« first day (2027 days earlier)      last day (3289 days later) » 
00:00 - 13:0013:00 - 00:00

Anonymous
00:00
@JourneymanGeek comes with cleaning supplies
cleaning supplies?
Anonymous
sponges
That joke's even worse than my sudo women joke.
@I'manartist : or yahoo's potentially using HEAD to obtain the filesize instead for same purposes.. not sure why they would waste bandwidth. Wonder if HEAD request triggers clicks. (pondering mind)
i fail @ names
00:01
Not really....
@Dave I'd use first few letters + tab
yea i love the autocompletion
though since we have a name collision, I'm using one of @OliverSalzburg's scripts so I can hit : and up so I can reply
@ + i + tab. It's just 3 keystrokes
yup
@JourneymanGeek But yet, none of your replies are actually replies :-/
00:03
@qasdfdsaq sometimes. Sometimes not. Maybe
Eh, pretty much all of them except that one!
Bob
Bob
@PatoSáinz MD5 is still considered strong against (second-)preimage attacks like you're suggesting.
I can pretty much never tell which message of mine you're replying to :-/
/me wonders if the 'some replies' is related to the rotating of the transcript log file.
Anonymous
@Bob k, nice to know
00:04
@qasdfdsaq I actually forget most of the time. I'm just using replies so I don't send notices to both daves.
Bob
Bob
@PatoSáinz MD5 is no longer considered collision resistant. That means that you can, relatively easily, find two inputs that hash to the same output.
Ah
We have two daves?
@Bob: This is true
Oh yeah. It shows two icons next to @Dave when you autocomplete
Anonymous
00:05
@Bob ya
My name's easy to distinguish between the two though.. There's non annoying dave.. and annoying dave... i'm probably the annoying one.
Bob
Bob
@PatoSáinz Preimage resistance means it's hard to find an input that hashes to a given output (for all possible outputs). Second-preimage resistance means it's hard to find a second input that hashes to the same output as a given input.
Database Dave and sneaky dave!
Lol, and I just pointlessly @'d both of them
YES dba dave @JourneymanGeek also it was a 72 hour migration.
Anonymous
00:06
@Bob I see
(since the other dave's ava is a shoe
boot to the head dave
Anonymous
I finally finished setting up my RHEL cluster
Anonymous
it works like magic
Anonymous
fockin proud
00:07
@PatoSáinz: Interesting! That sounded like a fun project.
Bob
Bob
@PatoSáinz Basically - MD5 is weak when the attacker can control both sides (e.g. generating a legitimate certificate and a forged one, and getting the collided signature signed, or generating two files with the same signature and getting one of them verified).
Anonymous
I followed that guide but I added my own twists
Bob
Bob
But it's still considered strong when the attacker can only generate one of the inputs and must match a given output.
oh nice
Bob
Bob
00:08
Still not recommended to use it, of course, since there's better available.
Anonymous
@Dave it's >10k words long iirc
Bob
Bob
But it's not as weak as the 'broken collision resistance' immediately implies.
Anonymous
it's a big toolstack to learn but after you are accustomed it's all fun
I want to do one of these en.wikipedia.org/wiki/Beowulf_cluster
Anonymous
@Bob oh I thought it'd be trivial to have two files with same hashes
Bob
Bob
00:10
@qasdfdsaq Just like the tale of two Lukes, giving us @CanadianLuke.
And the two Bobs I guess :P
@PatoSáinz not trivial.
@Dave Oh I have one of those
Had to do a training session on it today
Bob
Bob
@PatoSáinz Not quite trivial, but easier than desired -- when you generate both files.
Anonymous
@Dave sounds fun
Anonymous
in my case this cluster is for HA
Bob
Bob
00:10
When you're only generating one of the files, it's much much harder.
Anonymous
so it's not actually sharing a workload
@Bob Also when the file has to still contain a working Windows ISO
@qasdfdsaq: I hate you. I want one.
lol
Anonymous
and we actually lose 50% of total performance due to the need of all machines having to be able to run their own services and the other's at the same time
@Dave Why?
