#linuxcnc-devel | Logs for 2014-05-11

[17:04:36] <cradek> huh, I think for the first time in the 11 years I've had a gpg key available on my website and on the keyservers, a stranger has sent me an everyday email, encrypted.
[17:08:02] <cradek> seb_kuzminsky: I was just looking for you...
[17:11:39] <cradek> seb_kuzminsky: I've tried every way to get the livecd built with our kernel on it, but I think our kernel packaging is broken, possibly in more than one way. my attempts to get and rebuild it failed too - as far as I can tell there's no corresponding source from which to change/reproduce the deb I have
[17:12:09] <cradek> there's the corresponding linux-source package which just sticks a tarball in /usr/src
[17:12:31] <cradek> but that doesn't seem right - I expected to be able to use apt-get source like usual
[17:13:46] <cradek> hm if I extract it, it does have a debian directory...
[17:19:07] <pcw_home> I notice a lot of people have trouble with the hal net command needing a signal
[17:19:09] <pcw_home> I wonder if it makes things better or worse to have anonymous signals
[17:19:10] <pcw_home> either
[17:19:12] <pcw_home> net pin pin pin
[17:19:13] <pcw_home> or
[17:19:15] <pcw_home> link pin pin pin
[17:19:39] <cradek> we originally had linkpp, which would use the pin name as the signal name
[17:19:43] <cradek> it was utterly confusing
[17:20:17] <pcw_home> sometimes the extra name is just a distraction though
[17:20:56] <cradek> would those just not show up in the signals list of all the utilities?
[17:21:05] <pcw_home> most schematic capture/layout software have them
[17:21:37] <pcw_home> they typically show up like ***NNN
[17:24:02] <cradek> I'm not against the idea of anonymous signals - but I don't think it's a solution to the problem of people not knowing the hal concepts well enough to be able to use the net command
[17:24:45] <cradek> I'm not sure it would leave them less baffled, but I guess at least their hal file (that they don't really understand) would run
[17:25:12] <pcw_home> yeah it a beginner problem but i do have some sympathy since the extra names do add some clutter
[17:25:14] <cradek> does the net command give a crappy error that could be improved instead?
[17:25:19] <KimK> Hi gents. I like having the signal name because it serves as the "wire number", but instead of just being a number, it (hopefully) adds functional info.
[17:26:29] <pcw_home> It can but sometimes it just effectively means you need to make a clone of the pin name
[17:28:01] <cradek> I guess I have never found myself doing that. I've always used something descriptive.
[17:28:45] <cradek> but I suppose motion.spindle-whatever is descriptive enough, telling that the signal is for whatevering the spindle
[17:29:01] <pcw_home> well if you have multistage logic for example, labeling every inner term is more of a pain than it needs to be
[17:29:42] <cradek> true
[17:30:00] <cradek> you could still use linkpp in those cases, but it only takes two pins
[17:30:19] <cradek> I might be tainted because I use classicladder whenever any logic is remotely complex
[17:30:23] <KimK> Yes, that (cloning pin name) can happen, especially with a long list of simple repeated functions (limit sws, etc.?) Maybe I should add something to the docs about helping yourself by providing good labels/names?
[17:30:25] <cradek> it has the internal variables for you
[17:30:47] <pcw_home> kind of like a schematic you really dont want to label every wire
[17:31:56] <cradek> it would be not very hard to have: net anonymous pin pin pin
[17:32:12] <cradek> halcmd could treat that specially and make special unique names
[17:33:02] <pcw_home> not a big deal, I just notice people stumble on that a lot
[17:33:18] <cradek> you'd change the results for, but not break, configs that happened to have a net called anonymous
[17:34:13] <cradek> I'd thoughtfully consider a patch that adds this
[17:34:32] <KimK> OT, I hope you guys are doing well
[17:34:45] <cradek> I am, same to you
[17:34:56] <KimK> Thanks
[17:35:27] <pcw_home> I'm good
[17:36:47] <pcw_home> finally got some hot weather after weeks of 50/60s
[17:37:21] <KimK> Good to know, we rely on you and your products more and more, it seems.
[17:37:38] <Tom_itx> we had over 100 last week
[17:37:46] <Tom_itx> mostly 90's
[17:38:18] <KimK> Hi Tom_itx
[17:38:25] <Tom_itx> hello
[17:38:43] <pcw_home> very rarely get that hot here (hot for us is 80s)
[17:38:55] <Tom_itx> been windy all day here
[17:39:46] <Tom_itx> probably get some storms later this evening
[17:41:07] <pcw_home> Same here (windy) but a big change in temperature
[17:41:08] <pcw_home> (our squash that were in suspended animation since they were planted
[17:41:10] <pcw_home> seem to have doubled in size in a couple days)
[17:42:14] <pcw_home> and the sheep are sleeping in the rain shelter for shade
[22:17:24] <skunkworks> logger[psha]:
[22:17:24] <logger[psha]> skunkworks: Log stored at http://psha.org.ru/irc/%23linuxcnc-devel/2014-05-12.html