« first day (2826 days earlier)      last day (2487 days later) » 
00:00 - 17:0017:00 - 00:00

00:00
We don't really need extra ROs. Heck if things worked perfectly we don't need ROs at all outside clever channel topics.
This room may be fine, but I've seen rooms that are much worse.
You do realise ROness is room only right? ;p
I do.
Just saying I'm available in case you need it.
00:25
@JourneymanGeek Yes. Me.
@FMLCat on the other hand, we can deal with you ;)
Only when you're around
Also I think telling @bwDraco to deal with you would be cruel and excessive.
When you're not... mwahahahaha evil laugh
I'm not sure to who
@FMLCat I work nights sometimes. You will never know when I'm lurking.
00:26
And yet I somehow always manage to stir up shit while you're away
hah
You do realise the entire moderation framework here was designed to deal with one persistant user right?
(and its not you)
Not yet anyway
Also, I actually get what makes you tick. And by making sure that you get what you really want, by being productive....
You know what makes me tick? Cats.
we can totally XKCD 810 the shit out of you.
!!cats
00:28
@JourneymanGeek That didn't make much sense. Maybe you meant: cat, ncats
!!xkcd 810
!!XKCD 810
00:29
rip
Lunch time.
00:48
@JourneymanGeek @FMLCat is chaotic neutral and @bwDraco is Lawful lawful
Bob
Bob
hi
So - I'd like to point out
that positional power means very little here.
Even if you arn't an RO, and you feel you can defuse a situation, go ahead.
If you arn't confident, you can assist any external mod who comes in with context.
@ThatBrazilianGuy ha
We're a civil place. We don't need wandering cops in full body armour to keep the peace.
00:54
@JourneymanGeek Images don't load for me. :|
Bob
Bob
I think I might've used RO powers ... like twice? And that's just moving messages to another room. Because that's all I can do.
@MichaelFrank probably slashdotted
@Bob and kick
Bob
Bob
Sure, kick-mute exists. And there's almost never a good reason to use it.
@JourneymanGeek Most people respond well enough to a verbal warning. And kicking without said warning/notice seems kinda pointless - it doesn't tell them what they're doing wrong.
A persistent offender can't be dealt with via kick-mute, since it's so short...
@Bob ... that's literally the main point I was trying to make ;)
Bob
Bob
@JourneymanGeek What, that RO tools are useless? :P
Really, all I want/need is a delete button.
00:57
;p
Not really likely to happen.
@Bob I've been RO for two? three? years now and I've used RO powers to moderate a person / defuse a situation once.
@Bob well and that anyone here has the same ability to regulate chat
Bob
Bob
@JourneymanGeek I've not actually seen any consideration of it, or reasoning not to.
I hardly if ever use my mod powers too
Bob
Bob
They've just stuck their fingers in their ears and ignored the requests.
00:58
@Bob there's this meta post...
lemme go dig it up
hm
can't find it
01:13
@JourneymanGeek And I've grown up now. I'm not going to start modding anyone or raising flags unless it genuinely gets serious.
So I just went through the whole twitter thing... That sounded like a 'fun' conversation... :|
ya
@MichaelFrank all of them were
"fun"
01:31
So chat transcript doesn't highlight replies and point out who is an RO or mod, which is sorta annoying.
it dosen't highlight who's a mod? 0_0
I don't think so.
ah guess not
Yea, technically I imagine it's just pulling from the DB and doesn't do the smarts on the UserIDs that chat proper does.
or generating a static page
cause I suspect transcripts by design are probably designed to be very light
01:44
As a transcript should be really. Although I think being able to identify authority figures in a transcript is probably a good idea as well.
I've been of the opinion it typically dosen't matter
I'd hope, 99.99% of the time, the fact that I'm a mod dosen't affect how people interact with me ;p
Moderation is a team sport, etc. I suppose that's a good point.
Well, more that running a healthy community is a process that needs to involve everyone, not just the mods
active moderation is a very small part of that needed when things go wrong
And conversations like this do more than a hundred kicks or suspensions ;)
 
