code logs -> 2013 -> Sun, 15 Dec 2013< code.20131214.log - code.20131216.log >
--- Log opened Sun Dec 15 00:00:04 2013
00:00
<@froztbyte>
because the software *on* the machine isn't upgradeable
00:00
<@froztbyte>
so it's 1:1 the same in antiquity, but without all the fucking stupid retardation and pain that you find in rhel country
00:00
<@froztbyte>
I mean, seriously, /etc/sysconfig?
00:00
<@froztbyte>
fuck you guys.
00:00
<@TheWatcher>
HEh
00:00
<@TheWatcher>
^--
00:01
<@froztbyte>
I have this theory
00:01
<@froztbyte>
LCD
00:01
<@froztbyte>
Lowest Commoner Denominator
00:01
<@froztbyte>
for something to be accepted widely, it has to be usable by idiotic plebs
00:02
<@froztbyte>
(but those, of course, are basically just applying wrenches)
00:02
< RichyB>
well, yes
00:03
<&McMartin>
If it can't be widely used, then it won't be widely used.
00:03
<&McMartin>
This is a tautology with added sneering at The Masses.
00:03
<@froztbyte>
I have an alternate idea
00:03
< RichyB>
If you design something that needs a lot of study to use at all, you've put up a high barrier to entry to using your thing.
00:03
<@froztbyte>
which basically throws away this retarded assumption that everyone is equal and should have all their feelings treated equally
00:04
< RichyB>
The product that usually wins in the market is whichever one can be most easily used.
00:04
<&McMartin>
Or most cheaply used
00:05
<@froztbyte>
RichyB: but that idea immediately breaks down
00:05
< RichyB>
If you want to improve the state of the world a lot, make the product which can be most easily used and which also can most easily be used correctly.
00:05
<@froztbyte>
debian is vastly easier to deal with because it's not full of insanity
00:05
<@froztbyte>
I believe the only reason rhel is "easy" is because there's enough trained monkeys
00:05
< RichyB>
[Citation needed]
00:05
<&McMartin>
froztbyte: That statement is false on its face~
00:05
<@froztbyte>
pop quiz
00:06
<@gnolam>
RichyB: you're still committing the economist fallacy of assuming perfect knowledge.
00:06
<&McMartin>
"Correct = the way I do it"; "Sane = ways to do it compatible with the way I do it; sometimes strictly synonymous with 'correct'"
00:06
<@froztbyte>
where do your syslog entries go?
00:06
<@froztbyte>
McMartin: no, I'm not making that assumption
00:06
<&McMartin>
... with that question? You most certainly are.
00:06
<@froztbyte>
McMartin: I've run an environment where gentoo actually made sense.
00:06
< RichyB>
gnolam, which sentence? "The product that usually wins in the market is whichever one can be most easily used" can't possibly be a fallacy, I think, because it is an observation rather than a prediction.
00:07
<@froztbyte>
I have run every single major distro
00:07
<@froztbyte>
debian comes out ahead
00:07
<&McMartin>
That's nice
00:07
<@froztbyte>
(I've run a couple of minor ones too)
00:07
<&McMartin>
The answer to your question is "I've never had to give a shit. I win."
00:07
<&McMartin>
I'm an end user.
00:07
<@froztbyte>
rofl
00:08
<&McMartin>
If I *did* regularly have to give a shit, I would be running something that took care of itself better instead.
00:08
< RichyB>
gnolam, the point of the other bit is that, since the most easily-used product will win, try to make the most easily-used product, and if your product's design can be made to naturally direct people towards correct instead of incorrect usage too (*) then you will probably have improved the world somewhat.
00:08
<@froztbyte>
(*cough* debian *cough*)
00:08
<@froztbyte>
but yeah
00:08
<&McMartin>
If I have to *occasionally* give a shit, I will be websearching for instructions first
00:08
<@froztbyte>
too many people contributing bad software to the world
00:08
<&McMartin>
froztbyte: Ubuntu and Fedora both also meet this bar handily.
00:08 Harlow [Harlow@Nightstar-b902fba7.chi.megapath.net] has joined #code
00:08
<@froztbyte>
may they all choke on some cocks
00:09
< RichyB>
gnolam, unless your product is, like, moon-based death lasers, in which case winning the market is probably massively harmful to everyone except your customers. ;)
00:09
<&McMartin>
As evidenced by the fact that I'm running them as consumer-grade OSes without incident.
00:09
<@froztbyte>
McMartin: I don't believe you on fedora, but I'll give you the benefit of the doubt
00:09
<@gnolam>
RichyB: As to where that reasoning leads, I point to GNOME. :P
00:09
< RichyB>
froztbyte, I run Fedora.
00:09
<&McMartin>
froztbyte: Fedora is what made me abandon gentoo, as it happens
00:09
< RichyB>
gnolam, I don't think that GNOME has succeeded in either direction.
00:10
<@froztbyte>
McMartin: the reason I don't believe you is because I had a couple of inter-release fedora upgrades (from fedora core -> fedora days, and along) which just fucked out hilariously badly
00:10
<@Tamber>
It is nearly the end of 2013, why is distro-warring still a thing?
00:10
< RichyB>
gnolam, BTW, I continue to rave about KDE4; it is amazeballs.
00:10
<@Tamber>
(No, don't answer that. It's rhetorical.)
00:10
<&McMartin>
froztbyte: I started in FC3.
00:10
<@TheWatcher>
Tamber: EMACS!
00:10
<&McMartin>
I did not take every update.
00:10
<@froztbyte>
McMartin: FC2 here
00:10
<&McMartin>
FC2 and FC4, in particular, were ones I did not take because they were awful~
00:10
<@froztbyte>
McMartin: iirc, FC2->4->6->7->8->9->12
00:10
<&McMartin>
Also... either F10 or F11
00:10
<@Tamber>
TW: See, the problem there is I *use* Emacs. But, *sigh*, as a matter of tradition: ED, YOU HEATHEN!
00:11
<@TheWatcher>
Hee
00:11
< RichyB>
Tamber, I use ed from inside emacs.
00:11
<@froztbyte>
McMartin: and after every stage my system was hosed for a handful of days :/
00:11
<@gnolam>
Anyway. My point is that /actual/ performance (by any metric) has very little to do with market share.
00:11
<&McMartin>
froztbyte: Right, so, um, just making sure here: You do realize that it's F20 that's hitting in three days
00:11
<@froztbyte>
McMartin: as I said, benefit of the doubt :)
00:11
<@Tamber>
RichyB: Worst of both worlds, eh?
00:11
<@Tamber>
:)
00:11
<@froztbyte>
McMartin: yeah, I know
00:12
<&McMartin>
Kinda surprised 6->7->8 were a problem; 5 through 9 was one of the smooth patches for me.
00:12
<@Namegduf>
I notice that a lot of distributions have improved over the last five years or so.
00:12
<&McMartin>
whatever was from 9 was a disaster, though, and then I popped back in around fedora 13
00:12
<@froztbyte>
quite so
00:12
<@froztbyte>
a lot of rounding work has happened
00:12
<&McMartin>
And then that system (which is from like 2003) has taken every update since.
00:12
<@Namegduf>
I kind of wonder if everyone condemning the previous distributions they used as crap is just being tricked into attributing to their new distribution what is mostly improvements that happened for all over time.
00:13
<@froztbyte>
and some people seemed to actually agree about some things
00:13
<&McMartin>
Just, um, usually not immediately, because I never do~
00:13
< RichyB>
Tamber, I am not joking about that. Sometimes I'm working inside an ssh session inside shell-mode, and since shell-mode deliberately implements a DUMB terminal instead of a vt100 emulator, it's occasionally easier to just run "ed" on the remote machine than to bother to get a real terminal in there or to get tramp to edit it.
