« first day (468 days earlier)      last day (221 days later) » 
00:00 - 20:0023:00 - 00:00

12:10 AM
I did squats today. I wanna go to bed.
 
I did my leg day as well
 
@PaulWhite v. unwieldy
I'd like a refund of my bounty please
the commentor has convinced me of the error of my ways
 
How is that comment still there
 
I tried flagging it with a "no u" report reason but it wasn't long enough
 
I bet he has a blog called Hibble's Quibbles
 
12:27 AM
@ErikDarling sounds like a good comment
 
If one were the type of person to leave comments
 
"BTW: Our MS Support contract expired and will not be renewed."
looks like he just @ed Gallardy
 
Where is that one?
 
@JoeObbish I don't read all the comments you all do
linky pls
 
@JoeObbish unrealistic, even
 
12:46 AM
Terribly
You should be ashamed
 
Yes it sounds strange to us also but to have it signed off by security we are trying to find an authoritative source to say it is not part of the GUI and it is expected to be managed by the OS. BTW: Our MS Support contract expired and will not be renewed. — Herkimer 9 hours ago
anyone know of an "authoritative source"?
 
An authoritative source dot com
 
> SSMS 2000
LMFAO
Thank god they don't have support anymore, that would be a terrible customer
2
 
One more close vote...
 
1:03 AM
When will SSMS add support for banana smoothies?
I bet ADS has it as a plug-in
 
Guys
GUYS
When will ADS have an SSMS add in?
 
lol
also, can we be real, "our security team requires every application to have its own usage time out lock lock screen"
 
I don't get why you wouldn't just have a generic login timeout
 
like... let the OS do OS things?
that's crazy talk
 
Of 18 seconds
 
1:09 AM
don't you know our security team is the only and bestest ever
 
Install SSMS 18. The regular crashes will serve the same purpose
 
😊
 
hey-yo
 
I wonder how the Security Team copes with Notepad
 
it's a real piece of work, that notepad
calculator too
 
1:13 AM
I heard they're not even encrypted
Probably a Huawei back door
 
The SSMS timeout sounds like a great idea until it goes flat in the middle of modification query.
 
2:15 AM
I did it. I closed that question. Because Paul refused to.
Josh for moderator 2020
2
 
 
1 hour later…
3:29 AM
 
 
3 hours later…
6:32 AM
2
A: XQuery doesn't update XML data

Mikael ErikssonIf you have the XML in table you should use UPDATE instead of SET and there you can put the namespace declaration outside the XML_DML expression using WITH XMLNAMESPACES. with xmlnamespaces(default 'DataService/1.0.0.0') update DataImpTable set serviceData.modify('replace value of (/SMObjInfo/Ce...

 
 
3 hours later…
9:50 AM
And my Office icons changed again. This is weird.
 
10:32 AM
2
A: Chat.se, chat.so and chat.mse seem to be down

Dean WardOne of our ISP links to the NY data center was flapping which meant any traffic in or out of the data center to certain parts of the world was not being routed correctly. For now we've disabled that link and all traffic is being routed over our other ISPs

> flapping
 
Guys, I marked a question as a dupe all by myself. Like NewNess--
 
10:47 AM
@JoshDarnell I could get behind this, what's your platform?
 
Newness for all, I expect
 
11:00 AM
@PaulWhite that answer should be accepted. I upvoted it. I thought about the fact I was using a variable for my test, but knowing almost nothing about XML, I figured it wasn't a big deal.
 
I'm starting to think Mikael knows his way around XML
 
11:33 AM
Updating XML in the database is a strange thing though
Storing, shredding, or generating XML? Sure, fine in the right circumstances. But updating?
 
TL;DR
too much detail
screenshot was awesome tho
 
Great visual aid
 
I often wondered what perfmon would look like if I ran it on an old CRT monitor
Bookmarked for when I have the time spare to dig in as it deserves
 
11:48 AM
That's the new "pink eye" Instagram filter I think
Very hip
If I were the type of person to leave comments I'd say that link only posts are discouraged.
 
ha!
clearly we've been over-thinking this blogging business
 
12:05 PM
We should follow the lead of someone whose business wins awards.
@JoeObbish your first post is quite popular on the twitter
 
I only take awards seriously if they're presented in AOL style
 
You don't take my wine awards seriously?
 
Well that wasn't an online thing
 
12:33 PM
today i will do it
today i will pay quarterly taxes
 
@ErikDarling it is?
 
12:47 PM
I don't see it
is it a Great Post, Brent?
 
Great post!!👏🏻👏🏻 https://twitter.com/erikdarlingdata/status/1128151051587158016
 
any chance I can get some "love" for this idea? feedback.azure.com/forums/908035-sql-server/suggestions/…
> The resource you are looking for has been removed, had its name changed, or is temporarily unavailable.
 
next thing you'll be asking for stats usage counts
they'll have to differentiate between loaded and used
what a mess
 
@JoeObbish There should be a nice big photo of you on your posts.
@ErikDarling yah, I saw a few of those suggestions. Seems harder than you'd think at first glance.
 
