code logs -> 2013 -> Thu, 21 Mar 2013< code.20130320.log - code.20130322.log >
--- Log opened Thu Mar 21 00:00:30 2013
00:05 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code
00:05 mode/#code [+qo Vornicus Vornicus] by ChanServ
00:05 You're now known as TheWatcher[T-2]
00:11
<&Derakon>
Yeah, the microscope computer is slightly more powerful than my laptop~
00:14 You're now known as TheWatcher[zZzZ]
00:22
<@Reiv>
Derakon: What rate of image gathering do you end up with?
00:24
<&Derakon>
Uh, for the inputs to this algorithm? We can do something like 5/sec.
00:27
<@Reiv>
And the algorathm can output 50 in 5-10 sec
00:27
<@Reiv>
Sssso it's still fast enough you could almost rig it realtime.
00:28
<&Derakon>
Mm, I guess I could run it that way; currently it's set to do the images first and then batch process them.
00:28
<@Reiv>
That's also legit
00:29
<@Reiv>
My thought was that if it started the batch before the photos were done (if there was the capacity, as it were) you could negate most of that lag.
00:29 * Derakon nods.
00:30
<@Reiv>
Obviously, the photos themselves take priority
00:31
<@Reiv>
But unless the computer is the bottleneck there (which is doubtful), you could easily enough nick the spare cycles in between shots.
00:31
<&Derakon>
The bottleneck is in waiting for confirmation that the sample holder has finished moving.
00:31
<&Derakon>
Which we have to do by querying the position twice and noting that it has not changed.
00:35
<@Reiv>
Reasonable
00:36
<&McMartin>
Hopefully it is not vibrating >_<
00:36
<&Derakon>
Well, not changed significantly. We ignore variation on the nanometer scale.
00:37
<&McMartin>
Heh, not quite what I meant, but now that you mention it, yeah, that too >_>
00:37 * McMartin was thinking "it's moved, then come back, between checks"
00:37
<&Derakon>
Ah.
00:37
<&Derakon>
Yeah, that's...improbable.
00:37
<&Derakon>
I suppose it does qualify as a bug though.
00:38 * Derakon closes; cannot replicate~
00:38
<&McMartin>
WORKSFORME
01:56 VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: Program Shutting down]
02:30 timmytwot [TimmyTwoTo@Nightstar-74b42d14.fios.verizon.net] has joined #code
02:31
< timmytwot>
http://www.youtube.com/watch?v=8WPafswrC64
02:31 timmytwot [TimmyTwoTo@Nightstar-74b42d14.fios.verizon.net] has quit [Connection closed]
03:00 Kindamoody[zZz] is now known as Kindamoody
03:12 syksleep [the@Nightstar-09f5d042.iinet.net.au] has joined #code
04:30 syksleep is now known as Syk
04:54 OrthiaLap [orthia@Nightstar-fae4ee48.ihug.co.nz] has quit [Ping timeout: 121 seconds]
05:16 himi [fow035@D741F1.243F35.CADC30.81D435] has quit [Ping timeout: 121 seconds]
05:19 thalass [thalass@Nightstar-f97b970e.bigpond.net.au] has joined #code
05:32 Derakon is now known as Derakon[AFK]
06:34 Alek [omegaboot@Nightstar-56dbba0f.in.comcast.net] has quit [[NS] Quit: ]
06:36 Alek [omegaboot@Nightstar-56dbba0f.in.comcast.net] has joined #code
06:36 mode/#code [+o Alek] by ChanServ
07:07 celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has quit [[NS] Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.]
07:14 thalass [thalass@Nightstar-f97b970e.bigpond.net.au] has quit [Ping timeout: 121 seconds]
07:16 Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: shutting down!]
