code logs -> 2013 -> Tue, 24 Dec 2013< code.20131223.log - code.20131225.log >
--- Log opened Tue Dec 24 00:00:28 2013
00:35 thalass [thalass@Nightstar-bce70i.eastlink.ca] has quit [Ping timeout: 121 seconds]
00:42 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has joined #code
00:42 mode/#code [+qo Vornicus Vornicus] by ChanServ
00:49 Reiver [quassel@Nightstar-ksqup0.co.uk] has quit [Connection closed]
00:50 Reiver [quassel@Nightstar-ksqup0.co.uk] has joined #code
01:05 Stalker [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds]
01:12 Turaiel[Offline] is now known as Turaiel
01:36 Turaiel is now known as Turaiel[Offline]
01:50
<&McMartin>
Hrm. when writing preconiditions and stuff in doc comments, do you spell out numbers and operators (i.e., "is positive", "is greater than zero") or use notation (">0")?
01:52
<~Vornicus>
I usually use notation.
01:57
<&McMartin>
Blergh, I have too many balls in the air at once here
01:58 * McMartin declares CELERITER MALE IUDICAMUS, charges forwards.
01:58
<&McMartin>
I've made no progress in like three days
02:02
<&McMartin>
Oh hey, awesome. ST. NICHOLAS SCIENCE. http://www.stnicholascenter.org/pages/real-face/
02:02
<&McMartin>
Looks like they're using CGI techniques to simulate musculature.
02:04
< RichyB>
You know I'm pretty certain that twitter's API used to be really easy to use
02:04
< RichyB>
now I have no fucking idea what the shit this shit is
02:18
< RichyB>
Haha, the example curl commands that their own "oath signing tool" generates, don't work.
02:23
< RichyB>
oh for god's sake, nothing even remotely connected to twitter even pretends to work
02:27
<@Tarinaky>
McMartin: Depends on intended audience I guess.
02:27
<@Tarinaky>
And what the people who have gone before you have done.
02:28 Turaiel[Offline] is now known as Turaiel
02:28
<&McMartin>
It's funny; I find myself leaning towards words in C because the "doc comments" are just comments and subject to the whims of the linebreaking text formatter
02:28
<&McMartin>
And it looks cleaner to me that way if it breaks mid-sentence than mid-formula or with a formula callout
02:29 * Tarinaky shrugs,
02:29
<@Tarinaky>
You're not wrong.
02:29
<&McMartin>
(IT's funny because C developers would be assumed to be more technical)
02:30
<@Tarinaky>
I'm not sure recognising == != <= >= < > counts as technical.
02:30
<@Tarinaky>
At least, not of the kind specific to C developers.
02:30
<&McMartin>
Sure.
02:31
<@Tarinaky>
That said. Whenever doing anything moderately complicated I usually end up embedding TeX math expressions in my doc comments.
02:31 * McMartin nods
02:31
<&McMartin>
In this case I'd want something more like box and pointer diagrams, but with luck I can keep it as words.
02:32
<&McMartin>
I'm reluctant to do explanatory documentation in too much depth until I get the design right, so for now it's "here's where the thought process is" and kind of scattershot
02:33
<@Tarinaky>
Heh, include a link to a .svg 'comment' :p[
02:34
<&McMartin>
When I get excessive it starts looking more like literary programming >_>
02:34
<@Tarinaky>
(A diagram is worth a thousand words >.<)
02:34
<&McMartin>
(It is, but if they aren't the *right* thousand words you're wasting your time)
02:35
<@Namegduf>
Hee.
02:35
<@Namegduf>
That's a great reply there.
02:35
<@Namegduf>
Documentation has maintenance costs.
02:36
<@Namegduf>
You don't want to be producing it all over the place with no plan on the idea that more == better.
02:36
<&McMartin>
For the bit I'm the most concerned about here, really, the correct documentation is a bibliographic citation.
02:36
<@Namegduf>
You want some idea what you're doing, who the audience is, and how they are going to actually end up finding the documentation at the time they need it.
02:36
<&McMartin>
With inline comments that are crossrefs.
02:36
<&McMartin>
Yeah
02:37
<&McMartin>
The literary approach is IMO best for the context "someone skeptical is auditing your code and you need to convince them in absentia that your design is sane"
02:37 Turaiel is now known as Turaiel[Offline]
02:37
<@Tarinaky>
Namegduf: That applies to all writing.
