| - chomwitt (QUIT: Ping timeout: 240 seconds) (~chomwitt@2a02:587:7a19:9600:1ac0:4dff:fedb:a3f1) | 02:24 | |
| - nsc (QUIT: Ping timeout: 256 seconds) (~nicolas@i5C74DFD7.versanet.de) | 03:28 | |
| + nsc (~nicolas@i5C74DEB8.versanet.de) | 03:30 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:03 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:05 | |
| - kop316 (QUIT: Remote host closed the connection) (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:06 | |
| + kop316 (m-6f6zq6@static.138.159.90.157.clients.your-server.de) | 04:06 | |
| + cow321 (~deflated8@user/meow/deflated8837) | 04:18 | |
| - q66 (QUIT: Ping timeout: 256 seconds) (~q66@q66.moe) | 07:18 | |
| ex-parrot | ok, hold your boots I'm gonna try the debian installer image | 08:39 |
|---|---|---|
| + chomwitt (~chomwitt@2a02:587:7a19:9600:1ac0:4dff:fedb:a3f1) | 09:09 | |
| - chomwitt (QUIT: Ping timeout: 268 seconds) (~chomwitt@2a02:587:7a19:9600:1ac0:4dff:fedb:a3f1) | 09:15 | |
| josch | ex-parrot: boots are fastened tightly! | 09:22 |
| josch | minute: what do you think of this? https://paste.debian.net/hidden/02479c37/ | 09:23 |
| josch | we could run it as part of reform-tools or reform-system-image test or both | 09:23 |
| ex-parrot | josch: not off to a good start, with the bpo image the extlinux menu comes up, kernel and initrd load, keyboard lights up and the display turns off :) | 09:30 |
| josch | nice :) | 09:31 |
| ex-parrot | just gonna try the non-bpo image | 09:32 |
| josch | ex-parrot: also note, that reform.d.n has been broken since middle of may due to new kernel | 09:32 |
| josch | i wanted to fix it again once rk3588 is out | 09:32 |
| ex-parrot | ah yeah, I wonder if that's why I ended up accidentally booting the standard debian arm64 kernel? | 09:33 |
| ex-parrot | ok the non-bpo installer starts correctly | 09:33 |
| josch | oh nice! | 09:34 |
| josch | ex-parrot: what platform are you on? imx8mq? | 09:34 |
| ex-parrot | yep | 09:34 |
| ex-parrot | reform number 37 | 09:34 |
| josch | okay, that should work well with non-bpo indeed | 09:34 |
| josch | uff! :D | 09:35 |
| ex-parrot | I was running the bpo kernel most recently in my install actually, and that worked great | 09:35 |
| josch | well yes, they both *should* work :D | 09:35 |
| ex-parrot | honestly aside from some display timing issues (I think?) the debian stock arm64 kernel also seemed to basically work, I could boot in to Gnome and do computery stuff | 09:35 |
| josch | yes, if you look at the imx8mq patch stack, only very few things are missing | 09:35 |
| ex-parrot | I thought I'd killed the SoM somehow until I realised | 09:36 |
| ex-parrot | great stuff | 09:36 |
| ex-parrot | I managed to get the last module needed for the PocketCHIP added in debian so it's possible to run a PocketCHIP now entirely on the stock kernel | 09:36 |
| ex-parrot | still a lot of work to actually get it installed, I need to finish that project :( | 09:36 |
| ex-parrot | in the interests of science I am going to do a basically next -> next -> next Debian install and see what I get | 09:37 |
| josch | ex-parrot: the interesting steps are at the end | 09:37 |
| josch | ex-parrot: i added some debconf questions about where to put u-boot and stuff | 09:37 |
| ex-parrot | ah nice! | 09:37 |
| ex-parrot | I've just been running an ext2 /boot and had everything sitting in there on the nvme and it seems to work | 09:38 |
| ex-parrot | I honestly thought I was booting off mmcblk0 and had a customised flash-kernel to drop the image in there but that must have been some other machine, I couldn't find any config relating to that when I was reverse engineering my own setup last night | 09:38 |
| josch | there was some discussion in #debian-devel whether the debian project could pay for a Reform for the person who adds vanilla Debian debian installer support to it | 09:41 |
| josch | fortunately/unfortunately everybody who would do that work already has a reform XD | 09:41 |
| ex-parrot | it seems so close | 09:41 |
| ex-parrot | yeah :P | 09:41 |
| ex-parrot | I basically succeeded with vanilla installer + a tiny bit of manual moving files around | 09:42 |
| josch | i think it is possible now that firmware-non-free is allowed in debian-installer | 09:42 |
| ex-parrot | yes! that will make things way easier | 09:42 |