07:27 thalass [thalass@Nightstar-f97b970e.bigpond.net.au] has joined #code
07:30 Reiver [quassel@Nightstar-3762b576.co.uk] has quit [Ping timeout: 121 seconds]
07:30 Reiver [quassel@Nightstar-3762b576.co.uk] has joined #code
07:37 Reiver [quassel@Nightstar-3762b576.co.uk] has quit [Ping timeout: 121 seconds]
07:37 Reiver [quassel@Nightstar-3762b576.co.uk] has joined #code
--- Log closed Thu Mar 21 07:38:03 2013
--- Log opened Thu Mar 21 07:38:10 2013
07:38 TheWatcher[zZzZ] [chris@Nightstar-3762b576.co.uk] has joined #code
07:38 Irssi: #code: Total of 34 nicks [22 ops, 0 halfops, 0 voices, 12 normal]
07:38 mode/#code [+o TheWatcher[zZzZ]] by ChanServ
07:41 Irssi: Join to #code was synced in 210 secs
08:15 You're now known as TheWatcher
08:22 Kindamoody is now known as Kindamoody|afk
08:33
<@Azash>
http://www.reddit.com/r/netsec/comments/1ansul/formerly_top_secret_nsa_cryptolog s_from_19741997/
08:45 himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code
08:45 mode/#code [+o himi] by ChanServ
09:26 thalass [thalass@Nightstar-f97b970e.bigpond.net.au] has quit [[NS] Quit: back laterz]
11:07 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving]
11:13 RichyB [richardb@Nightstar-228a334c.plus.com] has joined #code
11:43 Kyte|Work [c80e6b02@Nightstar-4fab16c5.mibbit.com] has joined #code
12:50 Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code
12:50 mode/#code [+o Pandemic] by ChanServ
13:25 Kyte|Work [c80e6b02@Nightstar-4fab16c5.mibbit.com] has quit [[NS] Quit: http://www.mibbit.com ajax IRC Client]
13:25 Kyte|Work [c80e6b02@Nightstar-992d69e3.mibbit.com] has joined #code
13:43 Kindamoody|afk is now known as Kindamoody
14:34 celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has joined #code
14:34 mode/#code [+o celticminstrel] by ChanServ
15:05 Kyte|Work [c80e6b02@Nightstar-992d69e3.mibbit.com] has quit [[NS] Quit: http://www.mibbit.com ajax IRC Client]
15:05 Kyte|Work [c80e6b02@Nightstar-36f67fd0.mibbit.com] has joined #code
15:39
<@celticminstrel>
Switching to rot.js breaks stuff because they inject functions into prototypes. :/
15:41
< Syk>
D:
15:43
<@celticminstrel>
I'm not going to let this stop me, though.
15:44
<@celticminstrel>
Because switching to it also made the thing less laggy.
15:47
<@celticminstrel>
Actually, I could make a ticket for this as well.
15:49
<@celticminstrel>
Because the project is on github.
15:56 Kyte|Work [c80e6b02@Nightstar-36f67fd0.mibbit.com] has quit [[NS] Quit: http://www.mibbit.com ajax IRC Client]
15:56 Kyte|Work [c80e6b02@Nightstar-36f67fd0.mibbit.com] has joined #code
15:58
<@celticminstrel>
Resting is quite a bit faster too, now.
15:59
<@celticminstrel>
So the bottleneck was display after all.
16:07 Kyte|Work [c80e6b02@Nightstar-36f67fd0.mibbit.com] has quit [[NS] Quit: http://www.mibbit.com ajax IRC Client]
16:16 iospace is now known as FauxRoscoe
16:21 ToxicFrog is now known as ToxicFrog|W`rkn
16:21
<&ToxicFrog|W`rkn>
Me: yay, I get to submit my first CL!
16:21
<&ToxicFrog|W`rkn>
Server: hold up there, buddy. The tests haven't finished running.
16:21
<&ToxicFrog|W`rkn>
Me: oh. How many tests are there?
16:21
<&ToxicFrog|W`rkn>
Server: well, let's see what you've changed...oooooooo. The answer is ALL OF THEM.
16:26 FauxRoscoe is now known as iospace
16:29
<@celticminstrel>
CL?
16:32
< RichyB>
Change Landing or something?
16:33
< RichyB>
I assume that it denotes "merge my feature branch into the master branch and mark it to be sent to QA and then onward towards production."
16:33
<&ToxicFrog|W`rkn>
Changelist.
16:33
<&ToxicFrog|W`rkn>
Basically a commit.
