#linuxcnc-devel Logs

Dec 01 2022

12:30 AM pere: linuxcnc-build: force build --branch=master 0000.checkin
12:37 AM linuxcnc-build: build forced [ETA 2h30m38s]
12:37 AM linuxcnc-build: I'll give a shout when the build finishes
02:36 AM linuxcnc-build: Hey! build 0000.checkin #9447 is complete: Success [3build successful]
02:36 AM linuxcnc-build: Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/9447
06:21 AM -!- #linuxcnc-devel mode set to +v by ChanServ
06:38 AM -!- #linuxcnc-devel mode set to +v by ChanServ
06:44 AM -!- #linuxcnc-devel mode set to +v by ChanServ
07:45 AM pere: Anyone know anything about the issue mentioned in <URL: https://lists.debian.org/debian-legal/2022/12/msg00000.html > ?
07:52 AM pere: andypugh: I believe <URL: https://github.com/LinuxCNC/linuxcnc-live-build/pull/18 > and <URL: https://github.com/LinuxCNC/linuxcnc-live-build/pull/19 > are required to get the bookworm live iso build working, and recommend including <URL: https://github.com/LinuxCNC/linuxcnc-live-build/pull/20 > for correct file naming.
09:49 AM -!- #linuxcnc-devel mode set to +v by ChanServ
10:41 AM linuxcnc-build: build #5123 of 4009.deb-precise-rtai-i386 is complete: Failure [4failed shell_2] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/4009.deb-precise-rtai-i386/builds/5123 blamelist: Hans Unzner <hansunzner@gmail.com>, andypugh <andy@bodgesoc.org>
11:50 AM Tom_itx is now known as Tom_L
12:51 PM andypugh: pere: No, I don’t know anything other than that we don’t know enough.
12:51 PM andypugh: There is also some code that the original auther refused to relicense.
12:52 PM andypugh: I am surprised that we are claiming LGPL though, nearly all the license statements in the code claim GPL 2+
12:55 PM pere: andypugh: right. the response in that thread claimed the original poster had misunderstood completely, and I do not know, but thought it best to mention it here.
12:56 PM CaptHindsight[m]: pere: https://www.mail-archive.com/emc-developers@lists.sourceforge.net/msg23329.html
12:56 PM andypugh: Actualy reading the (long, complicated) copyright file does actually have some Procter/Shackleford code with an LGPL license.
12:57 PM pere: andypugh: btw, at least part of the live boot screen lock fix worked. not quite sure if it was the removal of light-locker or the xscreensaver override, though. tried to express in the pull request that it needed testing, but can confirm that things improved, at least.
01:02 PM andypugh: And, it is indeed true that the source files (which may or may not have been released as PD) now have an LGPL (or sometimes GPL) in the header.
01:03 PM pere: as far as I know, that is ok.
01:10 PM andypugh: Who _can_ rule definitively on these issues?
01:11 PM andypugh: (Apart from this issue there is the perennial EtherCAT license issue. We don’t include anything EtherCAT because we don’t think we are allowed to.
01:11 PM pere: a judge when someone sue
01:12 PM andypugh: And they would sue me, personally, as the packager, presumably?
01:12 PM pere: there is also the <URL: https://sfconservancy.org/activities/ > that could give advice, I guess.
01:13 PM pere: andypugh: are you rich?
01:13 PM andypugh: And possibly it would go diffferently in different legal jurisdictions?
01:14 PM pere: (there is no point suing poor people)
01:16 PM andypugh: To sue anyone you would have to prove loss/damage. Who has lost, and what, by our possibly faulty spelling of copyright descriptions?
01:18 PM CaptHindsight[m]: pere: assert dominance, ruin their day etc etc
01:18 PM pere: no idea. you asked for definite answers, and as far as I know only a judge can provide those.
01:19 PM andypugh: Is it worth mentionong in that thread that the appropriation of EMC into EMC2 was done with the help and support of the original authors? We even had a hat with Will Shackleford on the phone at the Wichita meetup several years ago to discuss NML.
01:21 PM CaptHindsight[m]: andypugh: have you looked at how much work there is to get machinekits zeromq into LCNC to replace NML?
01:22 PM CaptHindsight[m]: i think they used zeroMQ
01:22 PM pere: is nml used elsewhere outside linuxcnc?
