Which RCS is the best James Blunt?
-
Not very POSIX compliant.
What does it mean for a DBMS to be "POSIX compliant"? I wasn't aware POSIX says anything about databases.
-
(May I add that there was a discussion about having a bug database proper. The consensus was that the mailing list Just Works as is.)
Wait, they ONLY have a mailing list? No bug database or knowledge base?
I now instantly hate this product.
-
I have enough short term memory for this, and can into \d and information_schema.
Well I don't, fuck you. I'm just as entitled to use a DBMS as a person with good rote memory.
-
What does it mean for a DBMS to be "POSIX compliant"? I wasn't aware POSIX says anything about databases.
Yeah, sorry, mistype. I meant ANSI/ISO compliant.
-
Blakey, you're also entitled to go and fuck yourself. Happy new year, while we are at it.
-
T-SQL is better than Oracle's variant of SQL. (The SQL spec specifically says empty string does not evaluate as null.) But yeah I could see that.
But also: 99% of the improvements in T-SQL make the language WAY better and more practical so I'm also ok with it. I wish pagination was easier. Oh and while they're implementing shit not in the spec anyway, PLEASE give us a
MEDIAN()function.AVG()(which is named "average" but implements mean, because whoever wrote the spec is an uneducated idiot apparently) is useless in so so so many fields. I wouldn't mind except implementing median "manually" in SQL is a bitch. (Select top 50% one sort order, then select top 50% the other sort order, then average the two results. PITA.)
-
Blakey, you're also entitled to go and fuck yourself.
I'm also right.
Accessibility is important. You can't go around just saying, "hey everybody should use the software like me," when the fact is: not everybody is like you. You just literally SAID the way you use PostgreSQL requires having tons of short-term rote memory. Guess what? MOST PEOPLE DO NOT HAVE VERY GOOD ROTE MEMORY AT ALL. Very very very few people do, in fact. Wow, great for you, it's like a superpower, but you're also saying "fuck you" to 95% of the population of planet Earth. So yeah, if you're going to say "fuck you" to 95% of the population of planet Earth, you gotta accept a couple "fuck yous" in return.
And the WORST part is that IT (at least the idiots making programming tools and open source bullshit) are like you. "Got mine, fuck everybody else" is their attitude. That's why systems that allow the use of spatial memory are practically extinct, and also a huge component of the reason people dislike and avoid computers.
Now I don't expect you to stop being a dick, but at least have a TINY BIT OF SELF-AWARENESS about it. I don't think that's too much to ask.
In the meantime, SQL Server continues to sell like hotcakes despite its cost primarily due to the fact that it's not developed by dumbshits like you. It's developed by people who are creating software for actual human beings to use.
-
So yeah, if you're going to say "fuck you" to 95% of the population of planet Earth, you gotta accept a couple "fuck yous" in return.
I'm OK with that. Fuck me, fuck you, fuck everyone (obligatory Oprah pic). Everyone is stupid in their own little ways.Guess what? MOST PEOPLE DO NOT HAVE VERY GOOD ROTE MEMORY AT ALL. Very very very few people do, in fact.
Wait. It's like you're working with all that shit, seeing it pop up on your screen into your face all the time, typing it, and forgetting it the very next second? Are you a goldfish or what?And the WORST part is that IT (at least the idiots making programming tools and open source bullshit) are like you. "Got mine, fuck everybody else" is their attitude.
Because maybe they are good in building the actual database backends and they don't try to enter the UI/UX territory while being incompetent at it? (By the way, sql*plus is total utter shit; last I checked, it even hadn't history. psql shines in comparison. Compare the budgets...)
-
But also: 99% of the improvements in T-SQL make the language WAY better and more practical so I'm also ok with it.
Yeah, I'm also OK with it.
What annoys me when they implement the same features that standard specifies, except they use different syntax. Unfortunately, I can't remember ATM what it was that annoyed me the last time I seriously worked with SQL Server.
Probably not a big deal anyway.
-
Everyone is stupid in their own little ways.
Right; but your attitude is harming other people.
For example, PostgreSQL users can't look up known issues in a knowledge base or bug tracker because of stupid attitudes like yours.
Are you a goldfish or what?
Say I do have goldfish memory; does that disqualify me from being able to use a computer? Does that mean I deserve no respect or consideration as a human being? Think carefully about your answer.
Because maybe they are good in building the actual database backends and they don't try to enter the UI/UX territory while being incompetent at it?
They don't have to "enter" it, but they do have to acknowledge a need for it. If the big weakness of the product, a big reason Microsoft is making $$$ on SQL Server, is GUI tools, then they should get someone to produce GUI tools.
Now a normal person would say "duh". That's part of running a software product: identifying the areas where your product is weak, and improving on those at a higher priority than areas where it's already strong.
Not so in open source! In open source you just ship shit for decades, confused as to why nobody's fucking using it.
-
The reminds me, I've filed request on Microsoft Connect quoting another MVP for suggestion of including least()/greatest() in T-SQL, and ultimately it was marked as "Won't fix" when SQL2008R2 is released and they have to cleanup the outstanding requests.