00:11
@PatoSáinz Hey cool, that guide uses our servers. :D
Anonymous
so we price it adequately
Bob
Bob
@qasdfdsaq Actually, because a Windows ISO is so big that's arguably not too hard :P
Anonymous
@MichaelFrank fujitsus?
Bob
Bob
At least compared to much smaller files.
@PatoSáinz Yup
00:12
@qasdfdsaq: Want to crack some passwords bro.
@Bob Well... It has to defeat sfc post-install... and contain your desired payload, and in an effective location (e.g. kernel or services)
Bob
Bob
There's a fair amount of cruft in a Windows ISO that you could manipulate without breaking things - or at least not immediately breaking things.
@Dave Amazon is cheaper bro
Bob
Bob
@qasdfdsaq True - the payload makes it much harder.
Anonymous
@MichaelFrank by our servers you mean you work at fujitsu or you use them in prod?
00:13
I'm either going to crack passwords or.. calculate the next 300 prime numbers that arent found yet.
@PatoSáinz Both, I guess.
Anonymous
@Dave well they do pay some good money for finding primes
Amen to that @PatoSáinz
Anonymous
@MichaelFrank oh, cool!
Anonymous
yea, the guide really suggests using them: and their commercial offering also has them. They seem nice.
00:14
@PatoSáinz: Grats on your setup! I'm out of here.. I need sleep peace
Anonymous
@Dave haha thanks. It's been a ride.
Anonymous
@MichaelFrank what's your job title?
@PatoSáinz Technical Services Rep.
Anonymous
@MichaelFrank is that like on-site support?
@PatoSáinz Yep.
00:18
Sleeeeeeeep
Anonymous
@MichaelFrank nice
Bob
Bob
LOL
-8
Q: create a variants of MD5

Grijesh ChauhanI have also asked similar q here : To create a variants of MD5, I made following changes : MD5 uses a non-linear sin(i)* pow(2,32) ----> i plane to use cos(i)*pow(2,32) Instead original values of A, B, C, D that are four initial seeds( or states), that changes additively during ...

For the love of Scheneier, why??? — Gilles Oct 8 '12 at 23:50
Anonymous
@Bob loool
Its client requirement only. Also I want to know. — Grijesh Chauhan Oct 8 '12 at 10:31
I would have told that client to fuck right off. go forth and multiply.
Anonymous
@qasdfdsaq amen
Bob
Bob
00:24
> And twisting the function internals in the hope of making the function "better" is akin to performing brain surgery with a soldering iron: at least, it will make the failure indisputably spectacular.
@Polynomial Actually I just wanted how variant effect properties of MD5. I did just for fun/learning and understanding purpose. Random number generation I written wrongly I just wanted to check whether my variants are weaker then Original. I just posted my good question in wrong way- Thanks for your response. — Grijesh Chauhan Jul 8 '13 at 19:59
Something doesn't add up here.
"client requirement" and "fun/learning"
Well, put it that way, perhaps the client is requiring him to have fun and learn?
Huh, Galaxy S7 is getting Snapdragon 820 in the US and Exynos 8890's in the UK
Both using a Qualcomm X12 modem tho
Dang that's nice. I want it even more now.
Bob
Bob
@qasdfdsaq Do you prefer Exynos or Snapdragon?
Usually, Snapdragon because the kernel optimisations are more mature. However, some very good technical analyses on the Exynos 7420 (S6) found it to be a very advanced chip.
Bob
Bob
I've never had an Exynos phone.
Also, go to sleep :P
I believe my Galaxy S2 was an Exynos chipset.
But all subsequent Galaxy phones (S3, S4, S5) were Snapdragons.
One thing I really hated about the Exynos was the Samsung baseband/modem was utter shite. But the S7 uses an excellent Qualcomm modem with an excellent Exynos CPU so the best of both worlds.
Sleep? It's only 00:35.