| josch | it was not possible before where debian-installer only had to contain things from main | 09:42 |
| ex-parrot | I was just thinking that when I was reading the reform.d.o page | 09:42 |
| + q66 (~q66@q66.moe) | 09:42 | |
| ex-parrot | I am on the fence as to whether I will go forward using this Debian install or go help ec0 out testing OpenBSD, which is my other UNIX-like interest | 09:43 |
| ex-parrot | but in terms of actually doing useful work with the Reform I think really I need Debian... | 09:43 |
| ex-parrot | "Running reform..." | 09:43 |
| josch | yes, i was not creative when i named that :D | 09:44 |
| minute | josch: your paste looks cool! really creative way to do tests, nice | 09:59 |
| josch | minute: unfortunately i fear that the CI vm from hetzner do not support kvm, so this will be *very* slow | 10:00 |
| josch | all of this runs in 6m on a311d | 10:00 |
| josch | but since you seem to like it, i shall prepare a MR to see what happens | 10:00 |
| josch | if it works, we can start testing a bunch of stuff like this | 10:01 |
| josch | will of course not help with possible hardware-related regressions | 10:01 |
| josch | oh btw: i found some old ls1028a images on my disk | 10:01 |
| josch | back from march | 10:02 |
| josch | so i'll try that tomorrow | 10:02 |
| josch | (not today because it's too hot to go out) | 10:02 |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-64-230.tukw.qwest.net) | 10:11 | |
| + colinsane (~colinunin@97-113-64-230.tukw.qwest.net) | 10:16 | |
| - nybble (QUIT: Ping timeout: 255 seconds) (~sprang@user/csprng) | 10:23 | |
| ex-parrot | installer is almost done :) | 10:36 |
| ex-parrot | I did go leave it in the middle, it probably would have been done a while ago otherwisw | 10:36 |
| ex-parrot | ok, first boot I have run in to some trouble josch. I selected a LUKS with LVM install, and I just have "Waiting for encrypted source device..." looping on the LCD. I suspect maybe it's prompting me for creds on a serial port | 10:37 |
| josch | oh i see you complicated your test with luks :D | 10:38 |
| ex-parrot | "console=" is in the kernel cmdline five times | 10:38 |
| ex-parrot | I will fix that up and try again | 10:38 |
| josch | how do you "fix that up"? | 10:38 |
| josch | do you have a uart adapter? | 10:39 |
| josch | ex-parrot: the important part about console= is, that the last entry is console=tty1 | 10:39 |
| josch | because that's where the luks prompt will show up | 10:39 |
| ex-parrot | yeah, which it currently is, so I'm a bit suspicious that's not the entire problem here | 10:40 |
| ex-parrot | the initrd dropped me in to a console, so I can mount things manually and continue for now | 10:40 |
| josch | ex-parrot: this is the reform kernel and not the debian kernel, right? | 10:40 |
| josch | in the initrd, can you try find out why it didn't find the device listed in /etc/crypttab? | 10:41 |
| ex-parrot | yes | 10:41 |
| ex-parrot | suspiciously this is the bpo kernel | 10:41 |
| ex-parrot | I wonder if I have booted my old kernel... | 10:41 |
| ex-parrot | and that is why it can't find the device | 10:42 |
| josch | with the old initrd, yes | 10:42 |
| ch | which doesn't know what to look for, then | 10:42 |
| ex-parrot | yep cool, that'll be why | 10:42 |
| josch | u-boot will first search usb, then sd-card, then emmc for boot.scr or extlinux.conf | 10:42 |
| ex-parrot | I knew I had some stuff on the mmc for boot :) | 10:43 |
| ex-parrot | ok let me get this sorted out | 10:43 |
| ex-parrot | alright nice, shuffling over the contents of /boot on the nvme to the emmc has let me boot the new install | 10:45 |
| josch | oh you had /boot on the nvme? | 10:46 |
| ex-parrot | systemd decided to fsck for some reason, I wonder if the clock was wrong in the installer | 10:46 |
| ex-parrot | josch: yeah that's where the installer decided to put it | 10:46 |
| ex-parrot | which, I thought I had working previously, but maybe not | 10:46 |
| josch | ah right, you chose the defaults | 10:46 |
| ex-parrot | yeah just for interest's sake | 10:46 |
| josch | i think i need to add instructions that one has to choose manual partitioning and put /boot either on emmc or sd-card | 10:46 |
| ex-parrot | on other machines I've set this up with flash-kernel | 10:47 |
| ex-parrot | so /boot lives on the "non bootable" device and flash kernel copies over to the "real boot device" on updates etc | 10:47 |