00:13
<@Tamber>
*noddle*
00:13
<@Namegduf>
Exceptions exist, of course.
00:13
< RichyB>
Also I learned vim first, then switched to Emacs.
00:13
<@froztbyte>
Namegduf: everything does improve all the time
00:13
<&McMartin>
... except GNOME
00:14
<&McMartin>
Possibly also Unity
00:14
<@froztbyte>
don't let them hear you say that
00:14
< RichyB>
So I know and habitually use all three of the traditional Unix editors (and slightly prefer vi to vim). =D
00:14
<@froztbyte>
their feature removals are IN YOUR INTEREST
00:14
<&McMartin>
It's true!
00:14
<@froztbyte>
they know you don't know what to do with that one feature you use *all the goddamned time*
00:14
<&McMartin>
So I went one step ahead and removed *all* the features, and installed LXDE instead.
00:14
<@froztbyte>
winning
00:14
< RichyB>
McMartin, I apologise for PREACHING AT YOU like a recent convert
00:14
< RichyB>
McMartin, but have you tried KDE4? It's bloody amazeballs. =D
00:15
<&McMartin>
Man
00:15
<&McMartin>
I don't actually remember
00:15
<&McMartin>
I tried either KDE3 or KD4 as part of, IIRC, some SuSE install
00:15
<@froztbyte>
KDE3, likely
00:15
<@gnolam>
RichyB: I have tried KDE4. And that was after KDE3, so I went "What the hell is this shit?"
00:15
<@Tamber>
I tried KDE4 back when it was first out, and thought it was... a bit spartan.
00:15
<@gnolam>
I actually _liked_ KDE3.
00:15
<@gnolam>
KDE4? Jesus H Christ it was awful.
00:15
<@Tamber>
But then, I never touched it after that.
00:15
<@froztbyte>
I don't believe KDE4 and SuSE had overlapping ttimelines
00:15
<@froztbyte>
timelines*
00:16
<&McMartin>
And I literally could not interact with it, because something about the desktop metaphor had been changed in a manner I did not understand and could not discover interactively.
00:16
<@froztbyte>
gnolam: kde3 was pretty good in a lot of ways
00:16
<@Tamber>
(I pretty much just tumbled into e16, and that's where I think I'm going to stay. *shrug*)
00:16
< RichyB>
KDE3 was nice. KDE4 sucked for the first few releases after the 4.0.
00:16
<@froztbyte>
gnolam: I think that in recent months they got a lot of the kde3 niceties back along with the KDE4 goodies
00:16
<@froztbyte>
KDE4 up to 4.3 was basically super rough
00:16
<@gnolam>
Besides the otherwise for Linux standard fare of "you have to hand-edit arcane config files just to get a sane goddamned date format", it failed _every single usability thing I could think of_.
00:16
<@froztbyte>
iirc they branded 4.0 as a tech demo
00:16
<&McMartin>
I expect that I'm going to shift desktop environments once I upgrade the laptop to Ubuntu 14.04
00:17
<&McMartin>
Because Unity only had one remotely usable release, in my experience, and that release was in 12.04
00:17
<~Vornicus>
reminds me at some point I hsould choose a Linus
00:17
<~Vornicus>
x
00:17
<@Namegduf>
I'm a tiling WM sort.
00:17
< RichyB>
My experience with KDE4 is that, now, it has caught up with and exceeded the niceness that I remember enjoying with KDE3, and it's really very good once you turn off all the animations (since they introduce many small deliberate pauses to let the animation run).
00:17
<@froztbyte>
I found that I like tiling WMs on my (other) laptop, with the exception of hating the nipple mouse
00:17
<@Namegduf>
Yes, it's hand configured in Haskell, but it's documented and intended to be done that way rather than a thing you go into for debugging, and configuring as a whole is easier than debugging.
00:18
<@gnolam>
RichyB: with extra INTEGRATED MICROBLOGGING? :P
00:18
<@Namegduf>
And yes, it's actually totally unusuable if you don't know how I set up the keybindings.
00:18
<@gnolam>
Seriously. That was what KDE4 touted as its greatest feature.
00:18
<@Namegduf>
But it's efficient and the config is fast.
00:18
< RichyB>
gnolam, I haven't found the integrated microblogging feature but the integrated desktop Mandelbröt is pretty good.
00:18
<@froztbyte>
gnolam: hey, google's playing the social train~!
00:18
<&McMartin>
Namegduf: I want an almost-tiling WM, it turns out.
00:18
<@Namegduf>
Win+C for a new Chrome window on the current workspace, Win+P for a new terminal.
00:18
<@Tamber>
I have a shortcut button for a new terminal window.
00:19
<@gnolam>
While having things like absolutely no contrast whatsoever, and a package manager where you had to use a fucking magnifying glass to tell the difference between installed and not installed.
00:19
<@Namegduf>
Different monitors independently switch workspace and if I open multiple programs they divide up the space automatically.
00:19
<@froztbyte>
RichyB: it was 2009 or so. they played the "your life in your desktop" card
00:19
<@Tamber>
‚¶I never remember it exists until *after* I've opened the terminal with the menu.
00:19
<@froztbyte>
RichyB: realtime photo streaming of flickr, etc etc
00:19
<&McMartin>
I want my browsers, text editors, and terminals to be tiled, but I want applications that I start from the terminal that make their own windows to hover over that window like a proper desktop window.
00:19
< RichyB>
froztbyte, huh. Cool but silly.
00:19
<@Namegduf>
I don't minimise, I just switch workspaces and move things between them on occasion.
00:19
<@froztbyte>
RichyB: these kind of things are why plasma has the capabilities it does
00:19
<@Namegduf>
McMartin: You can exempt applications as the lazy way to deal with them tiling poorly.
00:19
<&McMartin>
Namegduf: Yeah, I suspect I want the reverse
00:20
<@Namegduf>
You can probably do that, too.
00:20
<&McMartin>
That is, I'd want tiling to be opt-in
00:20
<@gnolam>
4.0 and 4.1 made me turn my back permanently on KDE.
00:20
<@Namegduf>
XMonad configuration files are basically you composing functionality into the executable.
00:20
< RichyB>
froztbyte, yeah, Plasma's capabilities are quite nice to have even though I assume that they were massively over-touted or over-used at some point.
00:20
<@Namegduf>
They're compiled.
00:20
<@Namegduf>
You can do most anything you can express and turning "default on" to "default off" sounds easy enough.
00:20
< RichyB>
Namegduf, s/basically //
00:20
< RichyB>
Namegduf, your description was precisely accurate, no need to add "basically" to the sentence. :)
00:21
<@Namegduf>
You can write conditionals with the power of a full functional programming language. If you know Haskell. I don't but I can pretend sometimes.
00:21
<@Namegduf>
I got it to automatically send the first spawned instances of some programs to different workspaces once.
00:21
<@Namegduf>
So I could autostart things on the right workspace.
00:21
<&McMartin>
Heh
00:21
<&McMartin>
Yeah, I know a fair bit of Haskell though not enough to call myself proficient
00:22
< RichyB>
ehhhhh
00:22
<@Namegduf>
McMartin: The fancy but slower way is to write custom tiling rules for the software which work well. People tend to do that for IM stuff.
00:22
<&McMartin>
By this I mean "I solved several hundred project Euler problems in Haskell."
00:22
<@Namegduf>
Tabbed or something.
00:22
<@Namegduf>
There's lots of layouts you can pull in and embed in each other and whatever you like.
