« first day (2530 days earlier)      last day (405 days later) » 

12:01 AM
And that's despite the (almost) total mobilization of both political parties against the country.
looks like they're back to roughly where they were before a spike during the peak of the 2nd us-iraq war
is the average american even exposed to iranian PR/propaganda?
I think so, Javad Zarif is on every media outlet. I mean, we're not nearly as exposed to it as we are Saudi and Israeli propaganda. But neither the Sauds nor the Israeli's try to appeal to Americans directly.
The Israeli and Saudi game is largely a mirror image of America at a smaller scale -- staying out of the limelight.
i had forgotten about zarif. think there is also an iranian-american professor who is falrly pro-government and has done interviews with the bbc. still i have no idea what name recognition stats are like for these people.
there's also presstv, dunno if anyone watches it.
12:18 AM
I don't think presstv has seen any real success, I do think Javad has forced a much needed pivot in the Saudi/Iran issue.
The inverse has happened with Saudi Arabia in the past year, fyi. Also from Gallop: 37% favorable to 31% favorable, 55% unfavorable, to 65% unfavorable.
And, that's with all the propaganda about the heir.
friedman got a really negative reaction to his story about the heir
He deserved that reaction. That story was utter trash and naked apology.
Total propaganda.
the saudis very publicly being close to trump probably hasn't helped their image. or the fact that almost everything they've done in the last 5 years or so has been a disaster
yeah agree
Well, everything up to and including 9/11.
Just saying their immense wealth, massive military, and failed attempts at puppet-mastery are failing in the eyes of every nation. While smooth talkers in sharp suits and direct engagement seem to be working for the likes of Canada of Iran even though neither of those countries is doing anything useful or different except in rhetoric and appearance.
2 hours later…
2:44 AM
@sp_BlitzErik: hope that edit helps
@JoeObbish link
check your mail
I got nothing.
weird that you aren't getting notified
@sp_BlitzErik check out the comments of that for a laugh
Oh boy
2:55 AM
makes me appreciate stack exchange
Even the comments sections?
no, I mean the cancerous nature of those comments
That's a tough way to do support. I feel bad for that guy.
is it just the one guy throughout?
it is
I hope that's a fake account with pooled support
3 hours later…
5:45 AM
@sp_BlitzErik I posted an answer
I hope you're happy
2 hours later…
7:49 AM
Morning Europe, Good Evening Australasia, enjoy your meal Central Asia, g'night Americas
8:26 AM
Morning all
Just seen this. Doesn't matter if you believe in evolution or intelligeet design.
Either way, we as a species need culling
@gbn At least he got some new followers
9:08 AM
Q: PostgreSQL - Can simultaneous updates and deletes on a set of tables lead to AccessExclusiveLock being taken?

TheLostMindI am using Postgres 9.3. I have 2 tables : authn_session and customer. Every authn_session belongs to a customer (and hence has a customer_id as a column which is a FK to customer's id). Note : These tables contain additional FK references to other table(s) and indexes. Now, I start 2 differe...

Seems like an interesting question for the Postgres experts. BTW the OP is a moderator on SO.
But don't hold that against them.
9:30 AM
@gbn Evolution in action.
9:43 AM
@gbn contestant for a Darwin Award probably?
10:39 AM
@ypercubeᵀᴹ yep looks good now
10:50 AM
@PaulWhite happy saturday
@sp_BlitzErik Thanks! Perfect timing :)
the game is afoot, pablo
@sp_BlitzErik you are early today
@PaulWhite happy sunday
@dezso i had an odd dream and couldn't fall back to sleep
11:05 AM
@sp_BlitzErik ha ha ha even I do not have the patience (or scripting skills) for that
fancy debugger man who changes memory addresses doesn't have the patience
that's rich
11:25 AM
My wealth comprises entirely $100 bills
have you tried to use those to pay for anything yet?
Q: Service Broker cryptographic error

DianaI'm trying to set up a test environment for Service Broker communication between 2 SQL Server 2016 SP1 instances. I've created db master keys, certificates and endpoints in the master db for each instance, as in the code listed below. For now my goal is to synchronize data between 2 tables. On th...