| josch | ex-parrot: how did you do that? | 10:47 |
| ex-parrot | I can go grab an example later | 10:47 |
| josch | do you have a custom non-default boot.scr? | 10:47 |
| ex-parrot | the SPL can't read from nvme, but u-boot can right? so really only u-boot needs to be on emmc? | 10:48 |
| josch | u-boot can theoretically read from nvme once somebody implements it | 10:49 |
| ex-parrot | ah | 10:50 |
| josch | currently it can not, so it will load boot.scr, uimage and initrd from usb, sd-card or emmc | 10:50 |
| ex-parrot | sorry. my memory of how all this hangs together has gotten really faulty over the last 6 months :( | 10:50 |
| josch | ex-parrot: we should treat this as a problem with the documentation -- i have to write this down | 10:50 |
| ex-parrot | yeah, I should be writing proper lab notes for this work too | 10:50 |
| ex-parrot | I will endeavour to do so | 10:50 |
| ex-parrot | boot seems to have got stuck around starting sddm / cups | 10:50 |
| ex-parrot | but I can log in on tty2 | 10:51 |
| ex-parrot | something is repeating itself in the kernel cmdline, I've got ro twice, no_console_suspend twice, console=ttymxc0 three times and console=tty1 twice, cma=512M twice, pci=nomsi twice | 10:52 |
| ex-parrot | but at least the final console is console=tty1 | 10:52 |
| josch | ex-parrot: the repetition is because we cannot rely on u-boot putting in the correct values | 10:54 |
| josch | ex-parrot: but your u-boot is recent enough, so you get them twice | 10:54 |
| ex-parrot | ah nice | 10:54 |
| ex-parrot | ok, known issue | 10:54 |
| ex-parrot | sddm failed probably because it tried to start X, which I have had working previously on my debian install but not consistently | 10:55 |
| ex-parrot | manually starting KDE wayland worked fine | 10:55 |
| ex-parrot | it has some graphical glitches with blurred regions and the mouse pointer performance is kind of jank, but that happens on my M1 iMac too :/ | 10:55 |
| ex-parrot | https://img.hotplate.co.nz/NkCPSDNfDb | 10:57 |
| ex-parrot | and I'm out of time for tonight, but I'll make some notes on this tomorrow | 10:57 |
| ex-parrot | I reckon we could make flash kernel handle the nvme / mmc thing fairly elegantly, I may have already had it working on the rootfs I just blew away | 11:00 |
| ex-parrot | but I've got other examples from other machines, so I'll dig that out | 11:00 |
| josch | ex-parrot: these are my fixes for the web page so far: https://paste.debian.net/hidden/ec43d421/ | 11:04 |
| ex-parrot | looks good | 11:06 |
| ex-parrot | there's a small issue on https://reform.debian.net/repo/ too around the "option c" to enable backports | 11:06 |
| ex-parrot | it says to "execute" then there's just a raw source.list line | 11:07 |
| josch | right, that wording can be improved | 11:07 |
| ex-parrot | it's probably obvious to anyone actually following the instructions currently | 11:08 |
| josch | fixed locally | 11:08 |
| ex-parrot | alright, I'm out. I'll be back tomorrowish | 11:09 |
| ex-parrot | thanks for your continued excellent work josch | 11:09 |
| josch | ex-parrot: thank you for trying it out and reporting issues! :) | 11:09 |
| ex-parrot | depressingly I test software for a living, this is relatively friendly in comparison :P | 11:10 |
| josch | oh no XD | 11:10 |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-64-230.tukw.qwest.net) | 11:11 | |
| + colinsane (~colinunin@97-113-64-230.tukw.qwest.net) | 11:15 | |
| + mjw (~mjw@gnu.wildebeest.org) | 11:21 | |
| + chomwitt (~chomwitt@2a02:587:7a19:9600:1ac0:4dff:fedb:a3f1) | 11:25 | |
| - Asmadeus (QUIT: Ping timeout: 255 seconds) (~asmadeus@240b:13:8c80:d300::42:13) | 11:28 | |
| + Asmadeus (~asmadeus@user/meow/Asmadeus) | 11:28 | |
| - chomwitt (QUIT: Ping timeout: 268 seconds) (~chomwitt@2a02:587:7a19:9600:1ac0:4dff:fedb:a3f1) | 11:47 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-64-230.tukw.qwest.net) | 11:49 | |
| + colinsane (~colinunin@97-113-64-230.tukw.qwest.net) | 11:53 | |
| + chomwitt (~chomwitt@2a02:587:7a19:9600:1ac0:4dff:fedb:a3f1) | 11:59 | |
| - mjw (QUIT: Killed (erbium.libera.chat (Nickname regained by services))) (~mjw@gnu.wildebeest.org) | 12:22 | |
| * Guest3393 -> mjw | 12:22 | |
| + Guest7088 (~mjw@gnu.wildebeest.org) | 12:23 | |