1 hour later…
Bob
Bob
02:59
@JourneymanGeek Moderation in moderation? :P
man, why do all the interesting conversations about chat get discussed while I'm sleeping or working
are we being specifically warned for our stuff?
Naw
More talking about why we didn't need an RO top up
And a broader theory of how we do things
And various general network drama
maybe Ben Gruenbaum's chat QoL suggestions will be implemented sometime in the mid 2020s and this will be a moot point anyway
Ah ya.
That's something so Corp needs to work on.
it seems like they focus a lot on obsessively high quality features whenever they do anything, which means when we get new stuff it generally scales well, performs well, but takes forever and a day to come for fairly small changes
03:06
@allquixotic also the focus has been elsewhere the past few years
(careers, documentation... currently enterprise and channels teams)
And a result of that was the folks who're supposed to be looking after us feel... kinda cut to the bone.
(this is something I've brought up a few times)
losing an entire office full of sales people (no matter how important they are) affects us less than losing 4-5 of what, a dozen CMs?
Stack Exchange has been looking for ways to monetize the platform without compromising its integrity.
@bwDraco True but in a sense - its assumed that the sites themselves would run on their own, and you could scale up community indefinately.
But you can only have so many full-fledged moderators. Too many diamond mods and the community may no longer be able to function in a democratic fashion.
Random question, but has anyone ever experienced IE11 just stop downloading any files? Like, it doesn't even shows the Open, Save, etc popup?
03:19
You're assuming moderation is action
Moderation is awareness
A key idea behind Stack Exchange is that's designed to give power back to individual users.
I rely on the folks around me to tell me what I'm missing.
Our CMs are busy. I donno what C level does
Hence, it's designed to be community-driven as much as possible.
hah
waaaait
I kinda assume I know how the system here works? ;p
And yes it is!
but we all have blind spots.
and pesky emotions.
Could someone clarify why this edit was rejected? superuser.com/review/suggested-edits/752684
03:23
ehh
Was a judgement call?
I'd have approved it, but its kind of one of those things?
This is a hard call. I would likely have approved it, but it's a major change that can appear to deviate from the original intent of the answer.
An alternative would be to post your own answer, crediting the selected one, with that information.
and include a little more.
Generally, I approve edits that add meaningful information, as long as they do not stray excessive from the post as currently written.
(and keep jwh's one as the selected one)
The answer as it is, does not answer my question (i.e. how do I do X). Instead the answer links to a bug report with a solution. I've just imported the solution to the answer.
03:26
commented
So write your own answer and accept it ;)
or that ;p
NOTHING wrong with that
I've posted my own answer, but rejecting an edit that brings a non-answer into compliance is a waste of time for everyone involved. :/
03:44
@BardiHarborow agreed, but my next follow-up is, well, if they're going to have a non-answer, flag it as a non-answer and it'll get deleted, then post your own
having crap content on the site isn't an acceptable alternative just because people have such irrational protectivism over their "own content" (IMO we should be more like Wikipedia so I apologize for my fairly disdainful tone about rejected edits)
Bob
Bob
1
Q: PowerShell ISE runas user reports other user with whoami

user1330734I am looking to start a PowerShell ISE sessions as a domain user. runas.exe /netonly /user:domain.local\username "powershell ise" I am prompted for the password and the ISE environment launches. However, running whoami gives the following results: win-machine\user which is the local machine ...

