventions, but calling them such would similarly serve to debase you more than them.
11:33 < srk> L29Ah: then it needs kinematics + motion planning, time parametrization, transfering to interpolators which "just" follow the computed curves
11:34 < L29Ah> https://github.com/distrap/lambdadrive suggests you have FOC operational, nice!
11:34 < srk> yup, I'm in the process of porting that to stm32f765 but it can run on stm32f4 based ODrives already
11:35 < srk> it should be pretty stm32 agnostic after that, being able to target various hw platforms
11:37 < srk> hw support is generated from SVD files with https://github.com/HaskellEmbedded/data-stm32 which generates https://github.com/HaskellEmbedded/ivory-tower-stm32-generated#stm32-device-support-for-ivory-tower
11:38 < srk> originally only few MCUs of the stm32f4 series were supported
11:38 < L29Ah> i'm skeptical of CAN though: what's the expected lag of position reports from the motor controller to the motion planner/gcode interpreter? how do you plan to keep the machine in sync when one of the motor is having problems keeping up to the expected speed?
11:40 < srk> L29Ah: CANOpen defines Interpolated Position mode where interpolators are tracking the curves and if one of them can't keep up within specified tolerance it just reports that and the machine is halted
11:41 < L29Ah> so it's no better than steppers with stall detection
11:41 < srk> it is, you can still recover
11:41 < srk> as it knows the position at all times
11:41 < fenn> when a stepper falls behind it can't catch up
11:41 < srk> and it corrects small errors automatically
11:42 < L29Ah> fenn: yes, and closed loop control is interesting because it theoretically can, but i don't see the described design doing it smoothly enough to ignore the incidents at the user side, allowing the task to proceed at max available performance w/o tuning
11:42 < fenn> there will always be tuning
11:43 < srk> L29Ah: that's one of the goals - being able to not just run the motors closed loop but the whole machine
11:43 < srk> for 3d printers, extruder is the limiting factor and with feedback you can adjust feeds of the axes according to that
11:43 < L29Ah> yup
11:44 < srk> for mills, you can add e.g. acceloremeters and adjust feedrates based on that
11:44 < fenn> jrayhawk: i meant literally there was a lot of enthusiasm for dumpsters in my social scene in 2007, and there was a project called "dumpster" on that github account
11:45 < fenn> that said, i don't think there's a need for either CAN or haskell in a simple motion control system like a 3d printer
11:45 < srk> or load cells or whatever :)
11:45 < fenn> i mean it's easily doable in assembly really
11:45 < srk> fenn: it's not simple, the project is not limited to 3d printers, it depends on kind of 3d printer as well
11:45 < fenn> CAN is just complexity for its own sake
11:46 < jrayhawk> programming safety matters a lot when a head collision costs a million dollars
11:46 < srk> for me it's one less protocol to re-invent
11:46 < srk> (CANOpen)
11:46 < srk> CAN is pretty easy
11:46 < srk> jrayhawk: ever had a print failure after 20 hours of print or so? :)
11:46 < jrayhawk> fenn: also, rather than "dead", which is trivially falsifiable, the more useful and defensible claim is that libre does not naturally win anymore. https://www.youtube.com/watch?v=vBknF2yUZZ8&t=30m24s
11:48 < L29Ah> CAN is an extra and completely unnecessary trouble source as for me, and also synchronous buses are nicer overall on such a scale; not sure about haskell, probably rust would be a better choice for stm32, but the motion planner sounds quite complex for C and allows a powerful enough machine to run whatever language
11:49 < srk> I was following embedded rust for a long time and its not quite there yet :)
11:49 < fenn> jrayhawk: yeah that's basically what i was trying to say
11:49 < L29Ah> embedded haskell is even farther ;]
11:49 < fenn> "firms engaging in strategic openness"