Wasn't service broker deprecated several versions ago
That's not the first time I've heard that recently. I wonder why people are thinking that?
probably because its reached log shipping status where there's just not much to do with it
reading through this is strange
Use of #, ## as temporary table and temporary stored procedure names.
Use at least one additional character.
11:43 AM
@sp_BlitzErik Strange that # and ## are valid? Or something else?
DECLARE @ integer = 5;
I don't think I've ever seen that in practice though.
that's magical
On the Service Broker/Log Shipping thing: I'm no expert in either, but it is my impression that both are useful for their intended applications. LS in particular seems to have a solid fanbase. Has it been made effectively redundant by basic AGs or something?
i'm log shipping till death
I see SB got some small amount of love in 2017 so it's certainly not dead, just matured.
Ah I see so you meant both were mature technologies now with not much more to add
i seem to recall SB getting the ability to index, or rebuild indexes on its internal tables or something in a semi-recent version
maybe 2014?
anyway, never used it
11:50 AM
Sounds plausible
I used it once to prove that I could do it.
Getting it to work non-locally was "fun".
But I do like LS in principle because it is simple
A quality new things tend to lack sometimes
i try to drive people towards it because it doesn't take a team of dbas to manage
ags are fine to set up but as soon as you need to patch or something goes wrong it's a bloodbath
I'm all for avoiding unnecessary bloodbaths
are you though?
11:53 AM
In most cases
legendary cricket hooligan avoids bloodbaths
i'll be.
You're weird but I love you
it's my thoughtful voting, i know
It's a factor, certainly
@sp_BlitzErik After using replication, AGs are quite simple. Give me AGs any day
I'm not masochist enough to use replication unless I really need to on pain of death
11:57 AM
replication certainly is at the very bottom
i still like mirroring better than ags
deprecation be damned
Availability Groups are not my BAG
@sp_BlitzErik Especially now things get undeprecated
@PaulWhite i like it because it's the only HA or DR technology that doesn't require anything to touch the primary copy of the database
@PaulWhite this comment is allan hirt approved
@sp_BlitzErik Then I withdraw it and apologise unreservedly :)
all of the others have edge cases where the secondaries or the HA technology itself can cause an outage on the primary
12:00 PM
@JamesL Good point
@JamesL if you had me set up a new environment today, i'd still use a FCI and LS over AGs or anything else
Acronym Man saves the day
you're welcome
I was typing the same thing :-D
has anyone emailed you about finding a switch operator yet?
12:08 PM
@sp_BlitzErik Not so far, but I do have quite a few unread emails.
are any from me?
@sp_BlitzErik No, you have a special folder
@TomV That is an excellent example
there was something i was going to write you about but i'm at a loss to what it was
Perhaps it was an idea for what I should buy my wife for Christmas this year
Goodness knows I could use the inspiration
12:12 PM
new vape pen
make her a sandwich
@PaulWhite Shoe-socks
@sp_BlitzErik She doesn't
@sp_BlitzErik Let's call that Plan D
@TomV Hm
of course not, she doesn't have a vape pen
if you got her one she would
I cannot fault your logic
you just don't understand women.
the timing of that did not work out in my favor
12:15 PM
That's what I'll be saying on Dec 26
Dec 27 '15 at 13:02, by Paul White
@TomV I must be old, I quite like the idea.
I'm just reminding you :)
@TomV Yes I remembered thank you
get her a fun boy three shirt
Frankie Says Relax perhaps
Clothing is always an option, but (a) quite high risk; and (b) it's really hot
@sp_BlitzErik To match Paul's "the cure" t shirt
12:18 PM
paul looked dashing in that shirt
nearly asked him to dance
@sp_BlitzErik right. And any of us blokes here do?
@sp_BlitzErik i like AGs but it's really hard to get everyone to connect to the right name with a driver that supports multisubnetfailover and not do annoying stuff like block DDL on the secondaries
@TomV Bananarama t-shirt would go better
@gbn just enough to stay in my office most of the time :D
12:27 PM
for a lot of non-oltp apps here we just maintain multiple, independent copies of the app and database
@JamesL ah, yeah, that's no fun
@JamesL yeah. Even SSIS Catalog connections required the mirroring syntax, bit the AG syntax
12:52 PM
@sp_BlitzErik Erik Luther sp_King
that's not even how my name is spelled
@sp_BlitzErik buy new glasses
it's a gym day i'm wearing contacts so i don't look like a nerd and get wedgies
@PaulWhite does sql server's optimizer have a fallback algorithm (or a different heuristic/search space) after a certain number of joins are encountered?
@sp_BlitzErik Yes in several ways
1:38 PM
Completely off topic, some bloke is on the office patching and rebooting the water cooler
Had to reboot the office coffee machien earlier this week
Are they plugged into an ethernet jack?
@TomV Dunno. I ran away. It was bad enough with the Smart TV and phone reboots
2:05 PM
Q: What's the duration of U lock

BorkaAs far as I know, SQL Server holds X locks for the duration of a transaction no matter which isolation level is in use. What about the U lock? If a transaction reads 100 records in an UPDATE statement (U locks) but updates 10 records only, will we end up with 90-U and 10-X after the update is don...

quite durable, that one
@gbn We have a tap that needs a computer to work. Have to reboot it once in a while
morning and evening
Q: Calculations over UNION, merge query repetitive code

Kavvson EmpcraftI am struggling with this query a lot mainly because of the repetition which possibly could be reduced - hopefully. The general idea is to display sums of each table for a given month and display it as a pivot table. Current code is in my opinion very messy and repetitive but the main thing is ...