I normally go to sleep at like, 2am.
OMG the bot has become sentient
yay the repl works
ok, now to get console response
Egads, so the S7 does have one backward step from the S6/S5/S4, a fully exposed and non-recessed screen
Bob
Bob
The S6 screen is recessed?
Wait, so the screen isn't curved?
The edge has a curved screen.
Bob
Bob
00:42
@qasdfdsaq I meant a concave curve.
Must've mixed up the G5 in my head.
Nope, it's slightly convex at the edges
Bob
Bob
Ah.
A bit like the S3, where the screen protrudes out from the housing plastic that surrounds it.
I really disliked that, since it exposed it a lot more, and made it slip around a lot too.
> I suspect that the 1.4 micron pixel size here may not be entirely due to a desire to improve low light performance though, as a smaller pixel split in half may encounter strange quantum effects that affect image quality.
Eh? Reverse quantum dot camera distortion? That's new
@qasdfdsaq The S6 used an Exynos in all regions, which was a very odd decision. Samsung normally uses Snapdragons for the NA market and Exynos elsewhere—if I understand it correctly, the Exynos isn't designed to handle all the different bands we have here in the US, necessitating Qualcomm's modem.
Well the S3, S4, and S5 were all Snapdragons in the UK market. Only the S2 was an Exynos.
Exynos modems were always just crappy. Not just for bands, but for LTE support (they were generations behind) and reliability.
Bob
Bob
00:52
@bwDraco We get Snapdragons in AU
Actually, I always thought Exynos was primarily for the US.
Given they're combining Snapdragon modems with Exynos CPUs outside the U.S., and using the same Snapdragon radio in the US version, band support can't be the deciding factor.
Bob
Bob
@qasdfdsaq Perhaps power consumption? Dunno how good the modem integration is going to be.
What I'm reading is hinting at supply issues. There aren't enough Exynos processors to go around, so they use Snapdragon processors for the US market.
@Bob Well it's the same modem in both. Given the same CPU on different fabs have different power consumption characteristics (i.e. Apple) there's certainly going to be differences between the Snapdragon and Exynos. But quite why they don't just use the one with best consumption everywhere is a mystery.
Hmm, I heard mentions of the not-enough-Exynos-supplies thing before. It's been going round since the S3.
Remember that the US market for smartphones is insanely large.
00:57
So is the rest of the world.
Actually, come to think of it, China is the biggest smartphone market, and they're getting the Snapdragon too
Bob
Bob
Meanwhile, the AU market is comparatively minuscule and we got Snapdragon for the S4. Not too sure what happened with the other ones.
But if the two biggest markets are getting Snapdragons, why bother with doubling the hardware and software design requirements to make a second version for just "the rest" anyway?
Bob
Bob
I usually end up with overseas variants.
May be for marketing reasons. Snapdragon has better brand awareness than Exynos.
Bob
Bob
Exynos has always felt a bit like a NIH thing.
00:59
I mean it's not just a drop-in replacement, it needs a whole new motherboard, kernel, driver stack, and what not.
Meh. Samsung are weird.
I look forward to the Exynos though. The previous chip was very efficient. The previous Snapdragon... wasn't, but detailed analysis of the 14nm Snapdragons are thin on the ground right now.
Bob
Bob
Oh yea, the other thing is Snapdragon is using Qualcomm core designs again.
So that could be completely different (for better or worse) than the Cortex cores in Exynos.
Hm. Also interesting that it's apparently 2+2..?
Now that SD820 devices are out there and Qualcomm Kryo is in production products, what's next?
What about the new Hexagon 680 DSP? How well does it cope with things like 4K video playback?
(I've never fully understood the purpose of a DSP in this application.)
We always talk about the CPU, GPU, and cellular radio components in an SoC, but what about the other aspects of the chip?
I've also failed to understand how SoC fabs manage to create different parts of a chip that have different power characteristics on the same die.