@JoshDarnell May I suggest a minor edit? You describe PAGEIOLATCH waits as "latch waits" but it's probably easier for the OP to understand them as "IO waits"
 
12:57 PM
just make the edit
he's more new than you
PULL RANK
 
But then what's to stop me from going mad with power?
 
@Forrest Sure! Good call.
 
@Forrest paul
lack of a dragon
idk
 
@ErikDarling I didn't realize Forrest had edit privileges.
=P
Wanna add a link to sqlskills.com/help/waits/pageiolatch_sh while you're in the ninja window?
 
will do
 
1:11 PM
@JoeObbish I like your custom type easter egg
 
1:22 PM
this is very coool, @joe:
> If you’re wondering about the temporary stored procedure creation, it’s a habit that I’ve gotten into whenever I write a T-SQL while loop. While not applicable here, very fast loops can incur ASYNC_NETWORK_IO overhead due to the passing of DONE tokens to the client.
also, a small typo:
> roughly check if you are running of IDs
appears to be missing "out", as in :
> roughly check if you are running out of IDs
 
joe doesn't make typos
he adds easter eggs for dummies to fixate on
 
it was a good one
 
1:38 PM
Are table (heap) scans always allocation order?
2
 
hmm, I actually don't even know if there's a linked list to traverse
I'll DBCC PAGE and find out
It appears that there is no page linking, so I would presume allocation order is the only way.
 
Sweet.
 
1:58 PM
@Forrest Sean is like Donald Trump
he only reads blog posts if you put his name in them
@ErikDarling you editor, where you at?
 
working on becoming a world class sql server educator
also getting ready to get my passport replaced
the TSA owes me $200 i think
 
2:17 PM
Context for my allocation order question: sqlgene.com/2019/05/14/t-sql-tuesday-114-an-unsolved-sql-puzzle
I left a comment.
This World of Ours - that guy is a genius writer.
> If the Mossad wants your data, they’re going to use a drone to replace your cellphone with a piece of uranium that’s shaped like a cellphone, and when you die of tumors filled with tumors, they’re going to hold a press conference and say “It wasn’t us” as they wear t-shirts that say “IT WAS DEFINITELY US,”...
Security through futility.
 
"As part of that, I like to show that if you pull a single row from a heap, it has to read everything."
Ummm...
 
Heh, it's not phrased really well. If you look at the actual query, he's talking about reading a single row by ID without an index - meaning reading the whole table with a residual predicate being applied to each row.
 
2:33 PM
isn't that exactly the same as a clustered index though?
 
Yeah, if there's a unique value in a column that's not the clustered index. I guess his point applies to querying any unindexed column.
Maybe he has left out some context from his actual talk.
 
@Forrest lies and slander
 
3:01 PM
We are engaging in a new, efficient git branch naming strategy where the feature branches are named according to the story id in Azure DevOps.
"Hey Josh, will you take a look at the feature X I'm working on?"
Super helpful.
Now I have to find feature X in the sprint, and click on it to find the ID.
Could we at least name them "feature/6540-feature-X"?
No, too much typing.
I know this room isn't the right target audience for my rant.
 
Sorry, I don't quite get it
Something about a missing index?
 
so are any of you jokers going to show up on skype tomorrow?
 
I had something else planned :(
Otherwise I totally would
been thinking of skipping out on my plans...
 
3:17 PM
is tomorrow wednesday?
 
@ErikDarling yes
hth
 
i think i'm supposed to take my kid to a movie
if not i'll be there
can i use something other than skype?
 
you could use an airport
 
will there be tacos?
 
there's pizza from a hole in the wall type place
I doubt it's up to your snooty NYC standards
 
3:21 PM
...
 
@JoeObbish I was hoping to, but I'm going to see Endgame instead.
 
it's a real cornucopia of excuses in here
 
we had to find out THROUGH TWITTER
😒
 
you should be happy that twitter was useful for something
not that typing out clapping emojis isn't useful, of course
 
twitter: 1
joe: 0
 
3:26 PM
erik won't even let me see how many views my Great Posts are getting
he is a cruel wordpress administrator
 
Sounds like a real tyrant.
 
i have no idea how to do that
 
jk I don't care
 
uncaring blogger
i should have sent you to dcac
this person seems helpful
 
3:45 PM
Are you sure they're not just trying to trick you?
 
everyone is
especially joe
 
@Nick_Craver Hey Nick, thanks for reaching out. It's because our developers are clueless and have no idea what they're doing. But on a positive note, it's a great sign that our infosec is terrible and we'd probably be easy to hack. Thanks for being a valued customer, have a great day! /FM
Sounds like someone on their social media team needs an extra cup of coffee or something.
 
looks like a joke account to me
M zero N O
 
context?
 
Oh haha it is fake.
 