^ This was an interesting one
Didn't know about the /netonly switch but it seems useful... can keep your local profile (settings, etc.) while impersonating another user remotely.
@BardiHarborow I'm going to add a little to your answer off the comment that seems handy
I feel free to rollback if its irrelevant
@allquixotic The edit was rejected by the queue, not the answerer. Your point is valid though.
Bob
Bob
04:10
@bwDraco gonna install 1803 on the test system
hm... it's gonna be hard to time remotely
What are your specs?
Azure machine, I suspect?
Bob
Bob
@bwDraco VM with 4 cores of a Xeon D-1531 and 4 GB RAM allocated
Hmm...
Bob
Bob
OS drive on ZFS on striped SSDs
I really need to play with this stuff more.
Bob
Bob
04:12
@bwDraco KVM
ZFS, servers, and whatnot.
Bob
Bob
tbh I haven't really touched it in a while
I mean, I use it now and then for testing
What is it like to do ZFS in a virtualized environment? Do I want bare metal for this purpose?
Bob
Bob
but I haven't touched the VM config
@bwDraco If you want to play with ZFS, it doesn't really matter
I hate putting the Demon's capabilities to waste.
Bob
Bob
04:13
If you want to actually use it, it really should have raw access to disks
For play, learning, etc., virtual disks work
This desktop was designed with virtualization in mind. I just need to find the motivation to do it.
Bob
Bob
lol
it's been a while since I last ran a local VM too
ok, reboot started
this could take a while
...maybe I should've closed some programs
ok took 2 mins to get from restarting to configuring updates
Renovations on the house are finally resuming so I might just be able to set up a separate machine for virtualization/home lab stuff.
Bob
Bob
took 25s to actually reboot
another 12s to get to the boot time working on updates screen
:44346511 ehhhhh. I find it more economical to rent a server
had this discussion with @allquixotic before too
Well... it would be great if VirtualBox had proper nested virtualization support. Not sure if the improved AMD-V features in Zen allow for better performance running nested VMs, but it would really be useful for more ambitious projects.
Bob
Bob
04:22
@bwDraco VBox is pretty iffy
Things have improved greatly with the latest versions.
Perhaps a good motivator would be to tell myself, "we spent $4K on Astaroth for a reason, take advantage of that investment to learn skills for your career!"
And with my mother starting to give me more agency to do things on my own, I should no longer feel stuck in a rut.
(e.g. I'm now allowed to make more significant financial decisions on my own, without clearing her)
Bob
Bob
total time so far 14 mins, 7% (but it's speeding up)
I think it's CPU-limited at the moment
could be I/O though
Xeon-D doesn't have the greatest single-thread perf
It was mostly compute-bound on both systems, but having an NVMe SSD works wonders even when it's not obviously I/O-bound, since the NVMe protocol inherently requires less compute than SATA AHCI.
Which is why I got 10m 45s on the desktop. (POST is oddly long on this motherboard FWIW.)
Bob
Bob
@bwDraco Yea, uh... I'm at a disadvantage here since ZFS is relatively expensive
vs raw access anyway :P
then again, the ZFS stuff should be happening on the other cores
also I'm still thinking that watching it via remote console is slowing it down. idk.
also I've never bothered to optimise this VM
could probably optimise the disk emulation some
04:47
Windows 10 version 1803 appears to have greatly improved the memory compression feature. I'm seeing less paging under memory pressure and more memory saved from compression as reported by Task Manager.
This is a huge deal on a netbook with 2 GB of RAM.
Bob
Bob
25% at 35 mins
I'm gonna be out for a bit so I'll check when I get back
(Disk 1 is a portable SSD dedicated to PrimoCache)
It's getting late here so I can't really stay online much longer.
Then go to bed ;p
05:09
@Bob my main beef is even fiber to the home has an uptime around 95-97% instead of the 99.99% or better that you get in a DC... they install it so cheaply and without caring about the quality of their workmanship and you get line issues and reliability issues, and the hardware is cheap buggy crap, it's just faster ADSL basically
I'd pay a $1000 setup fee just to have enterprise-grade installation that will actually work reliably for a decade at least
these residential ISPs seem to take a special pleasure in saving every last quarter of a penny to figure out the best way to deliver a service that'll work fine on Day 1 of service without regard for how many support calls it's going to cost later down the line when the shoddy service just randomly cuts out or gets misconfigured on the service end, etc.
so yeah, dedi at home? !!no
ow
My line's reliable
my ISP is sometimes odd
05:29
@allquixotic WTF? Availability on my HFC cable Internet connection is closer to 99.9%.
95-97% means you're getting an entire day of downtime every month.
Bob
Bob
@bwDraco install finished ... at some point within 1h 15m
idk when
That took a while.
Bob
Bob
@bwDraco well, any time from 35 min to 1hr 15 min
I'm trying to log in to check when exactly
Hmm...
Bob
Bob
but rdp seems to be down... hmm.
and I'm in!
@bwDraco event log implies it took about 40 mins
05:38
That's about how long my netbook took.
nn
Bob
Bob
also apparently I only gave this machine 2 cores
not 4
whoops
@allquixotic NTLM:
> Microsoft Windows Server has detected that NTLM authentication is presently being used between clients and this server. This event occurs once per boot of the server on the first time a client uses NTLM with this server.