There's no benefit in doing big.LITTLE with two sets of differently-clocked Cortex-A53 cores if the silicon is the same across the die—you might as well use a homogeneous octa-core design and down-clock the cores when full power is not needed.
01:24
I've actually seen both types of designs. Snapdragon 615 uses big.LITTLE but both clusters use Cortex-A53 cores. Mediatek Helio X10 uses eight Cortex-A53 cores all capable of running at 2.2 GHz.
What the heck is the point of big.LITTLE if both clusters use the same core?
I just don't get it. Why?
01:39
@bwDraco typically a DSP does one 'simple' job well
for example listening in so it can power up the rest of the processor(s) only if needed
@JourneymanGeek Yup. "Ok Google" hotword detection even with the screen off without draining the battery.
@bwDraco or say mp3 decoding, or video decoding
so even if your main processor lacks the power to decode 4k, you could 'relatively' cheaply have a specialised DSP to just handle x264/5
4K video playback, advanced image processing, or just listening for "Ok Google"—while not draining the battery the way firing up the AP does.
Draining less of it
but the tradeoff is these processors may not be useful for anything except a narrow range of tasks
Purpose-built DSPs draw far less power than general-purpose APs.
01:42
From a desktop standpoint, this is what AES NI and quicksync do.
and faster
Every last minute of battery life counts, every last microwatt. That's why.
We're talking absolutes
and practically? Most phones probably spend a good chunk of the day connected to power
It also frees up processing power for more important tasks.
Like watching cat videos
Just enabled always-on "Ok Google" detection on my Nexus 5X. Not sure as to the extent it drains the battery, though.
...for some reason, it doesn't work. Disabling the feature.
01:53
work, dammit, work!
>_>
Either someone is trying to get Cavil to work, or the damn thing went rampant ;p
you didn't see anything
Bob
Bob
> bot.adapter.out.add('you didn\'t see anything')
Attempting to run "bot.adapter.out.add('you didn\'t see anything')"
$ null
Bob
Bob
@JourneymanGeek Just happens that some commands (like bot) kill it ... probably cause there's a cycle in there
02:02
cycle?
loop?
Bob
Bob
@JourneymanGeek Consider:
var a = {}
var b = {foo: a}
a.bar = b
Now, try to print that out.
Bob
Bob
It ends up looking something like this: {bar: { foo: { bar: { foo: ... } } } }
and it all goes kaboom?
Bob
Bob
Now, somewhere in webdriverio or in selenium or I don't know where it seems to be getting stuck in a loop like that.
I'm fiddling with timeouts to try to get it to break out in those cases.
test
I'm alive! (on PJS)
Hi @allquixotic!
Bob
Bob
only problem is getting the browser logs doesn't seem to work right :\
Hi @allquixotic!
Bob
Bob
whoops
uhm...
02:19
I'm probably not making the jump to VR. This laptop definitely does not have the graphics hardware for anything beyond low-fidelity VR and may drop frames.
It's a two-year-old machine. Top-of-the-line graphics hardware for the day, but can't cope with the latest games above or even at high and overheats if I do anything that runs the GPU at or near 100% for more than a few minutes.
GPU thermal throttling is one of my biggest complaints with this machine, and I have to put a fan behind the GPU exhaust vent to maintain stable performance.
The machine sits on a passive laptop cooler to aid airflow but that's not enough.
@bwDraco older, though heavily upgraded gaming desktop
02:34
Concerns about full desktop GTX 980 GPUs overheating when run in laptops.
Nobody mentioned thermal throttling as an issue at launch.
Laptop cooling is certainly getting better with larger fans and more advanced heatsink designs but there's no getting around the fact that you're trying to dissipate 150W or more from within a very constrained space.
Bob
Bob
@bwDraco Whoever would've guessed?!
I've honestly had it with my laptop's GPU hitting 93 °C and throttling during an Assassin's Creed III session.
In fact, I had overheating issues with my laptop mere months after I got it while I was playing Crysis Warhead.
The game would suddenly slow down dramatically (to the point where it became unplayable) after some time.