11:50 < srk> L29Ah: try both and let me know what you think :)
11:50 < srk> biggest reason for CAN is that it's differential
11:50 < L29Ah> i've tried a little rust, will be trying more
11:51 < L29Ah> seen a PoC haskell stm32 fimware and wasn't quite happy
11:51 < srk> I know the state of both projects, rust is all the hype now but haskell is around for a while so it will take time till it catches up
11:51 < L29Ah> srk: just keep your signal grounds straight and you won't need a differential bus :P
11:52 < L29Ah> srk: haskell doesn't work w/o a proper gc unlike rust
11:52 < srk> L29Ah: it's not haskell on embedded per se, it's generated C code from haskell eDSL
11:52 < L29Ah> ah yes, i'm yet to check ivory
11:52 < srk> L29Ah: https://ivorylang.org/ http://48.io/~rmarko/book/
11:53 < srk> latter is still wip but it contains few nice example
11:53 < srk> s
11:53 < L29Ah> tfw even a simple project documentation page needs crappy webfonts
11:56 < srk>  /o\
11:56  * srk off to watch ZuriHac talks, later o/
12:09 < fenn> jrayhawk: and a related point, which mako didn't mention. the existence of so many firms employing programmers with expertise in free software means that there's less un(der)employed free software programmers to actually work on the stuff we care about
12:10 < fenn> i don't even know how to explain the ridiculously sorry state of fdroid and the apps available on it
12:11 < fenn> or maybe people just got old and there was not enough in the new generation to replace them
12:13 < L29Ah> fdroid is doomed by basing itself on a hostile API and software stack
12:13 < fenn> by "old" i mean something like "got married, had kids, now spend free time mowing lawn"
12:13 < jrayhawk> Yeah, a lot of the GPL's success was brain-drain-resistance
12:13 < L29Ah> active people understand it and move to PMOS and other GNU/Linux projects
12:13 < fenn> never heard of PMOS
12:14 < fenn> .wik PostmarketOS
12:14 < saxo> "PostmarketOS (stylized as postmarketOS and abbreviated as pmOS) is a free and open-source operating system under development primarily for smartphones, based on the Alpine Linux distribution. /  PostmarketOS was launched on May 6, 2017 with the source code available on [...]" - https://en.wikipedia.org/wiki/PostmarketOS
12:15 < fenn> i never heard about the ubuntu OS either, until after it was EOL'd
12:15 < fenn> for phones
12:16 < fenn> ok so they ported X to a phone. wow. big deal
12:29 < Urchin[emacs]> wdat about Ada?
12:29 < Urchin[emacs]> &what
12:29 < Urchin[emacs]> *
12:29 < fenn> the language?
12:29 < Urchin[emacs]> yes
12:30 < Urchin[emacs]> it's for embedded stuff anyway
12:30 < Urchin[emacs]> and it has static verification stuff if you need it
12:48 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 272 seconds]
13:15 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
13:41 < jrayhawk> everyone and their mother was competing to make linux-based phone operating systems back around 2010
13:41 < jrayhawk> the emokit guy who used to hang around in here was doing one for Mozilla
13:41 < jrayhawk> stskeeps was around here for a while and his was actually moderately successful
13:42 < jrayhawk> in that i am using it today
13:49 < jrayhawk> but yeah, palm (webos), mozilla (firefox OS), ubuntu (touch), nokia (maemo, meego), samsung (tizen), intel (moblin, meego, tizen), kde (plasma active), to a very limited extent fdo (wayland, telephony), jolla (mer, sailfishos) almost all of them were very walled garden develpment, though.
13:49 < jrayhawk> also ubuntu was making their own compositor (like palm and mozilla) specifically because X11 sucks, and they were developing SDL/GTK+/Qt5 targets for it
13:50 < jrayhawk> i don't think anyone tried to target X11 post-maemo
13:59 < jrayhawk> and google, of course
14:07 < fenn> don't forget openmoko, inventors of the smartphone
14:08 -!- Codaraxis_ [~Codaraxis@ip68-5-90-227.oc.oc.cox.net] has joined ##hplusroadmap
14:12 < juri_> I miss my openmokos.