2:20 PM
@Philᵀᴹ in the pub?
@gbn the gen0 digital cats are selling for $7k+ now
obviously this sort of thing will go on forever
until folk get bored
Did anyone sell those Tamogotchi things to each other?
not sure. i was more of a tickle me elmo speculator
1 hour later…
4:03 PM
@gbn one of my favorite things to do was make IP address labels and stick them on things that didn't have IP addresses, like coffee machines and lamps.
morning / evening
@JoeObbish what had you posted an answer to?
@sp_BlitzErik your "fake news" business
you've been more obtuse lately
4:09 PM
@sp_BlitzErik no u
i can't follow your young and nimble mind
me too, thanks
oh yeah, speaking of which
I owe you an electronic mail
4:29 PM
> at a fraction of the cost of bottled water
bottled water must be rather expensive around there.
@sp_BlitzErik our new electricity meter proudly states a MAC address and a public key fingerprint on itself. It either connects via some wireless solution, or it does not, as there is no network cable to attach.
it's just there to make you paranoid
wrap it in tin foil
that won't prevent pregnancy
i've tried
I'm surprised that you're still married
that makes three of us
5:37 PM
@gbn By and large I try to avoid internet connected things where I can't reinstall the O/S. Even mobile phones are grudgingly accepted.
@PaulWhite Interesting factoid: Rotorua had a manual exchange up until 1970 because the sulphur compounds in the air would corrode electromechanical ones.
@sp_BlitzErik Remember these are $100 NZ peso bills, which are worth about three fiddy each.
@PaulWhite Lack frequently. I've long since concluded that simplicity has some intrinsic value because it has fewer failure modes.
I have nostalgic fantasies of a world where computer architectures were much simpler. When I were a lad I could go from a standing start to writing assembler code on a BBC micro between christmas and about april or may the following year.
Today, going from no programming experience to writing Win32 system code is not something most folks could do in three or four months.
@gbn's experience with CORAL-66 is similar to a few old greybeard stories I've heard about folks working on minicomputer architectures where they knew what the compiler would emit for given keywords.
Kit like that was simple enough that a reasonably competent programmer could be expected to gain a working knowledge of the whole system. Unix was about the last widely used platform on which you could do that without being Mark Russinovich.
Uphill both ways etc.
@PaulWhite I know the gift for the man who has everything ...
Fossilised dinosaur poo.
2 hours later…
7:22 PM
I'm considering starting using SQL-89 syntax in my answers. And NATURAL joins ;)
If the questioner isn't trying, I'll upvote them for style.
2 hours later…
8:56 PM
I wonder what the logic is on not letting people upvote answers that they can mark as chosen because of their own lack of experience.
How does that encourage good behavior any where?
9:20 PM
i believe it's to prevent fraud
users could ask daft questions under unregistered/dummy email accounts, answer and upvote
9:33 PM
Unless they're trying to prevent gaming the system specifically for a badge, that doesn't matter much. If unregistered dummy users create daft questions and mark the answer as chosen, the answer gets +15 REP.
make another meta post then?
i'm just here to game the system specifically for badges
I'm just sayin yo, getting chosen answers is easier than getting upvotes. And, it seems to be in part because people asking good questions can't upvote. =/
do you upvote the people asking?
Yes, almost always.
then it shouldn't take long for them to gain the privilege to upvote
9:39 PM
if only postgres had a bigger community on here
Well, I have any easy solution to that. ;) I've offered it multiple times. Put all questions about SQL on topic here. If a question is tagged with just a sql or an rdbms, prompt the user to see if they want to migrate it to DBA.SE from SO.
It probably instantly double the traffic here and alleviate misplaced questions on SO.
I'd even be for not-prompting the user. It would be dirt simple to infer the kinds of questions that belong here just from the tags provided by the user asking the question.
aren't there a lot of people on SO who answer database questions?
Yes, certainly, but does that matter, and if so why?
aren't there some rules about the kinds of questions that belong here vs there?
i.e. basic syntax
The basic syntax rule is really, really stupid and needs to be flushed.
That just adds confusion to the site.
9:51 PM
you're gonna be racking up the meta posts tonight
hope the lovely ladies of houston are patient
Not to mention, it's not enforced. I don't think we've closed a question and migrated it to stackoverflow in years for having "basic syntax"
i dunno, i seem to recall voting on a lot of migration flags
maybe i'm wrong
That basic thing is totally specific for this site. StackExchange should remove it for sanity, regardless of what a community wants.
I remember reading all the FAQs, and there wasn't another site in the network with that.
perhaps we're just more refined
i think my main opposition would be the already questionable question quality
It's the problem with growth in general, the worst way to deal with it is to leave it on the table.
10:09 PM
fair enough
maybe the user influx would help balance out the ridiculous ensuing review queues
but that's a big maybe
Usually you pay an upfront cost for an influx of users, and then the community matures and you'd get that help. Problem is now people are growing up with DBA questions on StackOverflow because they're "basic" and on topic there. Got to invest in that borg of reviewers.
check out this growth hacker
10:27 PM
Trying to get this PostgreSQL paragraph down..

