« first day (4290 days earlier)      last day (578 days later) » 

12:34 AM
Animal mineral vegetable rude boy
 
 
3 hours later…
3:35 AM
no u
 
4:20 AM
4u
 
 
2 hours later…
6:01 AM
Morning
@J.D. They were very LIFO
 
6:14 AM
@PaulWhite nou4u
 
seems so
 
@PaulWhite Why is this learn link showing up as docs?
 
@Zikato Original and best link provided for your superior enjoyment
 
7:12 AM
A chairde - Morning all!
Is this a shopping list or not? AFAIC, a shopping-list is "which tool is best/great for..." - searching for the mere existence of a tool does not a shopping-list make, IMHO. Open to correction on this!
Paul White posted a site where you could make up tables for presentation on various sites - here and elsewhere - does anybody have the link? I'll be sure and bookmark it this time!
 
@Vérace Do you mean markdown tables? ozh.github.io/ascii-tables
 
Also, another thing I was pondering last evening - I do a lot of pondering... there's a diagram showing the "phylogenetic tree" of the various RDBMS's out there - INGRESS normally being considered the granpops of them all and how the different systems came from that - PostgreSQL, Sybase, SQL Server... inter alia... anyone got a link to that as well?
 
@Vérace You're thinking about it the wrong way. Have a read of this:
22
A: Do we accept database comparison questions or are they offtopic?

Paul WhiteThese types of questions are harmful to the site and should be quickly closed and probably deleted. I know that SO has a no-shopping-list questions policy, and this feels like a shopping-list. It's not just SO, this has been established network wide for a long time. Just two meta.SE example...

 
7:29 AM
Re. tables - thanks for that! I'm looking for jupyter R notebook markdown - is it one of the options or should I just go with HTML? R markdown isn't specifically in the dropdown for output style, but I'm not mad keen on HTML! Thanks again...
 
No idea, sorry.
 
Re. shopping-list. The OP is asking about the existence of a tool for what she wants - not asking about (well, she did initially, but I've since edited) which is "best". That's a simple "yes and here it is..." or "Sorry, no, but no such tool exists...". The only problem I see is that a "no" answer could be rendered obsolete with the arrival of such a tool on the scene. However, this is also true of any number of features of <any_RDBMS> where an answer can be obsoleted by a new feature.
Anyway, it's not going to keep me up at night - if you feel it should be closed, then "go ahead punk, make my day!" :-)
 
7:49 AM
@Vérace This is the problem. People see the word "tool" in a question and immediately think of closing the question as 'shopping list'. If I felt the question should be closed, it would be closed. I don't get to vote.
The question is about solving a particular problem. The asker mentions a tool in passing; it is not the core of the issue. There is a risk people will answer with their favourite tool, but that's only one possibility. Better answers would be based on expertise and experience in solving the core problem: database deployment and maintenance.
 
I only noticed that you edited it after I wrote the above - I explained my rationale to a list member who seems to want it closed. I'm going to leave it there. I suppose if dba.se is to be a "community", I have to accept that not everybody will think the way I do and live with that - or else take my toys and go home! :-)
 
@Vérace We discuss policy on Database Administrators Meta, not on user's questions.
 
OK - I'll do that in future! Just wanted to get my fellow Heapers' thoughts!
 
You are welcome to do that too. Now I have the problem of what to do with your and Laurenz's comments.
Of course, I'll be proved wrong now and all the answers to that question will be favourite tool recommendations 😐
 
@Vérace what's Ingres? First time I hear that.
 
8:00 AM
Ingres Database ( ing-GRESS) is a proprietary SQL relational database management system intended to support large commercial and government applications. Actian Corporation, which announced April 2018 that it is being acquired by HCL Technologies, controls the development of Ingres and makes certified binaries available for download, as well as providing worldwide support. There was an open source release of Ingres but it is no longer available for download from Actian. However, there is a version of the sourcecode still available on GitHub.In its early years, Ingres was an important milestone...
> PostgreSQL evolved from the Ingres project at the University of California, Berkeley. In 1982, the leader of the Ingres team, Michael Stonebraker, left Berkeley to make a proprietary version of Ingres.
(from the linked PG wiki)
The first database I used was Informix
 
@ypercubeᵀᴹ Damn, Paul White beat me to it! It's used in Revenue in Ireland (God alone knows why...) but they're moving to Oracle AFAIK! I compiled and ran it once on Linux - v. complex build process - there's still source code on GitHub somewhere - but AFAICS, PostgreSQL has it beaten hands-down on every level.
 
thnx Paul. Reading that wikipedia page, yes, I vaguely remember reading it about it in the past.
 
I think it's slowly dying - but is has left a huge legacy - PostgreSQL (and all of its commerical derivatives), Sybase, SQL Server, Informix... It was originally Michael Stonebraker's project - he recently contributed to Postgres Vision saying what he thinks PostgreSQL's development priorities should be.
 
But the "normally being considered the granpops of them all and how the different systems came from that", that is truly first time i hear it. And I seriously doubt it. Did really Sybase and SQL Server use source code from Ingres, even in the first versions?
 