02:37
<&McMartin>
That goes in the .c files.
02:37
<&McMartin>
The "what the Hell is this, what do I do with it" goes in .h
02:37
<@Namegduf>
Tarinaky: It does. People however tend to forget it for documentation for some reason.
02:37
<&McMartin>
And then big-picture writing about how it's used goes outside of the source code entirely.
02:37
<@Namegduf>
I tend to archive and deprecate my design documents, rather than maintain them.
02:38
<@Tarinaky>
McMartin: I dunno. If you can fit it into a short sentence...
02:38
<&McMartin>
I should deprecate my current Monocle docs.
02:38
<&McMartin>
Tarinaky: Then the project is too small for that~
02:38
<&McMartin>
But no, really, on reflection, even there
02:38
<&McMartin>
That's what goes in README.md in githubese
02:39
<@Tarinaky>
"This method was originally added with X use-case/problem in mind."
02:39
<@Namegduf>
I have some simple "this raises these events which are hooked by these" diagrams.
02:39
<@Tarinaky>
Particularly if such a statement will assist with explaining why having some strange method with some arcane logic is perfectly sane.
02:40
<&McMartin>
Tarinaky: yeah, that's my second category
02:40
<&McMartin>
The "outside of teh source" is "why does this library exist"
02:40
<&McMartin>
And yeah, that's not a universal mechanism. I think the Python standard is to put that stuff in the module's __init__ docstring.
02:40
<@Tarinaky>
What if there's a smaller problem like profiling says that a particular method is slow and you want to optimise it at the cost of readability.
02:41
<@Tarinaky>
Surely that deserves a comment, with the bit of insane syntax, to explain why you thought this was a problem.
02:41
<&McMartin>
That's the class of "stuff in the implementation file that explains why the implementation is what it is"
02:41
<@Namegduf>
Comments are good because if they're sitting next to code they document they are easier to maintain. Still have some cost.
02:41
<&McMartin>
Which was class 2
02:41
<&McMartin>
Class 1 is "In the .h file: what is the interface, what is the contract, what are pre- and post-conditions, what types does itoperate on, etc"
02:41
<@Tarinaky>
McMartin: Then I think I have to disagree slightly.
02:42
<@Namegduf>
Comments I mostly add wherever I feel they have possible value.
02:42
<&McMartin>
Class 3 is "I need a library to do XYZ, does your thing do that"
02:42
<@Namegduf>
Claiming there's a non-intuitive process to it would be lying.
02:43
<@Namegduf>
I deliberately aim for non-overlap with the code, though. Why rather than what.
02:43
<&McMartin>
^++
02:44
<@Namegduf>
Or higher-level what.
02:44
<@Namegduf>
e.g. "Perform all necessary authentication and validation."
02:45
<@Namegduf>
Not really a realistic example, you would have odd code to have all that in one place, but that kind of concept.
02:47
<@Namegduf>
Overlaps with "why", to a degree.
02:47
<@Namegduf>
Also with sectioning functions a bit, even shortish ones.
02:48
<&McMartin>
Yeah
02:48
<&McMartin>
This function here is poorly sectioned. -_-
02:49
<&McMartin>
Going to have to decide wehther to just comment it as if it were three different functions, or make it three different functions and add a fourth which calls them in sequence.
02:50
<&McMartin>
(sync_object_trees, which comprises ensure_capacity, init_new_traits, process_object_lifecycle)
02:50
<&McMartin>
(And I guess that latter could be process_object_creation and process_object_destruction)
02:50
< RichyB>
where is your slay_the_unbelievers subroutine?
02:51
<&McMartin>
I'm thinking client code will implement that in response to collision events.
02:55
< RichyB>
Heehee.
03:00
<&McMartin>
I already know this will fail to properly handle visible/invisible/default render/custom render
03:01
<&McMartin>
But wihtout a framework to hack I can't make much progress anymore; my pencilwork is obviously just wheelspinning
03:31 Turaiel[Offline] is now known as Turaiel
03:40 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has quit [[NS] Quit: Leaving]
03:44 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has joined #code
03:44 mode/#code [+qo Vornicus Vornicus] by ChanServ
04:08 thalass [thalass@Nightstar-bce70i.eastlink.ca] has joined #code
04:29 celticminstrel [celticminst@Nightstar-gj43l1.dsl.bell.ca] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
04:43 VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [[NS] Quit: Program Shutting down]
05:41 thalass [thalass@Nightstar-bce70i.eastlink.ca] has quit [[NS] Quit: augh]
06:09 Kindamoody[zZz] is now known as Kindamoody
06:15 RichyB [RichyB@Nightstar-c6u.vd5.170.83.IP] has quit [[NS] Quit: Gone.]