16:34
< RichyB>
/?
16:34
<&ToxicFrog|W`rkn>
?
16:35
< RichyB>
Are ye not, like, using git and working in branches?
16:39
<&ToxicFrog|W`rkn>
The backend is Perforce (+ lots of Google specific extensions, integration with the test and code review systems, etc). There's support for talking to it from other VCSes, though.
16:40
<&ToxicFrog|W`rkn>
So what's actually happening is that I work in git, and when a feature branch is ready, that gets turned into a P4 changelist, tested, reviewed, etc, and once committed, shows up back in git as a merge of that feature branch back into master.
16:44
< RichyB>
Is a "P4 changelist" closely analogous to a github pull request?
16:44
<&ToxicFrog|W`rkn>
Well, it's a centralized VCS, so not really
16:44
<&ToxicFrog|W`rkn>
It's more like a git index
16:44
<&ToxicFrog|W`rkn>
i.e. a set of logically related changes that aren't a commit yet
16:45
<&ToxicFrog|W`rkn>
If you're using naked P4, the way it works is you edit a bunch of stuff, stage changes into a CL (which the server knows about and can thus e.g. show to other users for code review), and then submit that CL, turning it into a commit.
16:46
<&ToxicFrog|W`rkn>
Basically it's the git add -> git commit split, except all of the indexes are stored on the server rather than in your local repo.
16:47
<&ToxicFrog|W`rkn>
In terms of how it's used here, yes, it's kind of github-pull-request-y.
16:52 Courage [Moltare@583787.FF2A18.190FE2.4D81A1] has quit [Ping timeout: 121 seconds]
16:53 Kyte|Work [c80e6b02@Nightstar-992d69e3.mibbit.com] has joined #code
16:53 Courage [Moltare@583787.FF2A18.190FE2.4D81A1] has joined #code
16:53 mode/#code [+o Courage] by ChanServ
16:55 Kindamoody is now known as Kindamoody|afk
17:02
<@celticminstrel>
I don't remember installing it, but apparently I have a QuickLook plugin that formats Markdown documents.
17:06 d4de [olorin@687AAB.418D22.B05943.212CEB] has quit [Ping timeout: 121 seconds]
17:11
< RichyB>
ToxicFrog|W`rkn: hm. That sounds sane.
17:13
< RichyB>
I would not be happy to use P4 like that unless I knew that there was a person or better an entire team dedicated full-time to supporting it, but I guess centralised VCS can work just great if it's written by people who take pains to accommodate the same workflows that svn/cvs don't.
17:13
< RichyB>
I assume that Google have a "Perforce department" or something?
17:31
<&ToxicFrog|W`rkn>
Or something, yes.
17:32
<&ToxicFrog|W`rkn>
This isn't an ad hoc team specific thing, this is a fully integrated, well supported, company wide version control system.
17:36
< Syk>
does linus torvalds have to come and give another git talk
17:38
<@Tamber>
Yes, the git has to turn up to give a talk about gits.
17:38
<@Tamber>
And then everyone will roll their eyes and get back to what they were doing. :p
17:39 * Syk beats Tamber with a 40-byte file
17:40 * Tamber hits Syk with a driveshaft.
17:41
<@Alek>
cat Tamber > /dev/null
17:41
< Syk>
mv Tamber /dev/mordor
17:42 Syk is now known as syksleep
17:42
<@Tamber>
:> Alek; :> syksleep
17:43
<@Tamber>
:)
17:43
< syksleep>
slep tiem
17:44
<@Alek>
http://pics.nase-bohren.de/kilo-mega-giga.jpg
17:54
<@gnolam>
Brought to you by the people who don't know when to use PNG.
17:56
<@Alek>
http://pics.nase-bohren.de/linus_windows7.jpg
18:03 d4de [olorin@687AAB.418D22.D394B7.F781AC] has joined #code
18:04
<&ToxicFrog|W`rkn>
Augh
18:04
<&ToxicFrog|W`rkn>
Why does list.reverse() modify the list in place and return None
18:04
<&ToxicFrog|W`rkn>
WHY
18:07
<@EvilDarkLord>
Because you already get a reversed copy of the list with list[::-1]?