I then downloaded GPU-Z in an attempt to understand what was going on, and it became obvious that the GPU was overheating and throttling.
it is not like there is ANY game that uses the gpu sparcely, giving it time to cool, so if a computing device is going to have the GPU and the cooling ability to play the tough stuff , it has to keep up with it as a constant. it would not bother me that much kniwing there are 500+ good games that are not the top of the line or poorly optomised, one can play the "other" ones for the rest of thier life .
That's the biggest issue with gaming laptops. Yes, you can carry them with you easily, but you should not expect them to maintain full performance under several minutes of full-load operation.
Gaming laptops are becoming more popular than ever with the falling popularity of desktops and the shift to mobile devices, but there are concessions to be made with the more portable form factor.
At least the GPU is not running with everyday desktop apps (thanks, NVIDIA Optimus) so battery life, while not exactly great (the age of the battery is a factor), is still okay.
About 2-2.5 hours on a charge these days (used to be closer to 3-4 hours), but I do have a spare battery in the fridge.
02:50
If i was in your position i would not fight it, play the easy stuff, which includes hundreds of strategy games (in 3d and 2+d) plus all the games from a few years ago, that are now mostly bug free. Then start building a Desktop one piece at a time.
Plus, these laptops get really heavy to carry. I don't even want to bring my laptop to campus any more unless I really need it.
I built (and upgraded) all my desktops with "cash" which takes more time, but includes doing more research first, what is best performace for the money, what is not a lemon because it has been out for 6mo.
Right now there are probably tons of used parts that are very close to todays tech. sandy bridge is no sluff, one set of tech peoples upgrade for 10% , and end up selling the 90% for 30-50% :-)
One of my classmates is still running a Westmere desktop, ancient by today's standards.
For the millions of computers that have been upgraded to haswell and are upgrading to skylake , there are perfectally usable boards and chips out there from 1-2 years ago that are sold used, sometimes for dimes on the doller.
you could build your own desktop that would at least cool better than a laptop for minimal money, then the GPU can be upgraded, , later.
Great advice. I'm still on very limited funds but hopefully I'll have a real job soon that makes me enough money to move forward.
03:02
hello
How would one know if a program is indeed using a proxy
>_>
check the settings
Anonymous
just changed my thermal paste and heatsink
Anonymous
@BobEbert wireshark maybe?
Unless you're worried its leaking information
Anonymous
saw a 20ºC drop
Anonymous
03:04
cool
and I'm pretty sure that the question has been asked before on the site
@PatoSáinz 0_0
I opened wireshark but idk how to view if its using the proxy
Anonymous
@JourneymanGeek I had the stock intel cooler and dried thermal paste
Would you know?
I've wound up buying accessories like extra monitors and hard drives to augment this existing setup rather than actually building a desktop.
03:05
step 1. Make sure that the program is not using its own proxy.
step 2. Check proxy settings system wide
Anonymous
and the drop in CPU temps also shows in other temps (GPU, HDD, the case itself)
Anonymous
I mean it was smoking hot before: constant 75ºC
Anonymous
55ºC idle
I am using nmap with proxychains4
03:05
0_0
@PatoSáinz I have rediculously good luck with intel stock cooling
Anonymous
so I took liberty to turn a bit up the cpu clock :) (it's still locked tho)
Anonymous
@JourneymanGeek my thermal paste was dry
Anonymous
rock solid
Anonymous
and it's summer here
03:06
@bwDraco I use the 50% rule for just about everything :-) you take the initial price you were thinking something would cost, and add 50% to that , for accessories, wiring, batteries, perifreals, UPS , etc .
that usually takes a few years
lol
Anonymous
@JourneymanGeek bloody rock solid: awful
Anonymous
no more thermal throttling
Anonymous
and awful fans
@bwDraco the 'advantage' with a PC is outside the 'core' system, you don't need very much to start with, and you can upgrade over time
I got the best CPU I could afford, and enough ram for a few VMs. Every other component other than CPU/Motherboard is trivial to upgrade over time
Anonymous
03:10
@JourneymanGeek the new heatsink was a bloody PITA to install
Anonymous
the fact that my mobo is micro-ATX made it even more fiddlier (bloody pins)
almost all stock paste they put on in tossing this stuff together either is clay (intel stock) that doesnt really ever thin out like it should, or dries out like it is the 1980's and they are still using simple themal goops.