06:19 RichyB [RichyB@Nightstar-c6u.vd5.170.83.IP] has joined #code
06:39 Turaiel is now known as Turaiel[Offline]
07:21 AverageJoe [evil1@Nightstar-fb1kt4.ph.cox.net] has joined #code
08:08 ErikMesoy|sleep is now known as ErikMesoy
08:38 Kindamoody is now known as Kindamoody|afk
09:19 Vornotron [Vorn@Nightstar-sn7kve.sd.cox.net] has joined #code
09:21 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has quit [[NS] Quit: Leaving]
09:31 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has joined #code
09:32 mode/#code [+qo Vornicus Vornicus] by ChanServ
09:34 Vornotron [Vorn@Nightstar-sn7kve.sd.cox.net] has quit [[NS] Quit: Leaving]
09:43 Vornotron [Vorn@Nightstar-sn7kve.sd.cox.net] has joined #code
09:44 AverageJoe [evil1@Nightstar-fb1kt4.ph.cox.net] has quit [[NS] Quit: Leaving]
09:52 Vornotron is now known as Vash
09:53 mode/#code [+o Vash] by ChanServ
09:56 Vash [Vorn@Nightstar-sn7kve.sd.cox.net] has left #code ["Leaving"]
10:07
< ErikMesoy>
Weird. XCOM:EU seems to heat up the GPU in this computer to 100 degrees no matter what I do - graphics high or low, Vsync on or off, windowed or fullscreen, toggling Enable Framerate Smoothing. What is it /doing/ with the GPU?
10:08
< ErikMesoy>
Previously I thought this was usually lazy "draw frames as fast as you can" behavior which would be fixed with vsync to only drawing frames at refresh rate, but no. Or maybe there's something I don't understand here.
10:09
<&McMartin>
I'd now start checking to make sure I didn't have a fan jammed
10:10
< ErikMesoy>
When I put my hand on the side I can feel hot air coming out.
10:11
< ErikMesoy>
I'll see if I can do any dust cleaning.
10:11
< ErikMesoy>
This thing isn't built to open up nicely. >_>
10:13
<&McMartin>
Dust wackiness was basically what did that to me when I was playing Alpha Protocol some years back.
10:13
<&McMartin>
Just canned air didn't cover it for me, so I took it to the shop where I bought the components and they had this thing that looked like a shop vac but in reverse
10:23
<&McMartin>
hee hee
10:23
<&McMartin>
switch(which) {
10:31 Stalker [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code
10:50
< Xon>
ErikMesoy, I blame your gfx card
10:51
< ErikMesoy>
Xon: I blame insane troll logic programmers, because that temperature was on the main menu and it dropped 15 degrees when I got into a mission.
10:57
< Xon>
lol
10:58
< Xon>
ErikMesoy, it does mean the cooling for your card isn't working properly
10:58
< Xon>
(was that 100 degrees Celius?)
10:58
< Xon>
Celsius*
10:58
< Xon>
cos I can't think of any card which should run that hot if the cooling was actually working
10:59
< Xon>
even 85C is really god damn hot. well unless it is one of those AMD R9 290 =p
10:59
< Xon>
(then that is by design)
10:59
< ErikMesoy>
Yes, C.
10:59
< Xon>
yikes
10:59
< Xon>
I'm suprised it didn't hard lock on you
10:59
< ErikMesoy>
As measured by SpeedFan.
11:00
< Xon>
could be SpeedFan is reporting it wrong
11:01
< ErikMesoy>
Possibly. Also the card appears to be Intel(R) HD Graphics 4000.
11:01
< Xon>
tried something like GPU-Z or MSI afterburner and see if they report a different value?
11:01
< Xon>
yeah, no way that would be running at 100C and your computer would be still working
11:02
< ErikMesoy>
I'll try those
11:05
< ErikMesoy>
GPU-Z reports that the Intel card is running at 92C (wobbles by about 3 degrees) when doing something hot, also found a second card, NVIDIA GeForce GT 630M, running around 85.