| - colinsane (QUIT: Quit: bye) (~colinunin@97-113-64-230.tukw.qwest.net) | 12:31 | |
| + colinsane (~colinunin@97-113-64-230.tukw.qwest.net) | 12:36 | |
| - buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org) | 12:43 | |
| + buckket (~buckket@vps.buckket.org) | 12:44 | |
| josch | minute: failed to initialize kvm: No such file or directory -- should i still implement this? it will be very slow... | 12:46 |
| josch | also still no luck with creating a guix image on source.mnt.re | 12:50 |
| josch | I keep getting Throw to key `encoding-error' with args `("scm_to_stringn" "cannot convert wide string to output locale" 84 #f #f)'. | 12:51 |
| + bluerise_ (~bluerise@pc19f8b89.dip0.t-ipconnect.de) | 13:30 | |
| + chartreu1e (~chartreus@S0106908d78501d1d.cg.shawcable.net) | 13:31 | |
| + lexik_ (~lexik@93.185.97.218) | 13:32 | |
| - bluerise (QUIT: *.net *.split) (~bluerise@pc19f8b89.dip0.t-ipconnect.de) | 13:36 | |
| - chartreuse (QUIT: *.net *.split) (~chartreus@S0106908d78501d1d.cg.shawcable.net) | 13:36 | |
| - sir-photch (QUIT: *.net *.split) (~m-hy5poy@2a01:4f8:1c1b:4579::1) | 13:36 | |
| - laumann (QUIT: *.net *.split) (~quassel@user/laumann) | 13:36 | |
| - amk (QUIT: *.net *.split) (~amk@user/amk) | 13:36 | |
| - sterni (QUIT: *.net *.split) (~quassel@user/sterni) | 13:36 | |
| - kuno (QUIT: *.net *.split) (sid299442@id-299442.helmsley.irccloud.com) | 13:36 | |
| - lexik (QUIT: *.net *.split) (~lexik@93.185.97.218) | 13:36 | |
| + sir-photch (~m-hy5poy@2a01:4f8:1c1b:4579::1) | 13:44 | |
| + laumann (~quassel@user/laumann) | 13:44 | |
| + amk (~amk@user/amk) | 13:44 | |
| + sterni (~quassel@user/sterni) | 13:44 | |
| + kuno (sid299442@id-299442.helmsley.irccloud.com) | 13:44 | |
| - sir-photch (QUIT: *.net *.split) (~m-hy5poy@2a01:4f8:1c1b:4579::1) | 14:15 | |
| - laumann (QUIT: *.net *.split) (~quassel@user/laumann) | 14:15 | |
| - amk (QUIT: *.net *.split) (~amk@user/amk) | 14:15 | |
| - sterni (QUIT: *.net *.split) (~quassel@user/sterni) | 14:15 | |
| - kuno (QUIT: *.net *.split) (sid299442@id-299442.helmsley.irccloud.com) | 14:15 | |
| + sir-photch (~m-hy5poy@2a01:4f8:1c1b:4579::1) | 14:17 | |
| + laumann (~quassel@user/laumann) | 14:17 | |
| + amk (~amk@user/amk) | 14:17 | |
| + sterni (~quassel@user/sterni) | 14:17 | |
| + kuno (sid299442@id-299442.helmsley.irccloud.com) | 14:17 | |
| - laumann (QUIT: Quit: No Ping reply in 180 seconds.) (~quassel@user/laumann) | 16:53 | |
| + laumann (~quassel@2a0a-e5c0-2-2-0-c8ff-fe68-bef1.loves.ipv6.at.ungleich.ch) | 16:55 | |
| - mjw (QUIT: Quit: Leaving) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 17:42 | |
| + mjw (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:00 | |
| + nybble (~sprang@user/csprng) | 18:28 | |
| - chomwitt (QUIT: Ping timeout: 264 seconds) (~chomwitt@2a02:587:7a19:9600:1ac0:4dff:fedb:a3f1) | 18:30 | |
| minute | josch: maybe there's no utf8 locale? | 18:31 |
| - cobra (QUIT: Ping timeout: 240 seconds) (~cobra@user/Cobra) | 18:38 | |
| + cobra (~cobra@user/Cobra) | 18:41 | |
| ch | https://frame.work/gb/en/blog/introducing-a-new-risc-v-mainboard-from-deepcomputing | 19:03 |
| Twodisbetter | Yep, I bet the FOSS world is in love with that new mainboard already. | 19:13 |
| Twodisbetter | This makes a large part of what Framework is doing open source. | 19:13 |
| Twodisbetter | I had not idea when I first start "working" for them that this was a goal, but it is awesome to see. | 19:14 |
| minute | Twodisbetter: what exactly is open source there? | 19:16 |
| Twodisbetter | RISC-V | 19:18 |
| minute | Twodisbetter: RISC-V is just an instruction set architecture | 19:18 |
| Twodisbetter | Leadership there has said that they are driving to being able to release all of the schematics for their mainboards, but under x86 chips their have been NDAs, etc. | 19:18 |
| Twodisbetter | While this isn't completely there yet, it is just good to see them driving that way. Response has been big as well, which is good, becaue that means there will be continued push in this direction. | 19:19 |
| minute | Twodisbetter: i don't like this kind of imprecision and open washing. there's nothing open source here so far. the u74 cores are not open source cores. the JH7110 isn't more open than a typical ARM SoC. there are no sources for the motherboard. | 19:20 |
| minute | the framework is not an open hardware device | 19:21 |