No, I don't think so
 
8:11 AM
@ypercubeᵀᴹ did you ever get a chance to tackle that tricky SQL we were looking at? I'm right out of time at the moment - I shouldn't be posting here really what with doing a Masters added to my other daily chores :-). I'd love to get to grips with it - maybe when term is over. I think it'll call for "hammock-driven development" (Rich Hickey - inventor of Clojure - a really clever guy).
 
Sorry, no. Maybe this week I'll find time.
 
Those who were involved in the early project went on to found Sybase.
 
ok. that's different to the previous claim
 
 
1 hour later…
9:33 AM
@PaulWhite in your article sqlperformance.com/2018/02/sql-plan/…
> T-SQL does not provide us with a way to write a semi or anti join directly (though interestingly U-SQL does), so we have to use indirect syntax like this instead ...
the U-SQL link is broken ^^
 
Yes, I know. That's why I supplied the correct link in a separate comment earlier when I was phone-bound
I will fix it sometime this evening, thanks for the reminder
And of course the link is broken, because it is an MS link
 
haha
I was trying to find the relevant docs but it's a mess really - looking for SEMI JOIN syntax
but the title "JOIN (Databricks SQL)" is confusing. Does it apply to all Azure databases/products?
Is there a separate Databricks product/tool where SEMI JOIN can be used?
 
It's an utter mess. I have no idea.
Well, I do know that Databricks exists, and is different, but that's all
 
heh, checking "SQL Azure history" page
> Windows Azure was later renamed to Microsoft Azure, SQL Azure was renamed to Azure SQL, and both have since dramatically expanded services. ...
 
9:48 AM
Only Azure Data Lake Storage (ADLS) Generation 1 supports U-SQL
I have no idea if ADLS Gen 2 is different from Azure Databricks or Azure Synapse Analytics, and care even less
It's just stupid at this point
 
 
3 hours later…
12:21 PM
What's up with all the questions that look like they were copy and pasted from the Docs?
 
Such as?
 
The CU-GDR questions, the quorum question. I guess it's just the usual suspect.
 
@ypercubeᵀᴹ Fixed now
 
Summertime coming to a Paul near you, I hear.
 
Daylight saving? That already happened
 
12:32 PM
isn't it warming up?
 
Certainly won't be very summery this week. We're due a polar blast on Thursday
Spring is always like this though. Up and down. Sun, rain, wind, storms, everything.
Sometimes on the same day, sometimes within the same hour.
Possibility of some snow for Wellington on Thursday.
It's a relative to what you're used to ofc. Not cold by many standards.
No hurricane either.
 
Sounds a lot like Pennsylvania
@PaulWhite Very popular drink from what I hear
 
I'd prefer the drink to the weather, that's for sure
 
speaking of drink, I finished off a bottle of port in the hurricane party.
 
Party of one?
 
12:45 PM
Best kind of party
 
I can see that being true
 
 
2 hours later…
2:36 PM
wait did we stop doing dst
 
2:46 PM
@PaulWhite someone i know from nz is flying to the falkland islands
i can't imagine that flight
 
3:23 PM
Someone somewhere posted a link to a blog post someone wrote that they were leaving the MySQL development team and being critical about it. I can't find it now, does anyone know what it was? I'm not even sure it was posted here on chat.
 
it was
Dec 6, 2021 at 11:01, by Philᵀᴹ
Quite a good blog on MySQL, written by one of its [ex] developers https://blog.sesse.net/blog/tech/2021-12-05-16-41_leaving_mysql.html
 
merry christmas
 
 
2 hours later…
5:15 PM
> Kim Tripp (one of the programmers of SQL Server and a integral part of SQLSkills
did she work as a sql server developer?
 
5:32 PM
I know she worked on the documentation team, not sure if she ever worked on the engine
 
yeah
 
5:49 PM
morning
 
 
2 hours later…
7:30 PM
@ypercubeᵀᴹ Thanks yeah that was it. I don't think I was on here that far back though, and pretty sure I saw it in the last month or so, so might have seen it somewhere else.
What's your take on it anyway?
 
pretty honest I think
 
MySQL really does seem like a toy DBMS sometimes. Postgres just wipes the floor with it and it's also open-source. I thought of this blog today when I realized that MySQL doesn't do FULL JOIN, like, what??
It just doesn't seem to have a USP, it doesn't have a single feature I can see that other DBMSs don't have. Unless you count ONLY_FULL_GROUP_BY 😡
 
7:46 PM
hmmm
https://twitter.com/GrzegorzLyp/status/1576918416962686981
 
Looks like a bug in the optimizer, still playing with it though. I think constant-folding is the suspect here.
 
Yeah. Optimizer was too excited
 
8:01 PM
Does it repro in older compats? Inlining off?
 
in dbfiddle.uk it has same results (versions between 2016 and 2019)
 
and 2014 as well (was getting an error first, didn't like the CREATE OR ALTER.
 
it's not scalar in the first place. Can you disable inlining on an inlined function?
I think it's something with data type inference
 
9:03 PM
@ErikDarling but that's a different query.
I mean the MAX() will make the tvf to return 1 row, not 0 rows.
 
i know
 

« first day (4290 days earlier)      last day (578 days later) »