« first day (3206 days earlier)      last day (1655 days later) » 

5:48 AM
Morning all
 
Morning
 
Morning
 
6:49 AM
Morning
 
7:06 AM
Morning
 
7:50 AM
so @PaulWhite I did some spelunking around New Chat
The biggest open source competitor is basically just a Slack clone 😂
 
8:17 AM
@George.Palacios The free version seems to offer quite a bit more than slack does though
Ah but I missed that you have to self-host it
I may try it as a docker on my nas at some point
 
8:51 AM
@TomV I could potentially host. Doesn't look like it needs too much in terms of resource
 
9:31 AM
There's also this
 
 
1 hour later…
10:41 AM
What's this about yet another chat?
 
@Taryn <Menacing cello music>
 
@Philᵀᴹ Fallback option
It's mostly just speculation for now
 
@George.Palacios In this room?
 
@TomV I think it was mentioned in stop the merry go round
 
10:58 AM
@George.Palacios it's been a while since I lurked there
Sean's avatar is discomforting though
 
 
1 hour later…
12:15 PM
328
Q: I'm resigning as a Stack Overflow Community Elected Moderator

George StockerI'm resigning as a Stack Overflow Moderator. Thank you to everyone who has worked hard to make this Q&A site what it is. I'm resigning for three reasons: Stack Overflow Inc. has forgotten how to lead, how to persuade, and how to talk with the community. This has been a slow decline since 2014...

 
1:06 PM
Wasn't that the co-worker of Sara who was defending the actions at first?
I've lost track of the whole saga
How many mods have left in total so far?
 
@TomV 27, + 26 suspending activity, spanning about 80 moderator positions across the network according to one estimate.
 
@Zoethetransgirl So more than half
 
1:22 PM
Counting by mod positions, about 11%
 
Ah ok, the mods had a total of 80 positions when accounting for moderators who held the position on multiple sites.
Got it now, my English parser failed me
 
@TomV If he was, he at least also was the author of the most upvoted response to Sara's "Update & Apology"
 
1:42 PM
@TomV George was never an employee, just a long-standing SO moderator
 
@PaulWhite George worked with Sara at one point though.
 
@PaulWhite George worked with Sara at Jewelbots
 
Ah. Didn't know that.
Thanks!
 
Yeah that was it, right
I believe Ypercube mentioned it here but I couldn't find it any more
 
 
3 hours later…
4:36 PM
Currently, we partition data into physical databases based on the ones value of a monotonically increasing number - so a round robin distribution/sharding across 10 databases. The table in the database uses a composite key and we have ye-big-view to bring everything back together
I wanted to evaluate whether a partitioned view will improve performance as we always know the location value (the referenced monotonically increasing value) and it is part of the primary key
But, the check constraint must be simple to eliminate tables in a partitioned view so I can't use location_id %10 to check
Next thought was to make a persisted computed column that does the modulus and that would satisfy the check constraint but where would you put that in the composite key?
It's a constant so internet says we want a selective column first but mentally it feels wrong
 
5:07 PM
dba.stackexchange.com/a/167579/2639 Lol @ my conversation example
 
Are Erwin and a_horse inactives these days?
 
5:32 PM
I see less of them. But in fairness most people probably see less of me too
moar to learn elsewhere.
 
Both answered a question today
Though I can't remember jnk handing back the diamond. Must have happened silently
 
1
Q: Database Stuck 'In Recovery' with command XTP_DB_RECOVERY

InMemoryHardshipsWe recently converted a table to memory optimized data. Our backups have all ballooned (3x300GB files to 3x600GB files for Full, 3x50GB to 3x250GB for Diffs), and startup has been getting progressively slower. In an attempt to stave off these issues we converted the table in question to SCHEMA_O...

Interesting problem this
@TomV Yah, and it was a while back
 
Can two users have the same nickname?
JNK exists twice
 
@jcolebrand he pioneered the pronoun thing. Asking to be called a dude on his profile page
@McNets yes
 
@TomV ok
 
5:53 PM
@TomV there was discussion about one of his post
Oct 3 at 19:37, by Lamak
@Philᵀᴹ I noticed that I don't agree much with George, but that apology is a thousand times better than the thing we got
The co-worker info may was mentioned in another (hidden/external) chat room ;)
@McNets I think I saw Erwin post.
Sep 18 '18 at 8:47, by dezso
when did @JNK resign from being a mod? I somehow missed it.
 
@ypercubeᵀᴹ ok
 
Sep 18 '18 at 10:30, by Paul White
@dezso Sure. He'd been inactive for a long period of time due to changes at work etc (as he said himself in The Heap many times). Eventually there was a review and the diamond status was removed. Like any elected mod in good standing, he can reclaim it at any time, if his circumstances change.
( from your friendly heap-historian ;)
 
 
2 hours later…
7:58 PM
Thanks
 
8:23 PM
23
A: Calculate Hash or Checksum for a table in SQL Server

Cade RouxYou can use CHECKSUM_AGG. It only takes a single argument, so you could do CHECKSUM_AGG(CHECKSUM(*)) - but this doesn't work for your XML datatype, so you'll have to resort to dynamic SQL. You could generate dynamically the column list from INFORMATION_SCHEMA.COLUMNS and then insert int into a ...

Anyone have any idea why someone would have different checksums on data that they say looks identical in different databases? Comments on an old answer of mine.
 
8:37 PM
@CadeRoux it has to be a local machine generated seed for hashing
Now the question is does that live in master or on disk
I know that .NET does this on first-request but keeps it in the app domain during runtime. If you configure it in a config file you can force the same version across multiple machines.
 
Surely because there’s no ORDER BY? The thing you are checksumming wouldn’t be the same initial string if the orders of the resultset were different
 
8:53 PM
@Philᵀᴹ If column order is different yes, row order, no.
docs.microsoft.com/en-us/sql/t-sql/functions/… "The CHECKSUM_AGG result does not depend on the order of the rows in the table."
@jcolebrand This document doesn't hint at that docs.microsoft.com/en-us/sql/t-sql/functions/…
 
@CadeRoux right. But I don't think a lot of MSDN articles talk about configuration values that default when they aren't specified in machine.config or app.config
 
@jcolebrand It would seem like it needs to be portable. It's supposed to be deterministic because you can use it in a computed column and have it indexed.
 
9:22 PM
I just ran this in dbfiddle vs local on SQL 2012 and it matched: dbfiddle.uk/…
And all versions within dbfiddle matched, including Linux.
 

« first day (3206 days earlier)      last day (1655 days later) »