00:22
<&McMartin>
Yeah, in my case what I'm usually doing when developing is writing small games which run in a frame buffer or accelerated GL window
00:23
< RichyB>
One of the things that I find about Haskell is that I sometimes see people who don't think they know Haskell very well, producing Haskell code which I can read at the same rate as I can scan my eyes across the physical text.
00:23
<@Namegduf>
Ah, I see.
00:23
<&McMartin>
And the 'true' usecase is fullscreen, which tiles REALLY WELL ;-)
00:23
<&McMartin>
But I'd also like to have that debug log scrolling in the bacground while I develop
00:23
< RichyB>
Which IMHO is a very strong compliment to be able to honestly pay to a piece of code, for its clarity.
00:23
<@Namegduf>
Well, it recognises windows by their WM_CLASS.
00:23
<@Namegduf>
It's likely they have fairly consistent values for that.
00:23
<&McMartin>
RichyB: IIRC, I was one of those people~
00:24
< RichyB>
Namegduf, IIRC while WM_CLASS is the recogniser you'll most often use, it's possible to match by other attributes too.
00:24
<@Namegduf>
Yeah, true, you can.
00:24
<@Namegduf>
My netbook splits the screen in two and whichever has focus expands over the top of the other.
00:24
<&McMartin>
However, that puts me at "competent with caveats" in my self-evaluation, not "proficient". I'm mostly innocent of typeclasses and that unduly restricts the code that I can read and understand *myself*, even if the code I write is clean to other readers until I try something better done with judicious use of typeclasses.
00:24
<@Namegduf>
For effective multitasking on a tiny little screen.
00:24
<&McMartin>
Yeah
00:24
<@Namegduf>
It's neat.
00:25
< RichyB>
McMartin, :) Also a couple of people I've met on freenode.
00:25
< RichyB>
It's been a while since I last used xmonad but there's support for floating windows. Possibly in a -contrib plugin somewhere.
00:25 * McMartin nods
00:26
<&McMartin>
In practice, since all I really need is "divide the screen in two, float stuff on top of that", basically every window manager for Linux now has me covered, and OS X and Windows do too, at least as of Win7
00:28
<&McMartin>
(For me to consider myself competent at Haskell *without* caveats, I'd need to be able to understand the internals of custom monads as well as be able to just use them; write a custom monad of my own and know to never do this; define and use my own typeclasses for something worthwhile)
00:28
<@Namegduf>
RichyB: Built in
00:28
<@Namegduf>
I think it does it on its own for some windows which absolutely cannot be resized, possibly
00:28
<&McMartin>
(So I'd probably be just fine configuring XMonad)
00:29
<&McMartin>
Namegduf: ... heh. That probably includes most of my GLX contexts~
00:29
<@Namegduf>
McMartin: I'm not very sure about it
00:31
< RichyB>
Namegduf, oh yeah, you're right.
00:32
< RichyB>
McMartin, I strongly disagree with you on that; you do *not* need to understand the implementations of HS abstractions in order to be able to understand them!
00:33
<&McMartin>
RichyB: Oh! I didn't intend to imply that; quite the opposite, actually.
00:33
< RichyB>
IMHO you just need to understand precisely the algebra that they provide you and what time&space properties to expect.
00:33
<&McMartin>
Right
00:33
<&McMartin>
That's the level I'm at.
00:34
<&McMartin>
I'm just not comfortable saying "I'm competent with Haskell" without qualifying that that's the level I'm at with those.
00:34
<&McMartin>
Basically, without that, I don't think I'd be able to join a Haskell project and start contributing meaningfully without learning more of the language first, which is my threshold for "I can get away with putting this language on my CV"
00:36 redwire [redwire@Nightstar-27dppb.nl.bellaliant.net] has quit [Connection reset by peer]
00:36
< RichyB>
Okay. Well, if you want to get to "I can write a useful custom Monad implementation", I believe that the quickest path there is: 1. write a few Monoid instances (such as "instance (Monoid a, Monoid b) => Monoid (a,b)") to get used to typeclasses in practice, and 2. read a blog post about Free monads because they're the easy way to write monadic interpreters.
00:36
<&McMartin>
Yeah, that sounds about right
00:36
<&McMartin>
I'm living at the other end of the abstraction chain right now, though :D
00:37
< RichyB>
My current problem with Haskell in practice is that I learned how to use the C FFI at one point and became incapable of writing *Haskell* programs in Haskell instead of writing C programs in Haskell. ;)
00:38
< RichyB>
Realistically, more that I'm not using it in my day job and I'm too lazy to do much programming outside of that. :)
00:39
<&McMartin>
Heh.
00:40
<&McMartin>
And yeah.
00:40
<&McMartin>
To me "proficient" means "I could use it in my day job even though I don't"
00:41
< RichyB>
I can only unashamedly claim that of Python and Bash at the moment.
00:42
< RichyB>
Maybe C, if I was willing to accept really low productivity for a while.
00:42
<&McMartin>
That latter is what I think of as "competent"
00:42
< RichyB>
s/bash/Bourne shell/ (I avoid bash-isms on purpose)
00:42
<&McMartin>
If you're competent, you'll become proficient pretty quickly once you have to
00:43
<&McMartin>
Competent is "if we hire him, we won't have to train him except on the things unique to our product"
00:43
< RichyB>
s/him/them/
00:43
<&McMartin>
Generic 'he', sorry~
00:44
< RichyB>
No worries.
00:44
<&McMartin>
Proficient is "we can let you refactor parts of the code by the time you get it building"
00:45
<&McMartin>
My part-of-the-day-job stuff is C++/Java/Python these days, and I think I maintain a level of professional quality with C.
00:45
<&McMartin>
Perl and a handful of Lisp dialects end up one tier down
00:45
< RichyB>
Are you also working at Google?
00:45 You're now known as TheWatcher[T-2]
00:46
< RichyB>
Can't think of many other places where C++, Java and Python are the specific day-to-day mix.
00:46
<&McMartin>
No; I'm at a medium-sized virtual desktop management startup
00:46
<&McMartin>
Which operates at *all the layers of abstraction*
00:46
<&McMartin>
C++ for client-side applications, Java for server code, Python for internal keep-stuff-humming-along.
00:47 Harlow [Harlow@Nightstar-b902fba7.chi.megapath.net] has quit [[NS] Quit: This computer has gone to sleep]
00:47 Thalasleep is now known as Thalass
00:48
< RichyB>
I'm slightly surprised that your client-side apps aren't Python too, since it does an okay job on low-performance-demand Win32 things.
00:48
<&McMartin>
Some of the components of our product are things like "in-guest custom filesystems"
00:48 * RichyB blink
00:49
<&McMartin>
Our product is basically reverse-VDI; you use dumb servers to push out images onto proper computers and run them in VMs on the client, instead of streaming the desktop. This works for a different set of use cases than normal VDI does.
00:49
< RichyB>
From your description, I'd guess Citrix right now, only they're very well established and not a startup.
00:50
< RichyB>
*nodnod* good idea.
00:50
<&McMartin>
The company Moka5, inc., but what is sold and what the cool parts of the tech are not the same thing. ;-)
00:50
<&McMartin>
Becuase if the cool parts work they are *invisible*
00:50
< RichyB>
Yes.
00:51
<&McMartin>
The admin as a certain level of control over how the VM is run, even on the user's system, enforced via various policies and some tricks that occasionally make us run afoul of antimalware scanners
00:51
<&McMartin>
But then another part of this is that the admin can push updates "underneath" the work you've done yourself, as long as they're in different places, and they will be.
00:51
< RichyB>
That sounds like a great idea for corporate desktops. It'll be as good as if you were remote-reimaging every client every reboot.