| minute | why am i putting any effort into making OSHW when people don't care about the details anyway | 19:21 |
| ch | yeah don't see what is actually open there, but still interesting (also interesting on which parts they -do not- deliver) | 19:23 |
| noam | minute: because some of us do? :) | 19:24 |
| noam | Iffff they make that mainboard open source, that would be cool | 19:25 |
| minute | yeah. i'm just tired of people wanting to shove framework parts into my face. like, today i met the CEO of ifixit and they were also like, could you put these framework port modules into the reform | 19:25 |
| noam | minute: shoulda said "Sure, if they put Reform parts into the framework and open source enough of the schematics that we can make a framework/reform hybrid" or some such lmao | 19:26 |
| minute | i'm like, well, if someone wants to do that kind of thing, they could design a bay for it but i'm not interested | 19:26 |
| noam | Framework is having positive effects, and the repairability is an objectively good thing IMO | 19:27 |
| noam | but to call it open is disingenuous | 19:27 |
| minute | i agree. it is good that they care about repairability | 19:28 |
| minute | the way that the framework laptop is designed, the industrial production methods used, these are not attainable for makers, the devices are not reproducible | 19:28 |
| minute | it is a totally different way of constructing a thing | 19:28 |
| minute | constructing it to be manufactured in a big complex factory vs putting it in reach of makers | 19:29 |
| [tj] | framework replaced my mainboard for free when I killed it with an over voltage | 19:29 |
| noam | I think I _could_ make a Reform myself, with a great deal of effort, if I wanted to | 19:29 |
| [tj] | but there was not information to debug or fix it myself | 19:29 |
| noam | No chance I could do that for a framework | 19:29 |
| minute | noam: someone (jacqueline) has done it | 19:29 |
| [tj] | so much better than say dell, but not great | 19:29 |
| noam | minute: I know! | 19:29 |
| noam | Her reform is awesome :P | 19:29 |
| [tj] | the jh7110 documentation is an incredible joke | 19:30 |
| noam | [tj]: yeah, basically | 19:30 |
| noam | I mean, okay, lemme put it this way: framework puts the entire laptop industry to shame | 19:30 |
| noam | but that's because the laptop industry is horrible | 19:30 |
| minute | how was the replacement motherboard with better processor for some old thinkpad called? | 19:30 |
| noam | not because framework is Jesus | 19:30 |
| noam | :P | 19:30 |
| [tj] | I love this computer, but when framework goes out of business there won't be a community making new motherboards | 19:30 |
| noam | minute: frankenpad? | 19:30 |
| noam | :P | 19:30 |
| minute | hmm yeah? | 19:30 |
| noam | ori has one, I can ask him lol | 19:30 |
| minute | i mean that chinese board | 19:31 |
| noam | Right | 19:31 |
| noam | He has one :P | 19:31 |
| minute | some little company/team/person made a motherboard once... | 19:31 |
| vkoskiv | My understanding is that NDAs are so common in the hardware industry, framework couldn't publish motherboard schematics even if they wanted to? | 19:31 |
| minute | ok cool | 19:31 |
| noam | I think he has plan9 on it lo | 19:31 |
| noam | l | 19:31 |
| minute | vkoskiv: correct | 19:31 |
| noam | vkoskiv: I think that's true | 19:31 |
| [tj] | that was a cool effort, I don't think framework have sold enough to generate that sort of work from the community | 19:31 |
| noam | but it doesn't justify painting them as more than they are | 19:31 |
| minute | i've tried to get the design docs for intel compute element | 19:31 |
| minute | (which is probably cancelled now) | 19:31 |
| minute | so i have an NDA with intel | 19:31 |
| * mjw -> Guest2692 | 19:31 | |
| - Guest2692 (QUIT: Killed (molybdenum.libera.chat (Nickname regained by services))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 19:31 | |
| * Guest7088 -> mjw | 19:31 | |
| minute | but in the end they weren't giving me access anyway and also i couldn't have used that stuff | 19:32 |
| minute | for reform | 19:32 |
| minute | because the reference designs are under NDA | 19:32 |
| minute | lol. | 19:32 |
| vkoskiv | I get that, but I struggle to see what aspect about motherboard design is actually such secret sauce that publishing it would harm 'competitive advantage' or whatever | 19:32 |