01:24 PM andypugh: I don’t think that they aver actually did that. There was a lot of talk about it, but if you look in the Machinekit source NML still seems to be there.
01:25 PM CaptHindsight[m]: the MK server is down so I can't read their docs
01:26 PM CaptHindsight[m]: ah it was cached by google https://webcache.googleusercontent.com/search?q=cache:k5gbxRj7AXAJ:https://machinekoder.com/machinetalk-explained-part-3-technologies/&cd=1&hl=en&ct=clnk&gl=us&client=firefox-b-1-d
01:26 PM andypugh: pere: I am not sure. It might have been historically. There are still docs on it: https://www.nist.gov/ctl/smart-connected-systems-division/networked-control-systems-group/nml-programmers-guide-c-version
01:27 PM CaptHindsight[m]: https://www.researchgate.net/publication/243775807_The_Neutral_Message_Language_A_Model_and_Method_for_Message_Passing_in_Heterogeneous_Environments
01:29 PM CaptHindsight[m]: IF I was going to replace NML in LCNC I'd likely go with DDS that ROS2 uses
01:29 PM CaptHindsight[m]: https://www.researchgate.net/publication/309128426_Exploring_the_performance_of_ROS2
01:29 PM andypugh: CaptHindsight[m]: OK, so they did include 0MQ in machinetalk, but I am not clear if that replaced all of NML in the controller as a whole.
01:30 PM andypugh: 0MQ was probably a good choice back in 2014 or so, when it was chosen.
01:30 PM CaptHindsight[m]: Tormach did some work to bridge ROS2 with LCNC
01:32 PM CaptHindsight[m]: https://github.com/zultron/hal_ros_control/
01:32 PM pere: did they try to upstream it?
01:33 PM pere: the linuxcnc project have not been able to merge three year old patches from tormach yet.
01:34 PM CaptHindsight[m]: ah was a bridge for ROS2 to mchinekit hal
01:43 PM pere: andypugh: btw, I took the liberty to tag a few of my pull requests with 'affects 2.9' to indicate that they are intended for the 2.9 release. is this a good way to do it?
01:44 PM andypugh: Yes, that’s a good plan
01:44 PM andypugh: pere: The Tormach patches don’t fit, as they are based on Machinekit and don’t have the joints-axes changes
01:47 PM pere: andypugh: I talk about <URL: https://github.com/LinuxCNC/linuxcnc/pull/581 > from 2019.
01:48 PM pere: If my patches were treated like that, I would not waste much time sending more patches.
01:49 PM pere: I some times create a trivial change to a project I am interested in improving, to see if how it is handled, before I spend time improving the project. I assume others do the same, to avoid spending time on dead projects.
01:50 PM pere: this is part of the reason why I nag so much about getting the number of open pull requests down.
01:51 PM andypugh: Yes. So am I now, after making my statement from a postition of ignorance I am looking at the code and the conflicts are smaller than I anticipated. In fact I see little to no difficulty in sorting out that PR.
01:51 PM andypugh: Thoigh how well it it would then propogate to 2.8. 2.9, 2.10....
01:53 PM pere: aha. I suspect our users will like it.
01:55 PM pere: but I would never send a patch blob like was done in #581, as it make it close to impossible to review the changes.
01:57 PM pere: andypugh: I am testing the xscreensaver fix on bookworm. suspect the same should be applied in buster.
02:06 PM pere: andypugh: is it time to merge #2112 into master?
02:13 PM andypugh: Yes, done. let’s see what breaks
02:25 PM pere: andypugh: live locking issue believed to be fixed, patches in <URL: https://github.com/LinuxCNC/linuxcnc-live-build/pulls >.
02:29 PM pere: LinuxCNC_2.9pre-amd64.hybrid.iso is 3.3G. Is that a normal size?
02:37 PM CaptHindsight[m]: https://cdimage.debian.org/cdimage/unofficial/non-free/cd-including-firmware/11.5.0-live+nonfree/amd64/iso-hybrid/
02:37 PM CaptHindsight[m]: aren't they all this size for a Live + non-free?
02:38 PM pere: no idea. guess it depend on arch and package set
03:56 PM pere: seb_kuzm1nsky: could <URL: https://github.com/LinuxCNC/wlo/pull/28 > work for linking to translations?