00:51
<&McMartin>
It's exactly that, with a few caveats
00:51
<&McMartin>
And to get *around* those caveats we basically have a beefed-up thing like unionFS that works (a) on Windows and (b) on the Registry too.
00:51 Stalker [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds]
00:52 You're now known as TheWatcher[zZzZ]
00:52
< RichyB>
Huh.
00:52
< RichyB>
TBH that's just "oh neat tech" and "surprised that that works"
00:52
<&McMartin>
(And has configurable, multiple-writable layers, because it turns out one writable layer isn't enough)
00:52
< RichyB>
The thing that I actually like about what you're describing is the core it's-a-reverse-VDI thing
00:52
< RichyB>
because IMHO VDI is a bit insane on multiple levels
00:53
<&McMartin>
Yeah, it turns out to not only work but be mandatory, because if you remote-image a machine of every boot, you can't mass-duplicate it because they'll all have the same machine name/domain information/etc and you do *not* want to be running sysprep every boot. ;-)
00:53
< RichyB>
Let's take the user's interactive desktop and stick it on a server that's maybe 50ms away from them over an internet connection; what?
00:53
<&McMartin>
"And now let's give it to our jet setter employees"
00:53
<&McMartin>
There are places where we fail horribly and VDI is honestly the best solution - the one that comes to mind is office workers that work with sensitive data that isn't supposed to leave a specific location.
00:54
< RichyB>
Let's take the user's desktop that was formerly sitting on their desk in their office, eating ~50W of electricity and effectively being part of the building heating system
00:54
<&McMartin>
Then you want a thin client as incapable of possible of screen-scraping, and even if someone steals those clients, no harm done.
00:54
< RichyB>
and move it onto a server in a crowded machine room where it will need to be specifically air-conditioned at great expense; what?
00:54
<&McMartin>
(So, like, banks using VDI so that the disks with the sensitive data are literally in a vault - this is fine)
00:55
< RichyB>
Well yes that makes sense.
00:55
<&McMartin>
But yeah, we have a pretty good product, and it turns out the problems of getting this stuff deployed efficiently and working smoothly are both plentiful and interesting
00:56
< RichyB>
I think I would not have gone down the implement-win32-unionfs route. I would have gone re-prepare-on-every-reboot.
00:57
<&McMartin>
Both approaches have obvious fatal flaws; my understanding is that we moved to the win32-unionfs approach as part of 2.0
00:57
< RichyB>
Something like: take the base image, mount the NTFS using Linux' NTFS support, make changes, then schlep it over to the client.
00:57
< RichyB>
Obviously there are problems with that too.
00:57
<&McMartin>
Yeah
00:58
< RichyB>
So uh my first thought boils down to "do something that has the same outcome as sysprep but is faster, do it on every boot".
00:58
<&McMartin>
(For us, that problem is 'To scale better, that's the kind of thing that should happen client side, and it is a customer requirement that clients not necessarily have those kinds of privileges')
00:58
<&McMartin>
yeah
00:59
<&McMartin>
We're about 8 years old now; we are starting to have issues of Legacy, tbh.
00:59
<&McMartin>
Burning some of that Legacy out is part of my current job~
00:59
<&McMartin>
An impressive amount turns out to be salvagable, though
00:59
< RichyB>
Legacy code starts to be a thing on literally day 2 of your project. ;)
01:00
<&McMartin>
Anyway, yeah
01:00
<&McMartin>
We're also established enough that large customers I'm not allowed to name have managed to put it into some incredibly horrifying situations, where it doesn't do as well as they'd like
01:00
< RichyB>
Maybe there should be a culture of airing "oh god legacy" laundry more often
01:00
< RichyB>
every time I hear about someone else's, it makes me feel less bad about mine.
01:00
<&McMartin>
Heh.
01:01
< RichyB>
On the assumption that this might make you feel better: the company division I just left has its core product line on Python2.4
01:01
<&McMartin>
Honestly, the legacy here is more "we were wrong about the intended initial use case, and while we could and did adapt to what turned out to be the *actual* use cases, this resulted in the abstraction layer cuts happening in the wrong place"
01:02
<&McMartin>
So, in the worst case, that's "sure, let's go refactor several hundred KLOC"
01:02
< RichyB>
the division I just moved into has as its central-point-of-contact for 90% of its users, a similarly python2.4-bound program with executable code in its database
01:02
<&McMartin>
Which is why you send in someone who prefers the chisel to the sledgehammer, viz., me, so that he can say "we need to rewrite these two classes, and then we can delete every library that is no longer used" and you get something that looks like a roadmap with deliverables.
01:03
< RichyB>
"prefers the chisel to the sledgehammer" is a good turn of phrase, I'm stealing that one.
01:04
< RichyB>
McMartin, btw, *does* hearing that I sometimes maintain python2.4 code make you feel less bad by comparison? ;)
01:04
<&McMartin>
That's a different set of Legacy problems ;-)
01:04
<&McMartin>
We have a complete subteam working on our equivalent of those.
01:05
<&McMartin>
And we're in a unique position where external parties are a forcing function upon us.
01:05
< RichyB>
No? That's called "finding your product-market fit", I think?
01:06
<&McMartin>
A lot of our end-users are on Macs, and the way this generally works is that our management software ends up sandboxing their personal copy of VMware Fusion
01:06
<&McMartin>
So any time OS X or Fusion upgrades, every C++ library ever written breaks.
01:06
<&McMartin>
(I exaggerate slightly, but only a bit)
01:06 Thalass [thalass@Nightstar-gucq71.bigpond.net.au] has quit [Operation timed out]
01:07
<&McMartin>
More directly: "We have a Bring-Your-Own-Computer use case, which puts limits on how self-contained we can really be, and with also demands as much self-containment as possible"
01:07
<&McMartin>
I've earned a lot more respect for Microsoft's psychopathically obsessive attitude towards backcompat in this process >_>
01:08
<&McMartin>
Or rather, it's earned a lot more respect from me.
01:09
<&McMartin>
And because it turns out paying customers usually don't accept "this is an upstream bug" except in unusually extreme cases, this means we do a lot of system-level horrible hackery too, which needs to be kept up-to-date and then removed once it stops being a problem.
01:09
< RichyB>
So people are bringing their own hardware in and then your software is being used to give them centrally-managed desktop images inside a copy of VMW Fusion (if they have a Mac) or VMW Player (otherwise)?
01:09
<&McMartin>
Yeah, that's the default BYOC use case.
01:10
<&McMartin>
We also have a slightly misnamed "Bare Metal" product that runs on a stripped-down Linux system and does nothing but set up your VM and kick you into it.
01:10
<&McMartin>
Apparently some folks like that for mass-deployment and management with corporate assets
01:10
<&McMartin>
And I'm willing to accept that pushing an image update is probably faster than a lot of the other remote administration tools
01:11
< RichyB>
I don't think that's actually misnamed. e.g. XenServer and OpenStack's compute thing are also tiny Linux images that run just a hypervisor.
01:11
<&McMartin>
Fair enough.
01:12
<&ToxicFrog>
McMartin: on my more fanatical days, I contend that the backcompat spiders are a symptom and the cause is not requiring all software to be open-source and enforcing that with nuclear failsafes.
01:12
< RichyB>
that is very fanatical, since backcompat matterns between open-source projects too
01:13
<&McMartin>
ToxicFrog: A defensible position, but not ironclad; I would offer as my first argument that it's not sufficient the entirety of gentoo.
01:13
< RichyB>
not *within* them, which is why Linux can refactor drivers in the tree to their hearts' content, but *between* different open-source projects.