| + Guest2692 (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 19:32 | |
| noam | Do I think framework _would_ open source more schematics if they could? Honestly, I believe it's plausible | 19:32 |
| minute | vkoskiv: it's just oldschool chip industry bullshit | 19:32 |
| noam | Their behavior has been pretty good, as far as I knonw | 19:32 |
| noam | but it's one thing to say "I think they're good folks" | 19:32 |
| noam | it's another to say "they've actually done this thing" | 19:32 |
| minute | noam: open source the altium/orcad files? doubtful. pdf? probably yeah | 19:33 |
| minute | but pine does that too | 19:33 |
| minute | or fairphone | 19:33 |
| vkoskiv | When people have asked about my reform and I've explained the open hardware aspect, people seem to bring up Framework as a similar example, when it's not really the same | 19:33 |
| minute | vkoskiv: yeah i think this is too abstract for a lot of people to grasp | 19:33 |
| minute | vkoskiv: what works for me is to say, you can clone this thing | 19:34 |
| minute | if we go out of business, you can still make one | 19:34 |
| vkoskiv | Yep, I do then draw comparisons to software freedom, and that seems to click | 19:34 |
| minute | i mean, the same is true for like commodores and classic macintoshes, because you can reverse engineer them by looking at them | 19:34 |
| noam | minute: mm, point | 19:34 |
| vkoskiv | Microsoft goes bust, you could certainly reverse-engineer a lot of their binaries with enough effort, but there are legal issues there, and it's a lot of work | 19:34 |
| noam | (regarding pdf vs source) | 19:35 |
| noam | vkoskiv: yea, but if they go bust then the legal issues are way less bad | 19:35 |
| minute | but you'll never be able to make an m2 macbook after they're discontinued... | 19:35 |
| noam | because they suddenly don't have lawyers lmao | 19:35 |
| minute | abandonware :D | 19:35 |
| minute | hmm, pretty cool what people have been doing with frankenpads | 19:36 |
| Zaba | yeah I think framework just has good marketing and unfortunately many people have no mental picture of what the “sources” for hardware are anyway | 19:44 |
| Zaba | and like the whole industry is tied up by NDAs because companies just don’t disclose things by default. they need a reason to *publish* something. | 19:45 |
| minute | yeah | 19:46 |
| Zaba | because, well, if you publish something and that turns out to have been a bad move for *whatever* reason, it’s hard to un-publish things, so people err on the side of caution | 19:46 |
| Zaba | for many companies, accidentally revealing somebody else’s (like, their clients’) trade secrets is a serious concern, because they’re more than just a few people who know each other well and who are doing what they definitely know is original work | 19:47 |
| Zaba | and yeah… I also feel like a lot of enthusiasm for RISC-V is misplaced - just because the ISA is royalty free doesn’t mean anything about the CPUs, let alone the SoCs, is meaningfully free or open | 19:49 |
| Zaba | and in fact many of those chips seem to be coming from some very dystopian parts of the world | 19:49 |
| josch | minute: the C.UTF-8 locale (which the script sets) is always available as provided by glibc (which as far as i understand it guix is also using) | 19:50 |
| noam | I think framework is also genuinely a step up | 19:50 |
| noam | Like... they don't deserve some of the positive press | 19:50 |
| noam | but they are _genuinely better_ | 19:50 |
| noam | and so I think some people go "Wow, they are _so much better,_ they solve _all the problems" | 19:51 |
| noam | when really, yes, they solve some of the problems, and AFAIK they aren't making things worse, but they are not perfect | 19:51 |
| Zaba | Well from a consumer perspective I guess so yeah | 19:55 |
| Zaba | From a supply chain perspective I dunno, I don’t think framework is doing much of anything interesting | 19:55 |
| Twodisbetter | minute: I'm not trying to shove Framework in your face, I also am not calling their platforms open. I merely mentioned that there are talks at that leadership levels and this is the direction they want to go. They are not there, and probably wont be there for a long time, but to be angry about that seems weird to me. Wouldn't you want as many companies as possible to be driving in that direction? | 20:02 |
