« first day (3956 days earlier)      last day (1211 days later) » 

02:35
@Vérace-getVACCINATEDNOW you see all my questions. ;)
 
5 hours later…
07:12
Morning
yesterday, by mustaccio
Kiwis' latest bird of the year is a bat? Where is this world going?
 
2 hours later…
08:57
Morning
 
2 hours later…
10:53
Is there any incompatibility with SQL Server 2019 and Win 11?
A chairde - Morning all!
@EvanCarroll You're really burning the midnight oil over this one? :-)
Have you investigated the PostgreSQL hackers list(s) for these questions? I know there's Laurenz Albe and Craig Ringer here who are contributors - but I think you might get more bang for your buck there?
 
2 hours later…
12:59
@Vérace-getVACCINATEDNOW I suspect Evan's definition of bang differs from yours.
@mustaccio Not sure I understand this? I meant "bang" in terms of replies/bytes-written - what other sort of "bang" could there be (in this purely technical discussion)?
13:26
In case someone interested - Net 6.0 Launch
 
2 hours later…
15:26
@Vérace-getVACCINATEDNOW Yes. I've posted there too. But I don't normally wait around for more than a few hours before I investigate and find the issue myself. Mailing lists take far longer than that.
@EvanCarroll - depends - hit-n-miss - one can receive a response in anything from minutes to months... a bit like here really. Any quality responses on the pg-hackers' lists?
SQLite list pretty good though - in-depth responses < 24 hours!
16:15
@Vérace-getVACCINATEDNOW No disputing the highest quality of answers you can get come from the list. But I hate that format and they're never in minutes. I am on IRC though, that's the other extreme typically but not really with PostgreSQL. RhodiumToad is legend when you can bend his ear to work on a problem with you.
IRC you typically get instant results with low quality, on #postgresql you get instant results with high quality.
16:43
@Vérace-getVACCINATEDNOW : aws.amazon.com/blogs/aws/…
16:54
Bwahahhah
@ypercubeᵀᴹ that's huge.
@ypercubeᵀᴹ the real awesomeness is when the benchmarks come out showing it's faster.
That's going to be a real hit to the MS jugular.
17:12
@ypercubeᵀᴹ Thanks for the update, but what I'm reading is this: Over time, we are shifting Babelfish to fully open development on GitHub, so there is transparency from the start. Isn't this the long awaited ¿"start"? I think my gentlemen's bet stake is safe for the moment - it was the end of the year!
@EvanCarroll How do you know it'll be faster? Putting layers on anything is a speed drag!
17:36
@Vérace-getVACCINATEDNOW It's not an additional layer... It's not an emulator. It's a reimplementation.
Curious what kind of systems would be fully covered by that and which would need changes anyway such that they could just go to PostgreSQL instead. What features would you not have to use in SQL Server to ensure you are covered? We are pretty conservative in what we use overall, but when we absolutely have to depart from basic features... Like we use some assemblies for some string manipulation in one product and the MS data services assembly to do some XSL transforms as well.
At that point, our things would probably need an architecture review and look to just converting straight to PostgreSQL.
18:12
@CadeRoux Well I would assume quite a bit isn't covered. But details are scant. It would be worth playing with. Thing that probably aren't covered include CLR-invocation both user-defined and system (like GIS). I also assume things like hekaton aren't covered.
const char *unsupported_sp_procedures[] = {
	/* Catalog */
	"sp_column_privileges",
	//"sp_columns",
	"sp_databases",
	"sp_fkeys",
	"sp_pkeys",
	"sp_server_info",
	"sp_special_columns",
	"sp_sproc_columns",
	"sp_statistics",
	"sp_stored_procedures",
	"sp_table_privileges",
	//"sp_tables",

	/* Cursor */
	//"sp_cursor_list",
	//"sp_cursor",
	//"sp_cursorclose",
	//"sp_cursorexecute",
	//"sp_cursorfetch",
	//"sp_cursoropen",
	//"sp_cursoroption",
	//"sp_cursorprepare",
	//"sp_cursorprepexec",
	//"sp_cursorunprepare",
For reference
In any case, I don't think it's viable for us since SQL Server is a pre-req for the main OLTP system my analytics products consume, and I don't see them moving on - those on-premises instance costs are always born directly by the customer, and my DBs sit on the same server.
18:41
@EvanCarroll - I'd say one would be amazed by how little of any given vendor's functionality that is used by many apps. Worked as a consultant once upon a time for a company which provided ERP software to (inter alia) one of Europe's largest rail infrastructure companies - we're talking billions of widgets. I went to look for FKs to "get a feel" for some of the major tables. Ran query - nothing returned.
Thinking I'd messed up, I asked the head guy where my error was. Nope - wasn't an error on my part - the whole system didn't have a single FK - he said (glowing with pride) that all of their searches were on PKs and that they didn't need FKs... They were paying a fortune for Oracle - they could have use SQLite!
Or notepad!
There was a COBOL layer which automagically turned normal queries into queries that worked on the system... COBOL was never my strong suit!
Yeah, I agree. I bed what Amazon has done works for ~80% of SQL Server users and steals billions in revenue
19:14
Something with just tables, no code layer, should be already pretty easy to forklift to PostgreSQL if the data types are pretty basic. Babelfish might not even give you much.
19:33
Well, the syntax is different. Column quoting with [] and such. People are going to want to use their MS tools and query builders which will produce syntax PostgreSQL won't understand.
So if they have dumb tables but all their SQL is in the app somewhere?
But not EF
Or something that can already just go to PostgreSQL
19:58
OMFG - they've f***ing started - songs celebrating the birth of the person who is believed by many of (a) certain faith(s) to be some sort of good guy! Fine - but just don't inflict the singing of your acolytes on f*(&-ing me!
Just after Hallow'een!
Funny how you address your complaint to that same guy ("OMFG")
2
21:02
heh.
One does not challenge the gods lightly. If you find it fit to cross them, go hard. If they exist their adherents leave me to believe that "it was a joke" will not fly with them.
 
2 hours later…
23:02
Oh god the new mayor of New York =(
so not inspiring another idiot spewing American exceptionalism
"There is no Polish Dream. There is no German Dream. There is an AMERICAN DREAM."

« first day (3956 days earlier)      last day (1211 days later) »