Hello room

@JackDouglas you actually did it I see.

@swasheck The more you know...

@swasheck I don't think that necessarly applies here.

This is on an index. The index is on two columns. RowStatus then LoanMasterID. Row status has a 4 possibly statuses 1,2,4, and 255. 1,2,4 are all different types of expired/deleted rows where as 255 is active.

LoanMasterID is just as it sound the unique ID for any given loan. This table has 153901294 rows in it so when you do DBCC_ShowStatistics for this column you get something that looks a little bit exactly like this.

Sorry it was 1,2,5

In this scenario there are only for 4 rows because that is a 100% of the posibilities to sample.

@MartinSmith This is what he's looking at running -- 599 GBP for two days. pastebin.com/F9DP28Av

It's currently over 95 F in central London. It's the hottest day in July here on record.

Bummer. It's unseasonably cool here.

I'm digging it.

@ypercube I did, though it's not gone yet, which I presume is because it's slightly too awesome :)

Time to spill some coffee on this pile o junk and get a new laptop.

@Zane well that's new information

Is anyone else failing to get emails from Microsoft when you request a CU?

E.g. I filled out the form at support.microsoft.com/en-us/hotfix/… with a few different addresses and nothing has arrived.

Not sure but ill be happy to check now

Now I'm getting "the system is unavailable" messages

Oh well. I didn't really want to install CU1 tonight anyhow.

Yeah, im getting unavailable

Thanks for confirming

no problem

@swasheck Indeed.

still in developer mode?

Hi @Zane. Sorry for not replying yesterday.

I don't think I could help much with the histograms anyway.

hi heapers

Bad enough when you're showing someone how bad all their code is and they aren't taking any adivice.

It's even worse when they are smacking hteir lunch in your ear.

@bluefeet Hello!

@ypercube It's cool once I realized it was a cardinality issue it was in my ballpark.

@Zane I usually resort to proving my point, rewrite it, show them the stats before and after. That usually shuts them up

@KrisGruttemeyer I'm getting to a point where I can't rewrite every shit query that comes across my door step. I'd never leave the office.

@bluefeet How was your weekend?

@Zane it was good, yours?

@bluefeet Cleaned the kitchen!

Very exciting stuff.

sounds like tons 'o fun

My puppy has turned me into an old man quite quickly.

it happens

@billinkc you here?

I might need to pick someone's brain on something in the near future

What kind of something.

@Zane At least you see them. These devs here just deploy it then I'm left to deal with the aftermath the next morning

It's taken me 1 year to even get a list of what they are actually changing before the deploy

@Zane I'm trying in SQL to figure out how fast (the velocity) of something is. Example, how quickly comments are added to a question

and my sql skills are failing me

Alright, can someone sanity check me? Go into an already created SP, Script -> ALTER TO -> New Query Window. Do you get extra blank lines above the ALTER PROCEDURE and below END?

Then, try just hitting Modify

@bluefeet Oh so it won't be me then. I'm more of an execution plan kinda guy.

@bluefeet hi!

@swasheck have you ever put both 2012 and 2014 in DAG (for a migration)?

@Zane estoy aqui

Oh I was just going to make a comment about when you open a SSIS project and it hides a message behind all of your other windows and you just sit there like a moron waiting for it to load.

I usually sit here like a moron until the day's over

@billinkc ???? you still just sit there like a moron

@JamesLupolt nope

@swasheck As long as the checks clear, I'm content for the drool to pool under me

@billinkc no doubt. works for me

@swasheck Me neither

once you failover to 2014 you're stuck

i'd surmise

@swasheck yep, that's how it worked with mirroring and is what i'd expect with a DAL

@bluefeet Sounds like a fun query. Need to get back to work-work but LEAD would help figure out when the next comment came in. Not quite velocity, but this shows any questions that had more than 10 comments in a 6 minute period for the last week data.stackexchange.com/stackoverflow/query/edit/331089

@billinkc <3 I've been playing around with different things trying to figure it out. This is great

Do you have to work against data.se or do they give you direct access to a sql instance?

a version of data.se - no direct access to a server

wimps

I've been trying to do something similar without LEAD/LAG

@bluefeet glad to see you still get to write some code.

@Zane a bunch of sql'ing still. Gotta research impacts on changing things.

I did some user retention stuff my first week - cohort analysis which I'd never done before

first you guys have to fix data.se

fix what?

If something is wrong, then post a request - meta.stackexchange.com/questions/ask

:)

I'll fix you, @swasheck. And your little dog too.

@bluefeet need more requirements

select
    p.id,
    comments_per_second = 1.*count_big(1) / datediff(second,p.creationdate,sysdatetime()),
    comments_per_second2 = 1.*count_big(1) / datediff(second,max(c.creationdate),sysdatetime())
from posts p
join comments c
  on p.id = c.postid
where p.posttypeid = 1
group by p.id, p.creationdate

@swasheck interesting as well

thx

@bluefeet lag and lead would get you acceleration, that is, the rate at which the velocity is changing

I'll probably wind up looking at a combination of both of these. They definitely help with what I'm trying to analyze

you guys are da best

This is true

Especially since the Chilean is not in channel

@billinkc You've got his country right – you two must've been getting on lately then :)

Grrrr

Dear C#:

> You are my best creative partner and we will build many more great projects together. And as far as I’m concerned, your special son PowerShell can live with us for as long as he needs, even indefinitely. Everyone around the shop loves him and knows he gets a lot done, in his own peculiar way; we just ask him to sit on a towel.

(emphasis mine)

blog.tnwdevlabs.com/2015/07/01/…