| Twodisbetter | Furthermore, they are not calling themselves open at all. There is no marketing saying they are. They are simply open sourcing what they can as they can. | 20:02 |
| minute | Twodisbetter: > 19:13:52 <Twodisbetter>This makes a large part of what Framework is doing open source. | 20:03 |
| Twodisbetter | minute: also the ability to build the product yourself is something that appeals to a VERY small market. My guess is that over 90% of the people who have bought a Reform have no intention of ever doing that kind of thing even though they can. While I think it is totally awesome that this is a design mantra for MNT, I also don't see it as bad if other computer companies aren't doing that, simply be | 20:04 |
| Twodisbetter | cause a very small part of the population can even take part in that. | 20:04 |
| Twodisbetter | minute: I don't feel like that is wrong either. My comments about them I mean regarding a lot fo what they are doing as open source. | 20:05 |
| Twodisbetter | minute: compared to MNT, it is not a lot at all. Compared to HP or Dell it is a ton. | 20:05 |
| Twodisbetter | minute: I think it is great to see a more corporate entity caring about such things. That is all. | 20:06 |
| Zaba | for focus on the *act* of building it yourself is myopic | 20:06 |
| Zaba | it’s all about the supply chain | 20:06 |
| Twodisbetter | minute: MNT is awesome, and I have never indicated otherwise. Trust me, I love what you and the team are doing! | 20:07 |
| Zaba | now, it’s true that most people have no idea how most things they interact with are made and how they come to be, and that’s… kind of the thing | 20:07 |
| minute | i am closing this tab now and will try to get some work done | 20:08 |
| noam | Healthy /thumbs up | 20:09 |
| Twodisbetter | Healthy /thumbs up as well | 20:24 |
| AbortRetryFail | 👍 | 20:41 |
| AbortRetryFail | Oh darn, no unicode emoji on this box. :x | 20:41 |
| Twodisbetter | Those bridging from XMPP see it though. Nice thumbs up. | 20:43 |
| noam | I see it on IRC, too | 20:47 |
| noam | It just depends on the client :) | 20:47 |
| Twodisbetter | noam: what client are you using? | 20:48 |
| noam | SourceHut :P | 20:48 |
| noam | emersion's web client, they have an instance hosted | 20:48 |
| noam | (with a bouncer) | 20:48 |
| sigrid | it's just a unicode character. why wouldn't it work with any client, given a non-lacking font | 20:54 |
| AbortRetryFail | If things filter it out? | 20:54 |
| sigrid | I don't think there are any things that would do that | 20:54 |
| jfred | I do really wish MNT would get more attention the way Framework seems to. Feels like in tech journalism it's specs above all else though | 20:56 |
| jfred | Maybe the Pocket Reform could change that somewhat since it's got a pretty unique form factor, and using the same modules as the big Reform is a big deal | 20:58 |
| jfred | Would need to get some in the hands of reviewers I guess | 20:58 |
| - hairu (QUIT: Remote host closed the connection) (m-uotkmd@user/hairu) | 21:56 | |
| + hairu (m-uotkmd@user/hairu) | 22:01 | |
| AbortRetryFail | its not just specs. shiny also matters. | 22:17 |
| AbortRetryFail | i'm pretty happy with my Reform. I don't think I would have been able to fix my sound problems early on the way I did if the same thing happened with a Framework laptop. | 22:18 |
| AbortRetryFail | ACTION needs a reminder to re-test mic input with the new kernel... | 22:19 |
| minute | i was working on a worktime logging project and realized i designed the first reform2 motherboard almost exactly 5 years ago... june 2019 | 22:37 |
| josch | birthday celebration time? :) | 22:38 |
| minute | Twodisbetter: sorry i was grumpy earlier. i am a bit overworked at the moment and am very sensitive towards products+marketing in the same category being lauded for open source while not going all the way. if there are no precise category for what "open", "open source", "open hardware" means it feels devaluing to my/our efforts in that area. | 22:40 |
| minute | i would like things to be evaluated under similar lenses | 22:41 |
| minute | i mean, cool that they made a joint venture with another electronics company that made am experimental motherboard for them. | 22:42 |