11:06
< ErikMesoy>
I think the card may be reporting its own temperature wrong.
11:06
< Xon>
yikes, both are running hot
11:07
< ErikMesoy>
I think there's a microhell in my computer hosting all the daemons. :P
11:08
< Xon>
laptop I assume?
11:08
< ErikMesoy>
Yes.
11:08
< Xon>
(due to the GeForce GT 630M)
11:09
< Xon>
the 630M is running hot, but the Intel one really sohuld not be that hot
11:09
< Xon>
what is your CPU temperature?
11:10
< Xon>
if that is crazy high, then you definitely have a cooling issue
11:10 * ErikMesoy runs some more experimental tests. WHAT THE HELL, PROGRAMMERS.
11:10
< Xon>
(CPU-Z can measurements)
11:10
< Xon>
CPU-Z can display those measurements)
11:10
< ErikMesoy>
XCOM:Enemy Unknown in a mission, press Esc to bring up menu, minimize game: temp rapidly converges on 95C. (as reported by SpeedFan)
11:10
< ErikMesoy>
XCOM:Enemy Unknown in a mission, minimize game: temp rapidly converges on about 75C.
11:10
< Xon>
heh
11:11
< Xon>
I've seen that in games where the damn menu is the most intensive thing in the game
11:11
< Xon>
but generally maxing the gfx doesn't cause problems
11:11
< Xon>
(besides noise due to the active cooling in my desktop going to MAX)
11:13
< ErikMesoy>
Where on CPU-Z is temperature?
11:14
< ErikMesoy>
on the CPU tab I see Core Speed, Multiplier, Bus Speed, etc, not seeing temp
11:14
< Xon>
derp, I use that to figure out the exact module number of my motherboard not the cpu temp sorry
11:15
< Xon>
speedfan should report it
11:15
< ErikMesoy>
At the stable hot level, SpeedFan is reporting CPU temps of "all four cores 90+".
11:15
< Xon>
that is running on the hot side
11:15
<@TheWatcher>
Yeah, it really is
11:17
< Xon>
ErikMesoy, can you check Windows Power Options -> advanced power settings and see what the processor power manegment system cooling policy is set to?
11:24
< ErikMesoy>
Not seeing a cooling policy.
11:26
< ErikMesoy>
Aha, found it, I think. Power Options -> Power Saver Settings -> Cooling Method
11:26
< ErikMesoy>
Maximum Performance
11:27
< Xon>
i'm out of ideas
11:27
< Xon>
blowing out the dust & any hair build up (while it is off!) is probably a good idea
11:27
< Xon>
(I have cats, long haired cats. the hair; it gets everywhere)
11:28 * ErikMesoy installs HWMonitor for second opinion, confirms cores running in the >90C (>190F) range.
11:28
< ErikMesoy>
The kicker is, nothing actually stops working.
11:28
< ErikMesoy>
It just makes me worried and confused.
11:29
< ErikMesoy>
What is going *on* here?
11:29
< Xon>
is the keyboard hot to the touch?
11:29
< ErikMesoy>
Keyboard itself, no, lower right side of base adjacent to keyboard, yes.
11:31 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has quit [Connection reset by peer]
11:32 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has joined #code
11:32 mode/#code [+qo Vornicus Vornicus] by ChanServ
11:59
< ErikMesoy>
I'll just leave this here: http://i.imgur.com/saf6ugv.png
12:01 * Tamber trips over it.
12:02
<@Tamber>
Impressive. Are you sure it's not an AMD machine in drag? :p
12:17
< ErikMesoy>
...yes?
12:23
<@froztbyte>
lulz
12:36
<@Tamber>
=]
12:36
< ErikMesoy>
Is there some spot I should scrape the paint off to look for an AMD label underneath? :p
12:37
<@Tamber>
Nah, just leave it running long enough and it'll burn the paint off. :D
12:38
< ErikMesoy>
I don't think so, this thing runs comfortably at those 90+ temperatures for hours at a time.
12:39
< ErikMesoy>
Probably heat-resistant paint.
12:39
< ErikMesoy>
Or the logo is on the other side from the processors.