01:14
<&McMartin>
I'm not taking a firm position yet, but I'm leaning towards a position where open standards are more important than open source, in terms of having a healthy ecosystem
01:14
<&McMartin>
(This is cheating slightly because a requirement for open standards is an open source reference implementation)
01:14
< RichyB>
I go the other w-oh, reference implementations.
01:15
< RichyB>
I think that about half the time, you need an open standard and half the time, you're best with a totally nonstandard thing that's open-source, but I haven't thought about why hard enough to explain it.
01:16
<&McMartin>
I share that kind of attitude; note my "I'm not taking a firm position yet"
01:16
<&McMartin>
Certainly there will always be a place for bespoke software.
01:17
< RichyB>
There will probably always be a place for proprietary software provided you assume capitalism.
01:17
<&ToxicFrog>
McMartin: my main objection to that is that if someone doesn't follow the standards, or the standards change, and their software is closed source, you have to re-implement the entire thing from scratch.
01:17
<@Tamber>
Richy: In the latter case, because the standard was written in a vacuum by a committee of people who haven't tried to implement anything of the sort 'on the front lines', as it were, so it's mad as a box of frogs? :p
01:18
<&ToxicFrog>
Unless the original developers are (a) still around and (b) willing to update their legacy software
01:18
< RichyB>
Tamber, that'll do it.
01:19
<&ToxicFrog>
On a personal level this is mostly aggravating when I think about all the games that we have forever lost the source code for, but it is more general than that.
01:19
< RichyB>
e.g. I don't care if the protocol for loading a freshly-compiled binary onto a microcontroller board is some one-off thing that only this one program uses, but it would be great if this program was open source
01:20
< RichyB>
on the other hand, you can't reverse-engineer a multi-vendor standard from a single implementation without it sucking really hard because at the very least you'll accidentally standardise on a bunch of bugs.
01:21
< RichyB>
Imagine if, instead of an RFC for IRC, everyone just reverse engineered what one popular IRC client and one popular IRC server did, it'd be madness.
01:21
< RichyB>
(See: mIRC text colour codes, which are precisely that madness.)
01:28
< AnnoDomini>
What's wrong with mIRC colour codes?
01:32
< RichyB>
The grammar for them is idiotic.
01:33 Harlow [Harlow@Nightstar-2dbe3d64.il.comcast.net] has joined #code
01:33
< RichyB>
It's \003 [01[0-9]
01:33
< RichyB>
er
01:34
< AnnoDomini>
I know how to use it. Done it before.
01:34
< RichyB>
It's (with spaces between the octets for clarity): \003 [01]? [0-9] ("," [01] [0-9])?
01:35
< RichyB>
There's no way to specify "I want to set the foreground colour but not set the background colour for the string ',123'"
01:36
< AnnoDomini>
I see.
01:37
<&ToxicFrog>
Ahem
01:37
<&ToxicFrog>
,123
01:38
<&ToxicFrog>
^C<fg>^B^B<text>
01:38
< AnnoDomini>
Nice.
01:38
< RichyB>
ToxicFrog, it's bunny-ears-at-a-funeral-retarded that that was necessary but good show.
01:39
< RichyB>
er, I shouldn't be using ableist language like that but seriously mIRC's author is a bit awful
01:40
<&McMartin>
I've always interpreted those like "the ump is blind" - it's an insult because they are behaving in a manner that implies a societal duty of care rather than an exercise of the relevant responsibility
01:40
< RichyB>
McMartin, uh? Mis-channel?
01:41
< AnnoDomini>
RichyB: The current euphemism appears to be "developmentally challenged".
01:41
< AnnoDomini>
(But it doesn't matter. In due time, that too will be a cuss phrase.)
01:41
<&McMartin>
RichyB: No, it was with respect to "ableist language"
01:41
<@Namegduf>
I think the point is that you're not supposed to use describing someone as disabled as an insult, rather than that you're supposed to do so through a euphanism.
01:42
<&McMartin>
Right; I'm proffering an alternate theory of operation.
01:42
<@Namegduf>
Mostly you're just going to make that euphanism also become mostly known as an insult and so people have to move on to using something else to refer to actual people without being insulting.
01:42
< AnnoDomini>
Yes.
01:42
<@Namegduf>
Such is language.
01:42
<&McMartin>
Specifically: "Operating on the assumption that you are *not* in fact disabled, you should be better than this; this is why this is an insult"
01:43
< RichyB>
AnnoDomini, the core issue is not the naughty word, it's the bit where I referred to someone as disabled because I don't respect them to a certain extent.
01:43
<@Namegduf>
McMartin: As a model of human psychology and linguistics that fails to predict the actual responses we see.
01:43
< AnnoDomini>
Namegduf: Nay, such is stupidity of those who decide to chuck in more and more words down the contaminated pit and hoping that this time the word will continue to not be contaminated.
01:43
< RichyB>
oh Namegduf said it before me and said it better
01:43
<&McMartin>
RichyB: Well, there's your followup test: if it turns out the author of mIRC *were* in fact mentally damaged at the time of creation, would your rage remain, or would it transfer entirely to the people who standardized on it?
01:44
<&McMartin>
Much like someone would if they hired an umpire who was, literally, blind
01:45
<@Namegduf>
AnnoDomini: It's not like the supply of words is especially limited, and over extreme enough timespans you can even cycle.
01:45
< AnnoDomini>
Namegduf: I suppose so. It is wasted effort, though.
01:46
<@Namegduf>
It's more effortful to refer to people with a word that's also an insult and be clear you're not using it in the sense of an insult than to just use another construction, I think.
01:46
<@Namegduf>
People don't plot to make these things come about. They just do what's convenient for them and things happen.
01:46
<&McMartin>
There are like four strains of semantics here now
01:47
<&McMartin>
I suspect you are right but I can't untangle it into something useful.
01:47
< AnnoDomini>
Namegduf: For certain definitions of 'things happen', I'll agree.
01:47
<&McMartin>
("ableist language" is particularly sticky here, of course, because it *is* generally accepted that being disabled is not a value-neutral lifestyle choice)
01:48
< RichyB>
McMartin, can't answer that in the hypothetical, my brain won't stretch to it for some reason.
01:48
< AnnoDomini>
McMartin: (Indeed. Being disabled is a negative thing. I don't want to be disabled. Nobody should want to be disabled.)
01:48
<@Namegduf>
Much to the consternation of some deaf people, yes.
01:48
<@Namegduf>
(Who I don't agree with either)
01:48
<&McMartin>
That complicates matters, of course, because they are also a linguistic subculture (by necessity) and, well, linguistic subcultures totally exist and have that full set of baggage.
01:49
<@Namegduf>
Yeah, that sounds like a plausible explanation.
01:50
<&McMartin>
And it becomes grimly amusing from the outside because it's a linguistic subculture that is neither a superset nor a subset of the, er, unhearing.
01:51
<&McMartin>
("Hearing impaired", AIUI, is a strict superset?)
01:52
<@Namegduf>
Of deaf people? Yeah.
01:52
<@Namegduf>
It includes partial loss of hearing.
01:52
<@Namegduf>
Of people who use sign language? No, because of family and friends who can hear fine.
01:53
<&McMartin>
Right
01:53
<@Namegduf>
But also may know sign language and spend a lot of time in the culture as such.
01:53
<@Namegduf>
"allies"
01:53
< RichyB>
I think that there exist fully-deaf people who lack access to deaf subculture too, though?
01:53
<&McMartin>
Yes
01:53
<&McMartin>
17:46 <&McMartin> And it becomes grimly amusing from the outside because it's a linguistic subculture that is neither a superset nor a subset of the, er, unhearing.