I still put all my trust in (real) AS5 , because even though people say it is Ok and normal for the paste to dry out, i believe in the power of the Fat Fryer :-) the deep fried CPU scenario where the oils in the pastes make contact with the molecular hills and valleys, like no metal dust, or metal oxide dusts can.
. . and thermal pads suck, i dont care how well tested they are , mentally they are foam tape with oil in them, an insulator and a barely mobile fluid.
I don't quite find what I am looking for with the documentation @JourneymanGeek
I am sure themal pads are "the best they can do", but thier best sucks :-) it only serves the purpose of sloppy craftsmanship and machine assembly.
@BobEbert well, adjust the search query to taste then
and don't forget, we have a Q&A site to ask questions. This is more of a space to discuss things
@Psycogeek stock CPUs don't use thermal pads
they use thicker less runny thermal paste
Anonymous
03:25
@Psycogeek good enough for simple power transistors
Alright, probably going to bed earlier today. Good night.
And don't forget, the typical design life for a PC was ~5 years at one point
Just that we've hit a point where older PCs good enough for most things
@JourneymanGeek My father's old Dell Inspiron 531 is eight years old and it is still fully operational.
My R61s are still in active use
Swapped out the hard drive for cheap SSDs, and upgraded em to windows 10
I'm rather surprised that nothing has failed on this machine. We've seen some odd boot issues and whatnot and the HDD is showing its age but the machine still works.
Then again, the system sees only light use. Father is spending most of his time on his 2013 Nexus 7 these days.
Most of the time, I'm on my laptop, with some time spent on my tablet or phone.
My laptop is on virtually every single day. The old Dell gets turned on for no more than a few hours at a time once or twice a week.
03:41
funny thing is I have a pile of old hard drives from upgrades
they arn't dead, just slow
Feb 17 at 6:31, by bwDraco
7200 RPM desktop hard drives have hit a level of storage density (and hence performance) that allows them to practically match the VelociRaptor in most workloads.
Feb 17 at 6:29, by bwDraco
http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/71023-western-digi‌​tal-black-6tb-hdd-review.html
6 TB WD Black. Five platters at 1.2 TB each, new controller and more cache.
You'll be surprised at how far hard drives have come.
Sequential I/O speeds are remarkably high even towards the inside of the platters.
Bob
Bob
04:02
Why are you assuming they don't? To quote J. Applebaum, one of the TOR founders: (I'm paraphrasing from memory): "I think these days we have to assume: if the NSA want's be in a given system, they are in." — fgysin 18 hours ago
lol
see mossad quote (@JourneymanGeek)
;p
Quite honestly, if I wanted to keep something secret from the government I'd go back to oldschool books as one time pads, paper, and pop culture references.
Bob
Bob
Pigeons? :P
Anonymous
04:31
> ES6 performance sucks! Strong mode is a mode for ES6, you cannot use it without using various ES6 features. However, idiomatic ES6 code currently is substantially slower than ES5, across all browsers -- easily by 2x, often by 10x or more. Due to the sheer size of ES6, plus a number of unfortunate design choices, it will likely take years until implementations catch up with ES5 optimisations, and the hundreds of man years that went into those.
Anonymous
maybe it's not worth porting @ChatBotJohnCavil yet
Not any more
Have we talked about flagging yet?
Cause I'm JUST about to settle down to doing a run of the flags
Never mind.
Sorry.
I'm off to bed.
05:33
...or how eDRAM enables incredibly high performance.
Instead of relying on a convential memory bus like DDR3, eDRAM lets you use a custom, super-wide memory bus for insane memory bandwidth.