| Twodisbetter | minute: I understand. I didn't take it personal. | 22:43 |
| minute | we had a joint venture with spanish company rbz who designed an OSHW processor module for reform several years ago | 22:43 |
| minute | you can download the altium files and/or get that thing made at pcbway yourself with parts from mouser/digikey | 22:43 |
| minute | (ls1028a) | 22:43 |
| minute | we also had a vexriscv powered version of reform (using the oshw kintex-7 module that you can also build yourself) | 22:44 |
| josch | wow https://mntre.com/media/reform_md/2019-01-14-status_update_on_reform.html is such a crazy article to read in 2024 :D | 23:18 |
| josch | minute: did you end up trying out atuin? if yes, do you think it's something that should be in debian? | 23:19 |
| minute | josch: yes right? it is quite fascinating and i'm happy that i wrote with so much detail back then | 23:20 |
| minute | also about the software i was using | 23:21 |
| minute | on imx6qp :0 | 23:21 |
| josch | it's really cool you put so many photos | 23:24 |
| josch | i wasn't involved with the reform back then, so most of this is news to me :D | 23:24 |
| minute | josch: nice! i will see if i can restore the video that is missing there | 23:25 |
| minute | josch: ok i installed atuin on my reform using the install script, that was painless | 23:25 |
| josch | i only read atuin's features and they do sound enticing | 23:26 |
| josch | i often find myself in a situation where i know i ran a command but apparently keeping a history of 10k commands is not enough | 23:26 |
| + chomwitt (~chomwitt@2a02:587:7a19:9600:1ac0:4dff:fedb:a3f1) | 23:32 | |
| minute | ok, the video is now restored! https://mntmn.com/media/videos/reform-beta-hand-assembly.mp4 | 23:33 |
| josch | wow, those half-moon buttons :D | 23:35 |
| josch | the mainboard is so tiny! | 23:35 |
| minute | yeah :D oh it had a lid sensor | 23:35 |
| minute | yes, tiny mainboard and tiny som | 23:35 |
| josch | i have never seen a pouch cell lifepo4 | 23:36 |
| josch | how did the lid sensor work? | 23:36 |
| josch | it's on my list of mods that i need to have :) | 23:36 |
| josch | the heat sink!! XD | 23:36 |
| minute | the heat sink was too smol ;/ that's why there was a fan ;/ | 23:39 |
| minute | josch: the lid sensor was just a hall effect sensor and then in the screen frame there was a matching magnet | 23:40 |
| minute | we had a tinyAVR or whatever it was called back then, an 8=bit mcu, and that was reading that sensor and sending sleep/wake commands to the imx | 23:41 |
| minute | yes, it all worked fine, and also there was graphical uboot and everything. | 23:41 |
| minute | imx6 was really quite well supported | 23:41 |
| josch | right now, the magnets are in the body, so the hall sensor would have to go into the screen part... hrm... | 23:42 |
| josch | graphical uboot is so neat -- i'm missing that on the a311d | 23:42 |
| josch | it's a tiny thing of course... | 23:42 |
| josch | i thought about having an open issue about this but then i doubt that it's magically going to get fixed by somebody anytime soon | 23:43 |
| minute | josch: no it was just a tiny sticky magnet that was only for triggering the sensor | 23:43 |
| minute | like a coin | 23:43 |
| minute | josch: yeah it's probably more realistic to get rk3588 graphical uboot | 23:44 |
| minute | or edk2... | 23:44 |
| josch | the reform 2 went such a long way compared to the reform V0.4 | 23:51 |
| minute | yeah... funnily the pocket motherboard is a lot more like the reform 0.4 | 23:52 |
| minute | ports on one side, has a slot for cell modem :D | 23:52 |
| minute | tomorrow the first reform next motherboard pcb is supposed to show up | 23:53 |
| josch | \o/ | 23:54 |
| josch | after two weeks of shipment, according to UPS, tomorrow this might be at my door: http://handheldsci.com/kb/ | 23:55 |
| josch | i'll plug it into my reform keyboard/trackball combo and see what happens :) | 23:56 |
| josch | it says that it's compatible with built-in hubs... lets see... | 23:56 |
| Twodisbetter | josch: I would be interested in seeing how that goes. Nice piece of kit. | 23:57 |
| minute | josch: ohh, looking forward | 23:57 |
| Twodisbetter | I would like Bluetooth on my Reform as well, so if it works out, seems like the perfect solution. | 23:58 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!