01:56
<&McMartin>
Which is basically why I was claiming "you can't really lump deafness in with the other ones, it's a special case for clear external reasons"
01:57 * Vornicus wonders how you'd design it to not be silly
01:59
<@Namegduf>
Society, or?
02:00
<@Namegduf>
Or is this going back to something more on topic?
02:00
<@Namegduf>
XD
02:00
<&McMartin>
Speaking of
02:00
<&McMartin>
I'm gonna need realloc().
02:00
<@Namegduf>
That sounds worrying.
02:00
< AnnoDomini>
Vornicus: XML! XML everywhere!
02:01
<~Vornicus>
IRC color codes
02:01
<@celticminstrel>
It's sort of a pity that C++'s new/delete has no realloc equivalent.
02:02
<&McMartin>
Bracket the color codes with the escape, like boldface
02:02
<&McMartin>
celticminstrel: std::vector, unironically
02:02
<@celticminstrel>
Heh, I guess that's true.
02:02
<&McMartin>
But I don't get to use C++ here
02:02
<~Vornicus>
so ^C4,7^Cyaddayaddayadda... and then the end is...?
02:03
<@celticminstrel>
^O, right?
02:03
<@celticminstrel>
Or ^C without a following number. That might work too.
02:03
<@celticminstrel>
Oh, um... I'm pretty sure it's ^C4,7yadda
02:03
<&McMartin>
Yeah, that's what it is
02:03
<&McMartin>
The question is what it *should* be
02:04
<@celticminstrel>
Well, what Vornicus said would be a better way to do it.
02:04
<&McMartin>
yeah
02:05
<@celticminstrel>
Maybe something like ^C-,-^C to cancel colour codes?
02:05
<@Tarinaky>
irc colour codes aren't that bad are they?
02:05
<@celticminstrel>
They kind of are.
02:05
<&McMartin>
17:31 < RichyB> There's no way to specify "I want to set the foreground colour but not set the background colour for the string ',123'"
02:06
<&McMartin>
TF proved that wrong by proving the general case right~
02:06
<@celticminstrel>
Um, what?
02:06
<&McMartin>
It's the same trick as:
02:06
<&McMartin>
02:06
<@celticminstrel>
You can set just the foreground colour with ^C4stuff.
02:06
<@celticminstrel>
And I think you can set just background with ^C,4stuff.
02:06
<&McMartin>
Right, unless "stuff" can be interpreted as a background color.
02:06
<&McMartin>
say, because it starts with a comma.
02:06
<@celticminstrel>
Which would require it to start with comma... ah, yes.
02:07
<@celticminstrel>
That's the general issue with them, anyway.
02:07
<&McMartin>
And thus, an end-delimiter solves it
02:07
<@Tarinaky>
I always thought the bigger issue was that they existed at all.
02:08
<@celticminstrel>
You can't just insert an end-delimiter and have it work though, because ^C without a valid colour following it turns the colours off, as I recall.
02:08
<&McMartin>
So make *that* be ^C^C
02:08
<&McMartin>
This is assuming mIRC doesn't exist
02:08
<@celticminstrel>
No, I meant using the existing system, inserting a ^C as an end delimiter would fail.
02:08
<&McMartin>
Well, yes
02:09
<&McMartin>
If you have to keep working with the existing bad design, you use ^B^B as your delimiter.
02:10
<@celticminstrel>
Ah, that'd work.
02:11
<@celticminstrel>
I seem to remember some kind of proposal for a new way of doing them.
02:15
< RichyB>
A proposal for a better system is attached to the ctcp spec.
02:15
< RichyB>
Never took off.
02:17
< RichyB>
http://i-n-v-i-s-i-o-n.com/cms/index.php?option=com_content&task=view&id=41&Item id=50
02:18
< RichyB>
It's not *terrific*, though.
02:25
< RichyB>
giggling https://gist.github.com/quchen/5280339
02:34 Derakon[AFK] is now known as Derakon
02:38
<@Tarinaky>
o.O
02:38
<@Tarinaky>
Somedays I wonder why on earth I choose to be part of this sub-culture.
02:39
<@Tarinaky>
But then I guess I hate fun :/
02:40
<~Vornicus>
??
02:41
<@Tarinaky>
Trolling.
02:41
<@Tarinaky>
The calling of people faggots.
02:41
<@Tarinaky>
That kind of thing.
02:41
<@Tarinaky>
The fact that this is considered 'okay' and 'fun'.
02:42
< RichyB>
oh really I find trolling fucking vile
02:42
< RichyB>
even good trolling I find distasteful
02:43
< RichyB>
that was just funny to see a screaming asshole get talked down and talked into trying to do sometthing productive instead
02:44
<~Vornicus>
that was great
02:44
<@Tarinaky>
Steady now, I'm not sure learning Haskell counts as productive :p
02:45
<~Vornicus>
"i just want to get kicked out of a bunch of channels for fun. why is no one cooperating with me"
02:47
<@Tarinaky>
Any suggestions on how to tell a possibly beligerant-drunk to please stop slamming doors at 3am because I have a long car journey ahead of me today :
02:48
<@Tarinaky>
So far the only advice I've been given is 'throw a fire extinguisher at them'
02:48
<@celticminstrel>
That sounds like a bad idea.
02:48
<@Tarinaky>
Hence why I ask here.
02:49
<~Vornicus>
"hey fucktard, it's 3am, quit that racket already"
02:53
<&McMartin>
ToxicFrog: http://steamcommunity.com/groups/homestream
03:44 Harlow [Harlow@Nightstar-2dbe3d64.il.comcast.net] has quit [Connection closed]
03:48 Thalass [thalass@Nightstar-gucq71.bigpond.net.au] has joined #code
04:33
<@Tarinaky>
"The code was ‚well documented‚, in the way that all terrible code is well documented:"
04:34
<@Tarinaky>
"The original architect- not poor Sam, who only touched the project in its final days- didn‚t use design patterns to overcome common problems. Instead, he invented problems to justify trying every permutation of every design pattern- real or imagined."
04:35
<&ToxicFrog>
McMartin: AAAAAAAAAAAAAAA
04:47 Turaiel[Offline] is now known as Turaiel
04:58 Thalass [thalass@Nightstar-gucq71.bigpond.net.au] has quit [[NS] Quit: werk omg]
05:17
<&McMartin>
ToxicFrog: Anyway, I noticed that there is a group for announcements for this and that you weren't on it and you were complaining about wanting it shipping already, so...
05:17 McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [[NS] Quit: reboot]
05:22 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!]
05:24 McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code
05:24 mode/#code [+ao McMartin McMartin] by ChanServ
06:29 Turaiel is now known as Turaiel[Offline]
06:40
<~Vornicus>
okay. I think that's game startup sorted. I need to figure out what things I can test here.
06:41 VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [[NS] Quit: Program Shutting down]
06:52 RichyB [RichyB@Nightstar-c6u.vd5.170.83.IP] has quit [[NS] Quit: Gone.]
06:55 RichyB [RichyB@Nightstar-c6u.vd5.170.83.IP] has joined #code
07:33
<&McMartin>
High time I started actually breaking this stuff down into functions instead of notes on physical paper
07:37 Kindamoody[zZz] is now known as Kindamoody
07:46 Orthia [orthianz@Nightstar-avg.1ee.224.119.IP] has joined #code
07:46 mode/#code [+o Orthia] by ChanServ
08:28 Stalker [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code
08:37 * Vornicus fiddles further, finds a thing he needs to fix because he needs to get more information.
08:42 * Vornicus ...tries to figure out what things he needs to do to actually extract this information.
08:49 * jerith hands Vornicus the yak-razor.
08:49
<~Vornicus>
thank you sir.