12:50
< Xon>
lol
12:51
< Xon>
hmm, mixed feelings about this comment for some stuff I did for Total Annihilation Demo Recorder by one of the guys who picked up the stuff I released;
12:51
< Xon>
> your code looks like it was made by a Java nerd that found Delphi on HDD
12:53
<~Vornicus>
that's a new one
12:54
< Xon>
the orginal TA Demo Recorder was heavy on the spaghetti code
12:55
<~Vornicus>
Little surprise there.
13:32
<@Azash>
Merry Christmas to #code
13:32
<@Azash>
Hope you've built free software this year or Richard Clausman is not bound by the GNU Present License to interject in your house
13:33
<~Vornicus>
frankly I don't want that guy's freaky beard anywhere near my house
13:34
<@Azash>
What you are referring to as house is actually Chimney/House
13:35
< Shiz>
( )
13:39
< ErikMesoy>
Does unreleased free software, unlikely to ever leave my hard drive, count?
13:39
< ErikMesoy>
Also I agree with Vornicus.
13:43 VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code
14:10 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has quit [[NS] Quit: Leaving]
15:40 celticminstrel [celticminst@Nightstar-gj43l1.dsl.bell.ca] has joined #code
15:40 mode/#code [+o celticminstrel] by ChanServ
17:22 VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [[NS] Quit: Program Shutting down]
17:46 celticminstrel is now known as celmin|away
18:46 AbuDhabi_ [AbuDhabi@Nightstar-sco4t3.adsl.tpnet.pl] has joined #code
18:47 Turaiel[Offline] is now known as Turaiel
18:48 AbuDhabi1 [AbuDhabi@Nightstar-sco4t3.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds]
18:52 iospace is now known as io\parents
19:14 Orthia [orthianz@Nightstar-avg.1ee.224.119.IP] has quit [Ping timeout: 121 seconds]
19:38
<@Tarinaky>
https://isc.sans.edu/diary/Unfriendly+crontab+additions/17282
19:39
<@Tarinaky>
*NIX malware is more 'fun'.
19:40 Turaiel is now known as Turaiel[Offline]
19:41
< Syka>
hee
19:44 Turaiel[Offline] is now known as Turaiel
19:53 AbuDhabi1 [AbuDhabi@Nightstar-1bpasd.adsl.tpnet.pl] has joined #code
19:56 AbuDhabi_ [AbuDhabi@Nightstar-sco4t3.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds]
20:09 Turaiel is now known as Turaiel[Offline]
20:18 Kindamoody|afk is now known as Kindamoody
20:20 Orthia [orthianz@Nightstar-avg.1ee.224.119.IP] has joined #code
20:20 mode/#code [+o Orthia] by ChanServ
20:24 Reiv [orthianz@Nightstar-avg.1ee.224.119.IP] has joined #code
20:24 mode/#code [+o Reiv] by ChanServ
20:26 Orthia [orthianz@Nightstar-avg.1ee.224.119.IP] has quit [Ping timeout: 121 seconds]
21:04 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has joined #code
21:04 mode/#code [+qo Vornicus Vornicus] by ChanServ
21:19 celmin|away [celticminst@Nightstar-gj43l1.dsl.bell.ca] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
21:25 Kindamoody is now known as Kindamoody[zZz]
21:44 celticminstrel [celticminst@Nightstar-gj43l1.dsl.bell.ca] has joined #code
21:44 mode/#code [+o celticminstrel] by ChanServ
21:49 celticminstrel is now known as celmin|busy
22:01
<@Tarinaky>
Okay. Here's a tech problem.
22:01
<@Tarinaky>
With malware like Cryptlocker kicking about, how can I best secure the files I keep on DropBox?
22:02
<@Tarinaky>
Since they're automatically sync'd from multiple systems, with multiple OSs their exposure is going to be really high.
22:02
<@Tarinaky>
But manually copying them into a directory on each machine is going to defeat the entire point of Dropbox and those backups are going to be useless unless I'm really disciplined.
22:03
<@Tarinaky>
So what's my best option for 'versioning' my Dropbox automagically?
22:06
< AbuDhabi1>
Dropbox already does versioning, IIRC.
22:07
<@Tarinaky>
So if a malware machine uploads an encrypted/deleted version of a file to my Dropbox (and those changes propagate) I can revert the lost version?
22:09
< AbuDhabi1>
Probably. https://www.dropbox.com/help/11/en
22:09
<@Tarinaky>
Oh, cool.