14:12 -!- Codaraxis__ [Codaraxis@gateway/vpn/mullvad/codaraxis] has quit [Ping timeout: 264 seconds]
14:14 < jrayhawk> i actually preferred it when i had forgotten about openmoko
14:55 -!- filipepe [uid362247@gateway/web/irccloud.com/x-hwktjpueavgowazd] has joined ##hplusroadmap
15:05 -!- ffranr [~ffranr@213.167.139.102] has joined ##hplusroadmap
15:08 -!- HumanGeek [~HumanG33k@62.147.242.8] has joined ##hplusroadmap
15:11 -!- Human_G33k [~HumanG33k@62.147.242.8] has quit [Ping timeout: 256 seconds]
15:15 < gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=eafd63b8 Michael Folkson: Add Adam Gibson on Schnorr signatures >> http://diyhpl.us/diyhpluswiki/transcripts/london-bitcoin-devs/2018-06-12-adam-gibson-unfairly-linear-signatures/
15:15 < gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=c9ecc2ea Michael Folkson: Edits for Pieter Wuille at Scaling Bitcoin in Milan >> http://diyhpl.us/diyhpluswiki/transcripts/scalingbitcoin/milan/schnorr-signatures/
15:15 < gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=bce4d278 Michael Folkson: Merge pull request #115 from michaelfolkson/sipa-milan-edits >> http://diyhpl.us/diyhpluswiki/
15:15 < gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=9a740665 Michael Folkson: Edits for Tim Ruffing transcript >> http://diyhpl.us/diyhpluswiki/transcripts/cryptoeconomic-systems/2019/threshold-schnorr-signatures/
15:15 < gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=9b387508 Michael Folkson: Merge pull request #116 from michaelfolkson/tim-ruffing-ces-edits >> http://diyhpl.us/diyhpluswiki/
15:15 < gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=a3a2a474 Bryan Bishop: Merge pull request #114 from michaelfolkson/add-waxwing-schnorr >> http://diyhpl.us/diyhpluswiki/
15:38 -!- ffranr [~ffranr@213.167.139.102] has quit [Quit: ffranr]
15:39 -!- ffranr [~ffranr@213.167.139.102] has joined ##hplusroadmap
17:04 -!- filipepe [uid362247@gateway/web/irccloud.com/x-hwktjpueavgowazd] has quit [Quit: Connection closed for inactivity]
17:55 -!- ffranr [~ffranr@213.167.139.102] has quit [Ping timeout: 260 seconds]
17:55 -!- CryptoDavid [uid14990@gateway/web/irccloud.com/x-szinfyirvfzjwghi] has quit [Quit: Connection closed for inactivity]
17:58 < L29Ah> 22:16:41]<fenn> ok so they ported X to a phone. wow. big deal
17:58 < L29Ah> what
18:06 < fenn> maybe i don't understand what it is they did, but it didn't look like a usable OS from the few screenshots i could find
18:09 < L29Ah> check youtube
18:44 -!- darsie [~kvirc@84-114-73-160.cable.dynamic.surfer.at] has quit [Ping timeout: 258 seconds]
19:07 -!- Urchin[emacs] [~user@unaffiliated/urchin] has quit [Ping timeout: 258 seconds]
19:17 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has quit [Killed (orwell.freenode.net (Nickname regained by services))]
19:18 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has joined ##hplusroadmap
19:21 -!- L29Ah [~L29Ah@unaffiliated/l29ah] has left ##hplusroadmap ["Disconnected: closed"]
19:23 -!- L29Ah [~L29Ah@unaffiliated/l29ah] has joined ##hplusroadmap
20:22 -!- filipepe [uid362247@gateway/web/irccloud.com/x-dnscftquxhcoonmr] has joined ##hplusroadmap
20:56 -!- yashgaroth [~ffffffff@172.58.22.164] has quit [Quit: Leaving]
21:42 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 260 seconds]
21:51 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
22:04 -!- Urchin[emacs] [~user@unaffiliated/urchin] has joined ##hplusroadmap
22:41 -!- filipepe [uid362247@gateway/web/irccloud.com/x-dnscftquxhcoonmr] has quit [Quit: Connection closed for inactivity]
23:01 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
23:03 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 260 seconds]
--- Log closed Mon Jun 15 00:00:51 2020