08:50 AverageJoe [evil1@Nightstar-fb1kt4.ph.cox.net] has joined #code
08:52
<&McMartin>
Too lazy to actually write the functions now that I've broken it down into them
08:52
<&McMartin>
I think I will apply virtual battle axes to faces and then proceed to bed
08:54
<&McMartin>
Aha
08:54
<&McMartin>
MMF2 has a semi-standard extension that lets you hook Lua scripts to events
08:54
<~Vornicus>
in what game are you applying virtual battle axes?
08:54
<&McMartin>
That would in fact graduate it into an actual system
08:54
<&McMartin>
Game: Ys Origin
08:54
<~Vornicus>
aha
08:54
<~Vornicus>
Oh, right, you posted that screenshot
08:54
<&McMartin>
In which the default main character is an extremely anime young girl who battle axes everything in the face.
08:55
<~Vornicus>
which I need to go find again because Vash needs to see it.
08:55
<&McMartin>
She is also, as she accomplishes things of note, Actually Getting Some Goddamned Respect, which is very much *not* Every JRPG Ever
08:59
<~Vornicus>
Okay. Before I ask the user for regions (and thus wait for the callback to come in) I need to 1. detect what regions are available and 2. detect what regions are contiguous, so that we can either test whether the selected regions are contiguous or create a list of possible region selections.
09:01
<@Namegduf>
McMartin: Is it a good RPG?
09:01
<@Namegduf>
I vaguely remember having seen it.
09:09
<&McMartin>
It's, um
09:09
<&McMartin>
Not much of an RPG
09:10
<&McMartin>
It's more like the dungeons in Recettear at double speed
09:10
<~Vornicus>
Therefore: map loading, and region selection, must be two different parts; also, I must reduce the graph to regions.
09:10
<&McMartin>
Fortunately, I don't like JRPGs much but I *do* like running around with big fighty things
09:11
<@Namegduf>
Ah.
09:12
<@Namegduf>
I've not played Recettear either.
09:12
<&McMartin>
Um, let's see
09:12
<&McMartin>
Do you have a list of touchpoints I can use~
09:12
<~Vornicus>
have you played Secret of Mana or Diablo
09:12
<@Namegduf>
But I can get the idea. actiony monster smashing.
09:12
<@Namegduf>
I played Gauntlet.
09:12
<@Namegduf>
A recent one in the series.
09:12
<~Vornicus>
namegduf needs food badly
09:12
<@Namegduf>
Exactly.
09:12
<&McMartin>
OK, Ys games generally are very much like Gauntlet in that combat used to involve rubbing up aggressively against monsters which automeleed
09:12
<&McMartin>
Now you have to melee by hand, but so do they
09:12
<&McMartin>
And it's isometric viewpoint
09:12
<@Namegduf>
XD
09:13
<&McMartin>
What I hear is that Ys Origin is pretty mediocre by the standards of the series, in which case I'm looking forward to that other Ys game I got at random at the same time.
09:13
<&McMartin>
(I think it was some Steam thanksgiving sale pack)
09:13
<&McMartin>
As for the plot, well.
09:13
<&McMartin>
http://steamcommunity.com/sharedfiles/filedetails/?id=203698674
09:14
<&McMartin>
Another way of thinking of it would be "2D Zeldas without the special items and with having to be more careful about how you do combat filling in the gap"
09:15
<@Namegduf>
XD
09:16
<@Alek>
I remember Ys III.
09:16
<@Alek>
but this sounds vaguely different.
09:16
<&McMartin>
Alek: Yeah, it's got a more diablo-style 2d isometric-with-jumping...
09:16
<&McMartin>
...
09:17
<&McMartin>
... Flicky's Island/Sonic 3D Blast...
09:17
<&McMartin>
... kind of flavor
09:17
<@Alek>
ahahaa
09:17
<&McMartin>
Ys III was recreated using this engine and this is what "Oath in Felghana" is
09:17
<@Alek>
so basically, PS1 flavor 3D/isometric?
09:17
<&McMartin>
I guess
09:17
<&McMartin>
I can't think of any actual PS1 examples offhand; I spent that era playing 2D Metroidvanias and space shooters
09:18
<&McMartin>
Anyway, that screencap is also what combat looks like except the enemies are trying to put sharp or explodey things in you and so you return the favor
09:20
<@Namegduf>
Through weird associations I am now missing Baldur's Gate, which is a very different type of game.
09:20
<@Namegduf>
I should go back to that eventually.
09:21
<@Namegduf>
I need to do my overpowered fighter/magic user dual class.
09:21
<&McMartin>
BALDUR DASH.
09:21
< Shiz>
z.
09:21
<&McMartin>
A word-game time-management sim where you dig aruond getting diamonds in the forgotten realms.
09:21
<@Namegduf>
XD
09:23
<&McMartin>
Anyway: Yeah, I've been enjoying Ys Origin. I hear Oath in Felghana's better, though. If so, that speaks relatively well for it
09:23
<&McMartin>
But yeah, they're not like Diablo II or similar games that people will play near-exclusively for like ten years, no
09:25
<@Alek>
PS1, isometric? all I can think of is Vagrant Story.
09:25
<&McMartin>
I've heard of that, but I haven't actually played it.
09:25
<@Alek>
which was more 3D, but definitely an isometric view.
09:25
<@Tarinaky>
I am hungry but the shops aren't open yet.
09:25
<&McMartin>
Maybe I didn't understand what you meant by "PS1 flavor 3D/isometric"
09:25
<@Alek>
also, FFT.
09:25 * Tarinaky frowns.
09:25
<@Tarinaky>
Send food.
09:25
<&McMartin>
Mmm, FFT
09:25
<@Alek>
which is more of a classical tile-based isometric.
09:26
<&McMartin>
Yup
09:26
<&McMartin>
FFT and Einhänder are my two favorite Square games.
09:26
<&McMartin>
(At least, that they actually did. They don't get to retroactively have Deus Ex. =P)
09:27
<~Vornicus>
I enjoyed Vagrant Story
09:27
<~Vornicus>
I could never get into FFT; the first fight with mages killed me every time.
09:51 Kindamoody is now known as Kindamoody|afk
10:01 AverageJoe [evil1@Nightstar-fb1kt4.ph.cox.net] has quit [[NS] Quit: Leaving]
10:21 You're now known as TheWatcher
10:56 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has quit [[NS] Quit: Leaving]
11:45 Kindamoody|afk is now known as Kindamoody
12:42 teapot [teapot@Nightstar-ig4.h2p.244.103.IP] has joined #code
12:44
< teapot>
inb4 LURK MOAR; Does anyone here have experience configuring an NTP server without access to a reference source?
12:45
< teapot>
I'm up on ntp.org reading through everything, but I can't see an actual explanation to the ntp.conf file.
12:56 teapot [teapot@Nightstar-ig4.h2p.244.103.IP] has quit [Ping timeout: 121 seconds]
13:00
<@Tamber>
I know he's vanished, but the config file *is* pretty well commented~
13:00
<@Tamber>
And if you can't read the config file you're editing, you're going to have a bad day.
13:01
< Shiz>
maybe his distro ships a gimped config file
13:16
<@TheWatcher>
Tamber: I vaguely note that, while the ntp config is often commented, the entire thing is made entirely of spiders
13:17
<@Tamber>
Yes, but they're *commented* spiders~
13:17 dmiles_akf [dmiles@Nightstar-hdcn1r.or.comcast.net] has quit [Connection reset by peer]
13:19
<@TheWatcher>
It's a very good example of a piece of software developed by people who have absolute knowledge of the system they are implementing, and are utterly incapable of properly explaining it for normal human beings, utterly unwilling to make it behave in a sane fashion, and believe that incredibly obtuse and unhelpful status messages are all the debugging info you need
13:21 dmiles_afk [dmiles@Nightstar-hdcn1r.or.comcast.net] has joined #code
13:22
<@TheWatcher>
(why yes, I do have an axe to grind with the ntp devs~. Part of why I switched to the "Horrible, Renegade, Totally Not An NTP Implementation You Disgusting Horrors" OpenNTPd is because of the hilarious problems I've had with ntp.)