18:10
<&ToxicFrog|W`rkn>
It should be called reverse! or something~
18:10 * ToxicFrog|W`rkn is no longer used to functions that mutate stuff in place
18:10
<&McMartin>
Also, man, I miss systems where internal piecewise tests actually work >_<
18:11 * McMartin reverts some changes he refined over a few days because while it was basically adding a time delay to a component startup, this causes a complete system failure two minutes later when deployed as part of the full suite
18:11
<&McMartin>
This product really is teaching me bad habits
18:36
<&ToxicFrog|W`rkn>
Oh hey. We have a command line tool that checks your diffs against the ownership of the affected file and tells you who you should be asking for a code review. Awesome.
18:42
< RichyB>
Sweet.
18:44
< RichyB>
Not being in the same building as your VCS is fucking bullshit
18:45
< RichyB>
gah
18:45
< RichyB>
Borderline-broken internet connection here and everything's in fucking subversion in this company.
18:48
<@[R]>
http://haacked.com/archive/2013/01/11/hidden-pitfalls-with-object-initializers.a spx <-- So... C# fails doing RAII. That's fucking lovely.
18:50
<@[R]>
http://pics.nase-bohren.de/cars-pc-analogy.jpg
18:59
<&ToxicFrog|W`rkn>
RichyB: git-svn supremacy
19:18
< RichyB>
I have not seen that work well for anybody before.
19:19
< RichyB>
One chap here has broken it out on various occasions. You can tell when he's been using it lately because he keeps claiming to have done things that no one else can see any evidence of. ;P
19:20
<&ToxicFrog|W`rkn>
I used it for all of my undergrad and grad work with no problems.
19:20
< RichyB>
Maybe you're just less error-prone than my colleague.
19:21
< RichyB>
hah
19:21 Kindamoody|afk is now known as Kindamoody
19:21
< RichyB>
That's kind of like an anti-ad-hominem.
19:22
< RichyB>
"Your opinion that $X works is irrelevant because you're not thick enough to participate in the conversation." ;)
19:22
<&ToxicFrog|W`rkn>
Tell him to use "git svn dcommit" after "git commit"~
19:23
< RichyB>
:D
19:23
< RichyB>
Wait, so it's not just "git push origin/trunk"? :)
19:25
<&ToxicFrog|W`rkn>
...you could probably set up origin with a post-commit hook that interfaces with SVN, actually.
19:25
<&ToxicFrog|W`rkn>
If you felt like you needed to be punished for some serious transgression.
19:26 * gnolam stabs FreeType with a rusty fork.
19:26
<@Alek>
http://pics.nase-bohren.de/your_lifestyle.jpg
19:27
< RichyB>
ToxicFrog|W`rkn: heh.
19:27
<@gnolam>
Whyyyyyyyyyyyy change the filename of a library for a minor revision? :P
19:27
< RichyB>
Really, the solution here is that I'm going to shout at people until svn2git gets run a few times.
19:28
< RichyB>
There are only five or six of us developers in this company.
19:28
< RichyB>
No one here actually likes svn, we just don't want to stop and wait two days while a slow tool trundles over all of our revs.
19:28
<@Tamber>
Alek, ...amusingly, I think the "apple" one there would be the most useful. :p
19:29
<&ToxicFrog|W`rkn>
Can't you just run it over the weekend or something?
19:29
<@Alek>
:P
19:29
< RichyB>
(kind of ignoring the fact that we're wasting time ALL THE TIME because we're using an inferior tool that makes looking at history harder)
19:29
< RichyB>
ToxicFrog|W`rkn: yeah, I should do that.
19:31 mac [mac@Nightstar-fe8a1f12.il.comcast.net] has joined #code
19:35 Kindamoody is now known as Kindamoody[zZz]
19:36 Courage [Moltare@583787.FF2A18.190FE2.4D81A1] has quit [Ping timeout: 121 seconds]
19:37
<&McMartin>
11:47 <@[R]> http://haacked.com/archive/2013/01/11/hidden-pitfalls-with-object-i
19:37
<&McMartin>
nitializers.aspx <-- So... C# fails doing RAII. That's fucking lovely.