NTLM is a weaker authentication mechanism. Please check:

Which applications are using NTLM authentication?
Are there configuration issues preventing the use of stronger authentication such as Kerberos authentication?
If NTLM must be supported, is Extended Protection configured?
which looks very useful
@bwDraco that's pretty accurate, though it's usually in smaller chunks spread out
Bob
Bob
06:17
> it is possible to purchase editing rights to the Aadhaar database for a mere 2,000 rupees
O_O
06:44
@allquixotic, funnily enough, I'm actually an active editor over there, hence why I edited the answer instead of content forking it. :P
"content forking"
Thats a term I may need to steal in future
Bob
Bob
I'm a simple man: I see Alconja, I upvote. — IAmInPLS Sep 27 '16 at 13:49
 
1 hour later…
08:24
yey
Coventry is over the limit for bad air
Along with Manchester @Burgi, Liverpool @DavidPostill...
well as manchester is the defacto second city these days that isn't surprising
 
1 hour later…
09:53
yah
need moar wind.
10:44
le sigh
Hello
I hate lawyers
Good morning!
@allquixotic We call that "capitalism 101"
Oh sorry can't speak badly of capitalism. Ok, it's "business".
11:03
bleh... anyone a god with systemd services?
@rahuldottech Ah, I believe you're coming around to the Smiley PoV.
Welcome, to your new understanding.
21 hours ago, by djsmiley2k
oh lord. I hate the planet.
Dec 6 '17 at 13:56, by djsmiley2k
I mean I hate other people, but... sometimes it's good to talk
May 19 '17 at 19:14, by djsmiley2k
sometimes i hate you
Aug 29 '16 at 16:31, by djsmiley2k
I hate this place :/
My hate is strong.
11:31
okay, so not so much a systemd issue and more of a classic permissions issue. phew
@MichaelFrank I was about to say, I wrote one. Once.
@MichaelFrank you fixed it?
@djsmiley2k Yea
\o/
I looked at systemd
tried to debug a few things on centos boxes
went 'blarugh wtf is this' and never touched it again
journalctl is the key
11:40
so even tho my linux skills are high
yeah, was using that
it was like 'service started; service stopped'
yeah, thanks for that journald.
systemd is low
I know very little about Linux itself, but digitalocean has done a decent job of guiding me through most of what I've been trying to do.
We're all gonna DIE!.
> Please find attached a Flyer from Coventry City Council regarding a World Fusion Day on the 30/05/18. Its at the Canal Basin 10am - 3pm.
World Fusion day?!
Kinda sucks too
may be totally unrelated, but now it's gonna completely cast a shadow on any kind of diy bio work
k, hmmm, bored
@ Work D:
(on lunch tho)
k I need a website
(i don't, but hell, why not have one!?)
12:23
@djsmiley2k Egh, ethically grey area already
"You're dying, so what do you have to lose by taking this experimental drug?"
Not cool
I mean, if it's properly set up and investigated, grand
But to say "here, inject this and we'll see what happens" is irresponsible
Hang on a sec, need to cash my "Big Pharma $$$" cheque
@bertieb lol hehe
yeah, it's a weird one
and now he's dead, just gonna make a mess of the people doing good stuff :/
Like prosthesis is amazing. the 3d printed stuff they're making is outstanding.
12:38
Aye
It's a heckuva time to be alive