22:11
<@Azash>
You can right-click a file and choose compare versions
22:16
<@Tarinaky>
Yeah, I see it on Google now. Well, phew.
22:16
<@Tarinaky>
Now I only need to worry about the /rest/ of my data.
22:18
<&ToxicFrog>
Regular backups.
22:18
<&ToxicFrog>
Actual backups, not just mirroring.
22:19
<@Tarinaky>
By mirroring we mean?
22:20
<&ToxicFrog>
Copying all your stuff somewhere else periodically with e.g. rsync.
22:22
<@Tarinaky>
Err... I am confus. Isn't that backing up?
22:23
<~Vornicus>
TF is distinguishing maintaining a second copy of your stuff from making a /series/ of copies of your stuff.
22:23
<&ToxicFrog>
Yes, that.
22:23
<~Vornicus>
rsync is built for the former.
22:23
<&ToxicFrog>
Proper backups include versioning, because mirroring does not protect you from anything unless it's caught before the next backup cycle.
22:23
<@Tarinaky>
Ah, right.
22:24
<@Tarinaky>
I don't have the resources to maintain a series of backups.
22:24
<@Tarinaky>
I... actually don't have the resources to maintain a single mirror of /everything/ I want saved :/
22:24
<&ToxicFrog>
Proper backup software also includes deduplication.
22:25
<@Azash>
ToxicFrog: You mean like diffs?
22:25
< AbuDhabi1>
Tarinaky: What, Dropbox's 16GB is not enough?
22:26
<&ToxicFrog>
I currently maintain backups of about ~150GB of worth of data, spanning five systems, going back weeks.
22:26
<&ToxicFrog>
This takes up about 60GB of actual space, because there's a lot of duplication of data both across machines and across backup sets.
22:26
<@Tarinaky>
Dropbox only gives me 2.5G, which I've been using for docs, and my offline backup disk is only 500G or so.
22:26
<&ToxicFrog>
Azash: some backup software uses diffs, but this tends to result in really horrifying performance.
22:27
<&ToxicFrog>
Modern backup software like bup uses CAS.
22:27
<@Azash>
Just wondering about the idea of deduplication since I hadn't heard the term
22:27
<@Tamber>
TF: 'CAS'?
22:27
<&ToxicFrog>
Tamber: content-addressed storage.
22:27
<@Tamber>
Aha
22:27
< AbuDhabi1>
Tarinaky: You get +500 Mb per invite.
22:28
< AbuDhabi1>
Up to 16 GB.
22:28
<&ToxicFrog>
Azash: it's a pretty common term. It just means, if you see the same piece of data multiple times, only store it once.
22:28
<@celmin|busy>
I think deduplication means not storing multiple copies of identical files.
22:28
<&ToxicFrog>
ZFS can do it at the filesystem level, if you have shitloads of RAM.
22:28
<@Azash>
ToxicFrog: Yer
22:28
<@Tarinaky>
Who the hell am I going to invite :V
22:28
<&ToxicFrog>
git does it at the file level.
22:28
<@celmin|busy>
So Time Machine's hard-links probably count as that?
22:29
<@Tarinaky>
Anyway, a lot of the umm... 'large file junk' consists of files that have been written exactly once, are large, and binary.
22:29
<&ToxicFrog>
Tarinaky: so exclude those, unless they're critical.
22:29
<&ToxicFrog>
I exclude all of my music, movies, etc from backups.
22:29
<@Tarinaky>
Yeah, if I exclude all that, that leaves me with just the stuff I have on Dropbox atm.
22:29
<&ToxicFrog>
I can always re-create that, after all.
22:30
<&ToxicFrog>
What I'm backing up is system configurations (/etc /var) and user data (/home /root)
22:30
<@Tarinaky>
I'm... not sure I can recreate my music and film collection.
22:30
<@celmin|busy>
I think I exclude my browser caches...
22:30
< AbuDhabi1>
Tarinaky: Yourself. Dropbox's invite rewards can be cheated.
22:31
<&ToxicFrog>
Tarinaky: the backup includes the mlocatedb, which includes a complete index of my collection. Given that, I can recreate most of it with bittorrent and the rest by re-ripping from the original media.
22:31 * Tamber also mumbles something about a TARDIS.
