#emc-devel | Logs for 2008-08-13

[11:55:15] <cradek_> cradek_ is now known as cradek
[20:09:11] <cradek> jepler: I wonder if the max length of the backplot is too long.
[20:11:12] <jepler> mmmmmaybe
[20:11:37] <cradek> I'm not sure I've ever seen it undraw stuff, although I recall you added taht feature
[20:16:40] <jepler> yes but I made the buffer pretty big .. 100000 apparently
[23:30:21] <jepler> we either need to get our asses in gear, define what will be in 2.3 and finish those features; or we need to do something else like backport some features from TRUNK into a branch that we think are well-tested and separable
[23:31:44] <jmkasunich> probably should be the former
[23:31:47] <jepler> yeah ideally
[23:32:03] <jepler> I'm just realizing there is stuff on TRUNK that makes my life better -- but it's not in 2.2 because it's new features
[23:32:06] <jmkasunich> I'm afraid the joints-axes stuff that I was talking about won't happen
[23:32:39] <jmkasunich> I've got several machining projects that are absorbing my interest at the moment, so we shouldn't hold up 2.3 for stuff I said I'd do
[23:32:58] <jepler> this isn't intended as a call-out of anyone
[23:33:06] <jmkasunich> understood
[23:33:27] <jepler> I wonder what's broken in TRUNK
[23:34:05] <jmkasunich> no clue
[23:34:26] <jmkasunich> I should get a trunk checkout on the shoptask (if I don't already have one) and use it when I'm machining
[23:34:35] <jepler> that depends on your goals
[23:35:14] <jepler> I would recommend using a stable version for day-to-day work even if you're a developer
[23:35:42] <jmkasunich> then I won't do _any_ testing
[23:35:56] <jmkasunich> the only time I "test" is when I'm specifically looking for something
[23:36:06] <jmkasunich> either testing a fix, or trying to reproduce a reported problem
[23:36:23] <jmkasunich> (or testing a new feature)
[23:37:10] <jepler> well -- instead of doing that, I just spent 15 minutes backpoting two features in TRUNK I decided I wanted in 2.2
[23:37:25] <jepler> (the new no-popup operator messages and the less new unhome)
[23:37:38] <jmkasunich> they don't involve any ini changes, right?
[23:37:41] <jepler> no
[23:37:48] <jmkasunich> seems safe to me
[23:38:36] <jepler> well -- changing what happens when operator messages pop-up risks causing them to be lost, or causing worse user interaction problems than what we have now with the pop-up
[23:38:47] <jepler> "unhome" is likely to be lower impact
[23:45:14] <cradek> hmmmm what was I just talking about missing today?
[23:46:07] <jepler> cradek: you tell me
[23:46:23] <cradek> I would if I could remember
[23:46:23] <jepler> but you don't even use emc -- you just run boss all day
[23:46:50] <jmkasunich> something for the HNC?
[23:47:02] <jepler> I've got to say that not having the pop-up makes the "insert drill XXX" and "start spindle" operator messages much less of a hassle
[23:47:53] <cradek> yeah I like that. I will test it too.
[23:47:54] <jmkasunich> if you do "(MSG, do something idiot)" followed by M0, you can just hit resume once you've done whatever?
[23:48:10] <jepler> yeah -- you can clear the message whenever you feel like it
[23:48:10] <cradek> since it changes key bindings it's possibly messy
[23:48:17] <jepler> http://emergent.unpy.net/files/sandbox/notifications.png if you missed it
[23:48:40] <jmkasunich> "whenver you feel like it" includes never? (new message replaces old?)
[23:48:57] <jepler> they sort of "scroll up"
[23:49:09] <jmkasunich> the box eats up more and more of the preview window?
[23:49:26] <jmkasunich> or you just see the last few lines?
[23:49:59] <jepler> http://emergent.unpy.net/index.cgi-files/sandbox/multiple-notifications.png
[23:50:06] <jepler> the bottom one is the latest one
[23:50:12] <jmkasunich> cool
[23:50:21] <jepler> click "X" or press space to clear one, ctrl-space to clear all
[23:50:57] <cradek> I wonder if a black line (or rather, one pixel space) between them would help you understand that they are separate
[23:51:35] <jepler> some kind of border might clarify it
[23:52:42] <jepler> woo my spindle shipped
[23:52:48] <cradek> yay
[23:55:52] <jmkasunich> jepler: regarding debounce...
[23:56:01] <jmkasunich> are you going right into an FPGA pin?