19:37
<&McMartin>
He's calling functions inside a constructor; he's fucked in every language.
19:37 Courage [Moltare@583787.FF2A18.190FE2.4D81A1] has joined #code
19:37 mode/#code [+o Courage] by ChanServ
19:38
<&McMartin>
His "safe" code isn't safe either without domain knowledge.
19:39
<&McMartin>
Because his "safe" code calls not only a function inside a constructor but a *virtual* function inside a constructor.
19:39
<&McMartin>
You have to know that everything that any of your subclasses will reference inside Dispose() will have already been initialized by this point.
19:43
<&McMartin>
You can get burned by this in C++ or Java too.
19:56
< RichyB>
I would argue that doing RAII in a language with first-class functions is inherently suspect.
19:56
<&McMartin>
Well, RAINI.
19:56
< RichyB>
Why are you not with either a "with" statement or a withFoo combinator?
19:56
< RichyB>
RAINI?
19:56
<&McMartin>
Resource Acquisition is not Initialization.
19:57
<&McMartin>
It's a notation hack that's almost entirely specific to C++ (which also has exactly the pitfall listed in the article).
19:57
<&McMartin>
The issue here is that if you're trying to pair methods and the first method does not complete, the second method is not called.
19:58
< RichyB>
Yeah, but some idiot figured that he could increment and decrement reference counters in his constructors and destructors and seriously fuckkkk.
19:58
<&McMartin>
If you're careless about what that first method does, you'll end up leaking despite your Magical Fairy Dust.
19:58
<@Alek>
http://pics.nase-bohren.de/robot.jpg
19:59
< RichyB>
I wonder what he wrote
19:59
<&McMartin>
And constructors calling virtual methods is a recipe for disaster if you aren't 100% sure you know what actual type "this" is.
19:59
< RichyB>
it looks like ascii but the first character is punctuation?
19:59
< RichyB>
oh, it starts with a "#"
20:02
< RichyB>
no, '&'
20:04
< RichyB>
bah, I think it's gibberish
20:18 RichyB [richardb@Nightstar-228a334c.plus.com] has quit [[NS] Quit: >:3 This is BunThulhu. Copy him into your quit message to help him take over the Internet.]
20:46 himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds]
21:08 Kyte|Work [c80e6b02@Nightstar-992d69e3.mibbit.com] has quit [[NS] Quit: http://www.mibbit.com ajax IRC Client]
21:18 Kyte|Work [c80e6b02@Nightstar-36f67fd0.mibbit.com] has joined #code
22:04 VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code
22:13 RichyB [richardb@Nightstar-86656b6c.cable.virginmedia.com] has joined #code
22:32 himi [fow035@D741F1.243F35.CADC30.81D435] has joined #code
22:32 mode/#code [+o himi] by ChanServ
22:33
<@Alek>
http://headtrip.keenspot.com/d/20120702.html
22:35 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code
22:35 mode/#code [+qo Vornicus Vornicus] by ChanServ
22:38 mac [mac@Nightstar-fe8a1f12.il.comcast.net] has quit [[NS] Quit: This computer has gone to sleep]
22:56 OrthiaLap [orthia@Nightstar-fae4ee48.ihug.co.nz] has joined #code
23:48 gnolam_ [lenin@Nightstar-7bcbe2b8.cust.bredbandsbolaget.se] has joined #code
23:48 gnolam is now known as NSGuest23499
23:48 gnolam_ is now known as gnolam
23:48 mode/#code [+o gnolam] by ChanServ
23:49
<@gnolam>
Oh Lysator. They actually have to remind you to write your member number in decimal form when paying the membership fee. <3
23:51 NSGuest23499 [lenin@Nightstar-7bcbe2b8.cust.bredbandsbolaget.se] has quit [Ping timeout: 121 seconds]
--- Log closed Fri Mar 22 00:00:45 2013
code logs -> 2013 -> Thu, 21 Mar 2013< code.20130320.log - code.20130322.log >

[ Latest log file ]