3:55 PM
look at how gullible the new guy is!
 
Hey @monoprice, Why can't I have a * or a # in my password?
 
I bet OPs trick him all the time
 
@JoeObbish They really do.
 
4:09 PM
I'm glad I've gotten devs to start asking for advice before submitting stored procedures. The downside is I'm now staring at one with 4 temp tables, a cross join, and a full outer join.
 
get the strap
 
all of those sound fine to me
 
at least they didn't just chain together 8 ctes
 
same dev suggested sp_getapplock earlier, when his proc wasn't working right with simultaneous executions
 
i like this dev
 
4:15 PM
sounds like a problem solver
 
Great at Googling
 
@Forrest They probably came across Erik's blog post.
ahem
Had to clean up my implicit biases there.
 
I said "his proc" so you picked up correctly
 
Oh good.
 
i'd hate to see you go into your first sql saturday with that kind of stain on your reputation
 
4:25 PM
It would certainly make things tough.
 
4:36 PM
so imagine this
New intern starts today. Manager asks: "Have you ever used Windows?" Intern: "No, never" #GetOffMyLawn
an intern got hired at microsoft and no one asked them if they've used windows before the day they started
that's... that's some process
 
Top Linux Sentient Entity
 
ahem
 
implicit bias removed
 
Top GNU/Linux People
I'd think linux people wouldn't want to work at microsoft
probably uses macs
 
I'd just like to interject for a moment. What you're referring to as Linux, is in fact, GNU/Linux, or as I've recently taken to calling it, GNU plus Linux. Linux is not an operating system unto itself, but rather another free component of a fully functioning GNU system made useful by the GNU corelibs, shell utilities and vital system components comprising a full OS as defined by POSIX.
4
 
4:41 PM
This guy.
 
FINALLY SOMEONE SAID IT
 
🤭
 
@ErikDarling I didn't know @Zane started already.
 
🔥🔥🔥🔥🔥
 
I bet Zane uses a Sun workstation at home
 
4:48 PM
this is coming from a guy who owns a jitterbug
 
I don't even know what that is
 
more joe trickery
 
I'm the only OP that doesn't try to trick you
 
@ErikDarling lol
Those commercials...
 
the greatest trick op ever played...
 
4:50 PM
I don't watch commercials
 
was convincing the world they weren't trying to trick them
 
I use adblock on my GNU/Linux personal computing device
wait, is adblock free software?
 
> not reading the eula
it's like you want to give up your rights and privacy
 
5:02 PM
I feel deep shame
 
I can't believe that didn't get comments
@ErikDarling you aware of any fixes?
 
for which part?
 
any of it
 
5:08 PM
kind of seems like a disaster
 
hasn't changed afaik
yep
LIGHTWEIGHT BABY
 
I was hyped for this
 
i'm surprised you're not totally jaded by now
 
 
1 hour later…
6:19 PM
@JoshDarnell lol
@JoeObbish I have a PC that I use for most stuff and a poopy old laptop that barely works for my xcel stuff.
 
@JoeObbish it's your dream come true!
> Intel Reveals New Spectre-Like Attack, Advises Disabling Hyper-Threading
 
my main issue with HT is that it makes observability so much more difficult
if it makes the workload faster, great
 
7:17 PM
Am I the only one who can't understand what the 2nd sentence actually means here
> When you are using READ UNCOMMITTED, if your query reads rows using an allocation order scan (using IAM pages), you might miss rows if another transaction is causing a page split. This cannot occur when you are using read committed because a table lock is held during a page split and does not happen if the table does not have a clustered index, because updates do not cause page splits.
There are just one or two many double negatives in there.
This seems blatantly false, though:
> does not happen if the table does not have a clustered index, because updates do not cause page splits.
if a table has a clustered index, it still has page splits on inserts, even if the row being inserted is at the end of the index.
 
@MaxVernon That seems to be saying that page splits don't happen in heaps.
"does not happen if the table is a heap (aka does not have a clustered index)"
 
@JoshDarnell that's kinda what I thought, but like, it's wordy.
 
It is a hard to follow for sure.
Sounds like PR time.
Go Max!
=P
 
lol, I was reading the page to try to get a better understanding of when exactly a page split can lock the table. That didn't really help me at all.
I need source code access, dammit.
although one certainly thinks having source code access would still be a pretty big hill to climb.
@JoshDarnell and page splits certainly don't happen in heaps... there is nothing to "split". Inserts are just written in the order they occur.
I thought I had a pretty good command of the language, but that just doesn't make sense to me.
 
@MaxVernon I know =)
 
7:43 PM
5 hours ago, by Josh Darnell
Context for my allocation order question: https://www.sqlgene.com/2019/05/14/t-sql-tuesday-114-an-unsolved-sql-puzzle/
 
you're both fired
 
What a relief.
 
00:00 - 20:0023:00 - 00:00

« first day (468 days earlier)      last day (221 days later) »