05:45
mornin
weren't you off to bed? ;p
maybe he's doing the "shifting timezone" thing
also looks more like clever use of 'POP' than dram on the same die
In any case, it's some sort of MCM with eDRAM on the same package.
I wouldn't call that package-on-package. That's best described as a multichip module.
POP is when you have one chip stacked over another such that the chip on top completely covers the chip below.
This arrangement is most frequently encountered on mobile devices where the DRAM sits on top of the SoC.
The DRAM on the Sony sensor is basically chips put on the back side of the sensor. It's not POP.
06:49
I have just been restarted! This happens daily automatically, or when my owner restarts me. Ready for commands.
I have just been restarted! This happens daily automatically, or when my owner restarts me. Ready for commands.
I have just been restarted! This happens daily automatically, or when my owner restarts me. Ready for commands.
Bob
Bob
!!learn foo test
@Bob Command foo learned
I have just been restarted! This happens daily automatically, or when my owner restarts me. Ready for commands.
Bob
Bob
!!foo
07:16
I have just been restarted! This happens daily automatically, or when my owner restarts me. Ready for commands.
I have just been restarted! This happens daily automatically, or when my owner restarts me. Ready for commands.
I have just been restarted! This happens daily automatically, or when my owner restarts me. Ready for commands.
Bob
Bob
07:40
!!info
:S
I have just been restarted! This happens daily automatically, or when my owner restarts me. Ready for commands.
Bob
Bob
!!listcommands
@Bob help, afk, awsm, ban, coffee, color, convert, define, die, doge, domain, eval, export, findcommand, forget, forgetseen, github, google, hang, imdb, import, info, inhistory, jquery, learn, listcommands, listen, live, mdn, meme, moustache, mustache, norris, nudge, parse, refresh, spec, stat, stats, tell, timer, todo, unban, undo, unonebox, user, weather, wiki, xkcd, youtube, zalgo, ;p, \/s*, after5, ahh, areyoukidding, bababababat, baroo, beatingbloodoutofarockwithahalberd, bespecific
boberror, bunny, caat, cancer, caniuse, caution, chihuahuaunravelinginsidearollofbubblewrap, clevergirl,
Bob
Bob
Ok. Cavil is tentatively up for now.
There's still some weirdness with phantomjs, login and cookies that needs to be sorted out.
@allquixotic ^
For some reason phantomjs isn't saving cookies. Debug output tells me it's using the cookie file I set, but that's never created and values don't persist.
I'll just call that a PJS bug for now. Gotta go.
Oh, before I forget again
!!export
@Bob Exported to gist, id: 68ff2511a6fc1e24361d viewable at gist.github.com/68ff2511a6fc1e24361d
08:38
!!taytaytay
08:57
!!coffee
@Burgi TypeError: undefined is not a function (evaluating 'Number.isFinite(value)')
it r borked
Bob
Bob
09:30
!!tell Burgi help coffee
@Burgi coffee: Forwards message to coffeescript code-eval
10:27
"I can't believe Windows can break this badly" of the day: Start menu, and any W10 apps will not start.
gf's machine, all she's been using it for is Chrome/iPlayer
How does this even happen
One fix is to create a new user and migrate everything into the new account
But I can't add user accounts without launching the System Settings app, which does not launch
Dec 12 '15 at 17:39, by tereško
windows administrator guide: minor problem - restart; major problem - reinstall
Yup, that's about it
how do I find where a taskbar shortcut is pointing to?
drag it on desktop :D
@tereško If I drag it it just pops up the menu
10:56
So backing up and deleting AppData fixes the problem
 
2 hours later…
12:36
!!taytaytay
YES! THE WORLD HAS BEEN SAVED!
@JourneymanGeek That didn't make much sense. Maybe you meant: ;p
Has anyone used Git LFS yet?
00:00 - 13:0013:00 - 00:00

« first day (2027 days earlier)      last day (3289 days later) »