Bob
Bob
@DavidPostill idk if you want to redact your product id
@Bob Good point. Thanks.
And HNQ :)
13:01
Damnable HNQ on a question that is a dupe of several dupes.
And a subjective one at that. "Simplest" is not a definitive definition people.
monthly grumble about lack of decent (at least non-crappy) free send-SMS-from-computer apps on android
(Airdroid is overkill)
Bob
Bob
@Mokubai I felt like it was probably a dupe but couldn't actually find any :S
Most potential dupe targets are specifically asking for CLI answers
13:18
@Bob I saw a couple and put them on the list... for completeness and to suit the many different definitions of "simple"
"simple" depends far too much on end user skill and in some cases "press two key at the same time" could be beyond the end user. Sometimes the simplest method could be "check the box the computer came in"
Bob
Bob
@Mokubai Eh, I think since the OP explicitly called out keyboard shortcuts, that's clear enough for answers.
I didn't find a few of those in your list :\
The CLI ones are right out - they are generally not accessible to 'users' in that sense.
The Vista one is good though
hm... oh. I was searching with "check"
gah, should've just used google
13:38
IIRC my search was literally "Windows 32 or 64" to find a whole bunch of dupes. Granted there's a lot of the kind of questions he didn't want... but there were enough of the ones he did.
Bob
Bob
@Mokubai ha, my cursory search didn't find the non-hardware non-cli ones
that was the 10 min delay between comment and answer :P
I wasn't meaning to imply anything about bad searching, the search tool is fickle and a bit quirky at the best of times. I just get vaguely miffed by HNQs repeating old questions.
Bob
Bob
@Mokubai yup, I've been on both ends of that
HNQ annoys me more than it should.
I've had more than my (un)fair share of them, but they're hollow. The votes are meaningless.
@Mokubai I'm happy because I'm one step closer to [Legendary] ;)
13:44
Attributable more to luck than knowledge. Right place at the right time.
@DavidPostill which one is that?
Bob
Bob
@Mokubai Depends, really.
Some HNQs are like that. Totally luck, and an easy popular question.
@Mokubai Earn 200 daily reputation 150 times
Bob
Bob
Others HNQ because you've written a great answer.
Those are nice :)
Only 2 other people have it.
That they be.
@DavidPostill I still need to get [peer pressure]...
Bob
Bob
13:48
lol, same
short of intentionally flubbing an answer, I'm not entirely sure how :P
Yeah. And that would be disingenuous.
@Mokubai Yeah, I don't have that. Or Illuminator (only 1 awarded).
Hi. Can someone help me understand how my computer motherboard dictates maximum ram size...?
@deostroll For once there is a nice answer on quora. What component determines the amount of maximum RAM on a motherboard? - Quora
31
A: What limits a motherboard from supporting more memory?

MokubaiIn short, this is a mixture of limitations. One is in how many "address lines" the CPU itself supports (different from the 64/32-bit address space/register size), 32 lines would allow 4GiB of RAM, having a 33rd line connected would allow 8GiB of ram and so on. The other problem is in how many...