13:22 * Shiz is just glad he has no need to run an ntpd
13:23
< Shiz>
>>> Emerging (8 of 176) sys-devel/patch-2.6.1
13:23
< Shiz>
>>> Jobs: 7 of 176 complete, 1 running Load avg: 1.13, 1.29, 1.09
13:23
< Shiz>
My CPU approves.
14:02 Kindamoody is now known as Kindamoody|out
14:25 Stalker [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds]
14:44 VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code
14:46 celticminstrel [celticminst@Nightstar-gj43l1.dsl.bell.ca] has joined #code
14:46 mode/#code [+o celticminstrel] by ChanServ
14:55 Stalker [Z@Nightstar-ro94ms.balk.dk] has joined #code
14:59 Stalker [Z@Nightstar-ro94ms.balk.dk] has quit [Ping timeout: 121 seconds]
15:03
<@froztbyte>
man ntpd.conf~!
15:04
<@froztbyte>
(late to the party, but meh)
15:11
< Shiz>
ntpclient ftw :p
15:13 Stalker [Z@Nightstar-ro94ms.balk.dk] has joined #code
15:22 Kindamoody|out is now known as Kindamoody
15:40
<&ToxicFrog>
User comes to me with a bug report.
15:41
<&ToxicFrog>
An hour later I've fixed five other bugs and implemented two features, but not the bug they originally reported.
15:42 * Syka buys shares in the ToxicFrog Yak Shaving Company
16:09
<@froztbyte>
Haha
16:29 VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [[NS] Quit: Program Shutting down]
16:34 VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code
18:04 Harlow [Harlow@Nightstar-b902fba7.chi.megapath.net] has joined #code
18:10 Harlow [Harlow@Nightstar-b902fba7.chi.megapath.net] has quit [Client closed the connection]
18:12 Harlow [Harlow@Nightstar-b902fba7.chi.megapath.net] has joined #code
18:24 Turaiel[Offline] is now known as Turaiel
18:56 Harlow [Harlow@Nightstar-b902fba7.chi.megapath.net] has quit [[NS] Quit: BED]
20:49 Netsplit *.net <-> *.split quits: @iospace, Reiver, VirusJTG, @himi, Syka, @McMartin, @gnolam, Stalker, dmiles_afk, @Namegduf, (+1 more, use /NETSPLIT to show all of them)
20:50 Netsplit over, joins: RichyB, Reiver, VirusJTG, @gnolam, Syka, Stalker, @himi, @iospace, &McMartin, dmiles_afk (+1 more)
21:04 Orthia [orthianz@Nightstar-avg.1ee.224.119.IP] has quit [Ping timeout: 121 seconds]
21:05 Orthia [orthianz@Nightstar-avg.1ee.224.119.IP] has joined #code
21:05 mode/#code [+o Orthia] by ChanServ
21:05 Syloq [Syloq@Nightstar-ujl.ett.136.198.IP] has quit [Operation timed out]
21:08 Netsplit *.net <-> *.split quits: @iospace, Typherix, Attilla, AnnoDomini, @Orthia, VirusJTG, Reiver, @celticminstrel, simon_, @Reiv, (+17 more, use /NETSPLIT to show all of them)
21:09 Netsplit over, joins: RichyB, Reiver, @Orthia, VirusJTG, @gnolam, Syka, Stalker, @himi, @iospace, &McMartin (+2 more)
21:11 Attilla [uid13723@Nightstar-ed0oqj.irccloud.com] has joined #code
21:11 simon_ [simon@Nightstar-2og823.pronoia.dk] has joined #code
21:11 Typherix [Typherix@Nightstar-n91qrf.lnngmi.sbcglobal.net] has joined #code
21:12 ErikMesoy [Erik@Nightstar-3kot9q.80-203-17.nextgentel.com] has joined #code
21:12 ToxicFrog [ToxicFrog@Nightstar-gtjrvq.dsl.teksavvy.com] has joined #code
21:14 jerith [jerith@Nightstar-ngcnma.slipgate.za.net] has joined #code
21:14 jeroud [sid10043@Nightstar-a1k27g.irccloud.com] has joined #code
21:15 jeroud [sid10043@Nightstar-a1k27g.irccloud.com] has quit [Z-Lined: Your IP range has been attempting to connect too many times in too short a duration. Wait a while, and you will be able to connect.]
21:16 ToxicFrog [ToxicFrog@Nightstar-gtjrvq.dsl.teksavvy.com] has quit [Z-Lined: Your IP range has been attempting to connect too many times in too short a duration. Wait a while, and you will be able to connect.]
21:16 iospace is now known as IO\PACKERS
21:17 Xon [Xon@Nightstar-q4s.ku7.252.119.IP] has joined #code
21:18 Turaiel [Brandon@Nightstar-vku52k.resnet.mtu.edu] has joined #code
21:21 celticminstrel [celticminst@Nightstar-gj43l1.dsl.bell.ca] has joined #code
21:22 Alek [omegaboot@Nightstar-qa936g.il.comcast.net] has joined #code
21:22 Turaiel is now known as Turaiel[Offline]
21:23 AnnoDomini [abudhabi@Nightstar-d00ld9.gw-out.freebnc.net] has joined #code
21:23 Turaiel[Offline] is now known as Turaiel
21:24 Azash [ap@Nightstar-25v57p.net] has joined #code
21:29 jeroud [sid10043@Nightstar-a1k27g.irccloud.com] has joined #code
22:19 ToxicFrog [ToxicFrog@Nightstar-gtjrvq.dsl.teksavvy.com] has joined #code
22:19 Reiv [NSwebIRC@Nightstar-95746c1f.kinect.net.nz] has joined #code
22:19 PinkFreud [WhyNot@NetworkAdministrator.Nightstar.Net] has joined #code
22:19 ServerMode/#code [+oo Reiv PinkFreud] by *.Nightstar.Net
22:19 mode/#code [+o Alek] by ChanServ
22:19 mode/#code [+ao jerith jerith] by ChanServ
22:19 mode/#code [+o celticminstrel] by ChanServ
22:20 ErikMesoy is now known as ErikMesoy|sleep
22:25 mode/#code [+ao ToxicFrog ToxicFrog] by ChanServ
22:32 Kindamoody is now known as Kindamoody[zZz]
22:52 himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds]
22:58 * McMartin reads backscroll
22:58
<&McMartin>
It occurs to me that we are the opposite of LURK MOAR here; we want you talking soon, but idling afterwards so someone can have time to get around to it.
22:59 Stalker [Z@Nightstar-ro94ms.balk.dk] has quit [Ping timeout: 121 seconds]
23:06
<@Reiv>
What was this?
23:07
<&McMartin>
Newbie came in with a question and apologized for just asking his question like we want
23:08
<@Tamber>
Then pretty much immediately left.
23:31 Stalker [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code
23:32
<@Reiv>
Tsk
23:50 Vornicus [vorn@Nightstar-sn7kve.sd.cox.net] has joined #code
23:50 mode/#code [+qo Vornicus Vornicus] by ChanServ
--- Log closed Mon Dec 16 00:00:20 2013
code logs -> 2013 -> Sun, 15 Dec 2013< code.20131214.log - code.20131216.log >

[ Latest log file ]