22:31
<@celmin|busy>
Is that the same database used by the locate command?
22:31
<@Tarinaky>
ToxicFrog: Not all of it is still being seeded.
22:32
<@Tarinaky>
And bandwidth is expensive.
22:32
<@Tamber>
Nah, it's okay; y'don't need to keep a copy of everything. After all, there's a copy of it up on Megaupload!
22:32 * Tamber snickers
22:33
<@Tarinaky>
Some of this stuff took a lot of effort to /find/ like Millenium Done by The Merry Thoughts, which isn't even on Waffles. Had to get a friend with a what.cd account to download it for me.
22:34
<&ToxicFrog>
Tarinaky: anyways. For big stuff that doesn't ever really change, you can get away with mirroring, especially if you use something like rsync --ignore-existing
22:34
<@celmin|busy>
I tried to make up for a damaged CD through BitTorrent, but it was surprisingly hard to find them.
22:34
<&ToxicFrog>
(which ensures that if the file is later modified, the modified version won't be mirrored)
22:34 * Tarinaky nods.
22:34
<&ToxicFrog>
OTOH, for actual user data - i.e. mutable stuff - you absolutely want proper versioned backups.
22:34
<@Tarinaky>
I'm just complaining that I need more storage.
22:34
<&ToxicFrog>
500GB is plenty.
22:34
<@celmin|busy>
There was one torrent whose contents was links to the songs.
22:34
<@Tamber>
....ouch.
22:35
<&ToxicFrog>
For the important stuff, anyways
22:35
<@Tarinaky>
Yeah. All I can do is make sure I won't lose any money from the data I lose :/
22:35
<@Tarinaky>
It'll still be sad though.
22:42
<&ToxicFrog>
So, one of the advantages of real backups is that it means it protects you not just from data loss, but also things like "oops, I just accidentally invalidated all of my KSP saves"
22:43 ErikMesoy is now known as ErikMesoy|sleep
22:46 celmin|busy is now known as celticminstrel
22:49 Derakon[AFK] [Derakon@Nightstar-4k2ccr.ca.comcast.net] has quit [Ping timeout: 121 seconds]
22:50 Derakon[AFK] [Derakon@Nightstar-4k2ccr.ca.comcast.net] has joined #code
23:09
<@Tarinaky>
http://joelgrus.com/2013/12/24/why-programming-language-x-is-unambiguously-bette r-than-programming-language-y/
23:10
<@Azash>
To quote a SEng lecturer that once interrupted an argument about languages: "don't ask me to pick between beer and meatballs"
23:10
<@Tarinaky>
Beer.
23:11
<@Tarinaky>
Every time.
23:11
<&McMartin>
"also integrates well with [text-editor that's like 50 years old and whose key-bindings are based on Klingon"
23:11
<&McMartin>
... I think that's all of them, including Notepad++
23:11
<@Tarinaky>
Notepad++ is only like... 15 years old surely?
23:12 * AbuDhabi1 has Notepad2, Notepad++ and EditPad Lite all installed at the same time and plain old Notepad gets used sometimes too.
23:13 * Tarinaky uses vim in Cygwin on Windows like a bad motherfucka
23:13
<&McMartin>
Speaking of awesome IDEs that are still in pre-alpha, there's a big Light Table update coming next month
23:13
<@Tarinaky>
(Ka'plach!)
23:18 himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Operation timed out]
23:20
<@Azash>
Anyone tried IDEA or how it was called?
23:21
< AbuDhabi1>
I haven't tried IDEA but I've tried CONCEPT.
23:21
<&McMartin>
I've heard of IDEA but I know little of it. Was it one of the commercial ones?
23:22
<&ToxicFrog>
It has free and commercial versions
23:22
<&ToxicFrog>
I found that IDEA was very good, but also very heavy, in terms of both "amount of overhead needed to set stuff up" and "amount of system load"
23:23
<&ToxicFrog>
I.e. it is a project-oriented IDE, not an editor, and it needs 2GB of RAM minimum to run acceptably.
23:23
<&ToxicFrog>
These days I just use Sublime Text 2.
23:23 * AbuDhabi1 uses Code::Blocks.
--- Log closed Wed Dec 25 00:00:43 2013
code logs -> 2013 -> Tue, 24 Dec 2013< code.20131223.log - code.20131225.log >

[ Latest log file ]