I thought the 32bit cpu bus, was exactly why the 4Gib limit existed
and PAT? got around that using dirty hax
14:06
@djsmiley2k depends on the CPU implementation. Entirely possible to have a 32-bit internal bus fetching data from a 64-bit external bus, you'd just need special instructions that know how to access it. It's not entirely practical and is a lot simpler for programmers if you don't though...
Good old von Neumann bottleneck.
@Mokubai close it? ;p
HNQ is vaguely broken
@JourneymanGeek already did :)
> The other obvious solution [...] is moving memory into the same package as the processor.
@bwDraco the "obvious" solution if you're an idiot...
14:14
This is exactly what Intel tried with eDRAM. It's a shame they haven't done it again.
Jul 30 '17 at 2:23, by bwDraco
The eDRAM L4$ is capable of producing a massive improvement in IPC. Why in Oblivion are processor designers not doing this???
The modularity of CPU and memory is exactly what allows us to tailor computers to suit specific needs. We already have memory "on CPU" and putting more of it just makes it more expensive, not less.
The way I see it, you can't really remove the external memory interface... but you can put more cache onto the processor package. Intel has done this before. It makes a huge difference.
It can, but that only gets you so far.
It saddens me to see that Intel has not deployed this technology on any new desktop processors. They still have it for mobile chips, primarily to boost processor graphics performance, but a small group of people fortunate enough to have an i7-5775C know just how underrated the eDRAM is.
(and sadly, that does not include me)
Intel is just leaving performance on the table.
You still end up with yet another damn layer of cache that you have to manage and protect against hardware bugs
Cache is difficult. And annoying from a hardware perspective.
14:22
Also, AMD has recognized the value of fast memory and officially supports up to DDR4-2933 on their latest chips. Intel should follow suit.
Normal RAM you can let the OS manage. And you don't want normal RAM on the CPU package.
Anyhoo. /out.
 
1 hour later…
15:44
@Mokubai Well done for finding the dupes :)
salutes
We've long surpassed the point of diminishing returns on cache size and RAM speed. Increasing both makes no difference to pretty much all CPU tasks.
The reason they put eDRAM and the such on integrated graphics chips is because graphics is the only place where it actually helps
@FMLCat there are some benchmarks that show otherwise - if you're very regularly dealing with a common set of data whose size is greater than your L3 but not more than slightly larger than your L4, it'll be faster than going to system memory all the time
16:00
@allquixotic A few, very specific benchmarks and/or workloads, yes, which is why I said "pretty much all"
More typical benchmarks show practically no difference
Here they tested Cinebench with no eDRAM, then 128MB of eDRAM clocked all the way from 1400Mhz to 2200Mhz.
Then Anandtech tested performance with a bunch of applications with DDR4 memory all the way from 2400 to 3067MT/s
The biggest difference they got going from 2400 to 3067 was around 1%, whereas close to half the benchmarks saw RAM with tighter timings at 2400 outperform the 3067 stuff
Also note the laptop with DDR4-2133 and no eDRAM outperforming a desktop with the same CPU with 128MB eDRAM and DDR2800
Yeah, that's just one benchmark, but it's hard to find direct comparisons across a ton of benchmarks. The closest you get is them testing a 6770HQ NUC in one review and a 6700HQ laptop in another
16:49
@FMLCat NUCs are getting special treatment somehow, I think. What I mean is they're currently the only hardware platform receiving Vega M GH (though it may eventually find use in cheap HP/Dell AIOs in 2 years as a way to use up their stock)
@allquixotic Doesn't Apple like to chew up nearly all of Intel's high-end iGPU parts?
@FMLCat annoyingly they take the middle tier of them (presumably to save power/space, which I like, but still, I want it all :D)
like Skylake 13" MBP gets Intel Iris but not Iris Pro
Kaby Lake gets Iris Plus but still no Iris Pro, no 48 EUs
to be precise they seem to be taking GT3 chips
not GT4e
so if their power budget only supports GT3 I highly doubt they'd take M GH. Maybe in the 15" MBP, but I dunno what Vega M GH perf is like compared to the Radeon Pro 560 dGPU already in those
00:00 - 17:0017:00 - 00:00

« first day (2826 days earlier)      last day (2487 days later) »