> If data alignment is unimportant (either just for a specific function or because the data type specifies byte alignment anyway) then it's possible to avoid some of the overhead of PG_DETOAST_DATUM. You can use PG_DETOAST_DATUM_PACKED instead (customarily hidden by defining a GETARG_DATATYPE_PP macro) and using the macros VARSIZE_ANY_EXHDR and VARDATA_ANY to access a potentially-packed datum. Again, the data returned by these macros is not aligned even if the data type definition specifies an alignment. If the alignment is important you must
oh we're talking about the rules in here
what's the right thing to do about answers posted as comments? flag them?
"If data alignment is unimportant" is no good. Why would it be unimportant in that context and what does that even mean?
BTW, one disturbing thing about PostgreSQL that I'm sure SQL Server does more right is pack tables. PostgreSQL is stupid about this.
So there I complained about PostgreSQL. ^
@JoeObbish only if you're on mobile, i think
I've never been on mobile in my life
10:32 PM
the mobile site really weird. that ui is confusing too.
for instance, if you hit it on the desktop, it's not immediately clear how to get back to the desktop version and it stores the setting in a cookie or something horrible.
The link is on the bottom "full site", and cursor is broken on it.
we should probably ban mobile sites
they should all be angular/react native and handled on req values.
I agree
who can write good answers on mobile anyway?
in other news I'm a bit surprised no one answered that partial index question yet
You do it for the hat. And, then you edit it on the desktop.
I thought I was just being dumb
10:40 PM
@JoeObbish only the hat lovers, during hat season ;)
@JoeObbish which q?
@JoeObbish did you try filtered indexes that include and exclude the value you're deleting?
Not that I intend to answer. It's late for answers tonight.
@sp_BlitzErik yes, I believe so
that was the first thing that I tried, even
thinking through stuff
computed columns
the weird thing about it is that it's very restrictive in the first place about indexes
10:43 PM
wouldn't be xml or spatial
I was shocked when the old method wouldn't look at covering indexes
I can think of two things that make sense from a design point of view
filtered indexes and indexes that contain the first key of a two column foreign key
neither one seemed to be it
for all we know the feature doesn't even work
dumb question: did you try on 2017?
weird, none of those downvotes are mine
would you like some?
@sp_BlitzErik I was going through withdrawl.
10:46 PM
@sp_BlitzErik all of my testing was on 2017
google image search?
people still use 2016 for blog posts?
@JoeObbish for now
I don't want outdated content
there was a guy around for group by today asking about windowing functions on 08r2
gave me the sads
10:48 PM
yeah that's tough
R2 betrayed him
what else are you using from 7 years ago
i've lost like three wedding rings in five years
i think i swallowed one
@sp_BlitzErik sounds like the LOTR story
ah, you've met my ulcer
@sp_BlitzErik windows 7?
you're amish so whatever
10:51 PM
not me, other people
i re-wrote the parameter sniffing query from the precon
it's spectacularly bad now
for only 50 rows
what does it do?
(query plan)
I thought you were all supposed to use Extended Events? :D — sp_BlitzErik 5 hours ago
I immediately thought the same thing
man you OLTP guys have it rough
I get like ten of those per plan
this is more of a reporting style query
but i agree for the most part
anyway, time to go watch home alone with my kid to prepare her for the future
11:06 PM
@sp_BlitzErik sir, I believe that movie is more than 7 years old
11:45 PM
working on pg-bignum

« first day (2530 days earlier)      last day (405 days later) »