So we have to continue using long CASE WHEN ... THEN ... ELSE ... END statement to evaluate largest/smallest values in multiple fields.
-
For example, PostgreSQL users can't look up known issues in a knowledge base or bug tracker because of stupid attitudes like yours.
There's a wiki (https://wiki.postgresql.org/). Consider a knowledge base is covered.Also, there's quantile extension for PostgreSQL that delivers medians and does them fast (https://wiki.postgresql.org/wiki/Aggregate_Median for more comprehensive look at the problem).
They don't have to "enter" it, but they do have to acknowledge a need for it. If the big weakness of the product, a big reason Microsoft is making $$$ on SQL Server, is GUI tools, then they should get someone to produce GUI tools.
I think someone would be rich by making such a tool. But I think that before becoming rich, there's a lot of costs to sink on research. I fail to see how such tools would apply to the current power users (who mostly have their own tool sets and are satisfied with them). Most newcomers would probably want "MSSQL Management, but for PostgreSQL", and that's wrong IMO, because the two are too different. (On the other hand, ALL requests to "make a clone of tool X for vastly different Y" are wrong and should be discarded.)Also, PostgreSQL itself is not sold (it's not developed by any single for-profit company); PostgreSQL-based solutions are being sold (and it usually means highly customized GUIs for domain problems). A completely different perspective.
-
Say I do have goldfish memory; does that disqualify me from being able to use a computer?
To be fair, that would probably disqualify you from being employed as a database programmer or admin.
-
Right; but your attitude is harming other people.
So is yours. You assume 95% of all people are totally like you, while lacking solid data to back that up. And you are also an arrogant dick, at least on this forum. And you don't even work on open-source stuff to make up for being an arrogant dick. (shrug)
-
Most newcomers would probably want "MSSQL Management, but for PostgreSQL", and that's wrong IMO, because the two are too different. (On the other hand, ALL requests to "make a clone of tool X for vastly different Y" are wrong and should be discarded.)
How are they different? They are both relational database management systems, with similar feature sets and language supports. They are competing in pretty much the same market and targeting the same user base.
-
How are they different? They are both relational database management systems, with similar feature sets and language supports. They are competing in pretty much the same market and targeting the same user base.
Very different at the level DBA sees them — the area a GUI tool starts becoming remotely useful. They generate traffic at different points, they do logging/replication differently, they store data differently, they have different index types (apart of B-Trees) they have different configuration parameters that you need to be aware of. Throw in foreign tables support (which PostgreSQL has but MSSQL does not), and congrats, you open a can of worms.Also.
I've heard JetBrains wants to release 0xDBE, a database IDE which supports all of them. Schema management is covered, at least. Again, administration is too much of a bitch. I tried EAP versions, and they are rather good.
-
Very different at the level DBA sees them — the area a GUI tool starts becoming remotely useful. They generate traffic at different points, they do logging/replication differently, they store data differently, they have different index types (apart of B-Trees) they have different configuration parameters that you need to be aware of. Throw in foreign tables support (which PostgreSQL has but MSSQL does not), and congrats, you open a can of worms.
Yeah, OK, implementation details are different.
But I don't see anything here that would prevent PostgreSQL from shipping with something like SSMS and SSCM tools that come with SQL Server.
In fact, it already does ship with pgAdmin, that's doing like 60-70% of those tasks. The problem is that it's terrible.
I've heard JetBrains wants to release 0xDBE, a database IDE which supports all of them. Schema management is covered, at least. Again, administration is too much of a bitch. I tried EAP versions, and they are rather good.
Trying it out right now. Seems pretty similar to Navicat in some respects. Coding window is pretty good. Still haven't tried populating a few million rows into its grid view to see what happens.
-
T-SQL is better than Oracle's variant of SQL...Oh and while they're implementing shit not in the spec anyway, PLEASE give us a MEDIAN() function. AVG()
PERCENTILE_DISC
Of course...
MEDIAN
-
Very different at the level DBA sees them — the area a GUI tool starts becoming remotely useful.
I find having a GUI for running / developing / testing queries to be immensely helpful. YMMV, of course, but you're making a blanket statement about something that you can only really state for yourself, and I suspect that it would be helpful to you, too.
-
Trying it out right now. Seems pretty similar to Navicat in some respects. Coding window is pretty good. Still haven't tried populating a few million rows into its grid view to see what happens.
It does paging, idk if infiniscroll or not.I find having a GUI for running / developing / testing queries to be immensely helpful. YMMV, of course, but you're making a blanket statement about something that you can only really state for yourself, and I suspect that it would be helpful to you, too.
For queries, pgAdmin is decent enough. JetBrains' 0xDBE looks far better and has superior autocomplete for everything (and an ER diagram if you want; my schemas were complex enough to make these unreadable, though). pg_cli, written in Python, runs circles around psql and does a good job at inspection/autocompletion.
-
Trying it out right now. Seems pretty similar to Navicat in some respects.
I don't know what the killer feature of Navicat is. I don't feel compelled enough to throw money at them.
-
No idea, but their mobile website is terrible...
-
I don't know what the killer feature of Navicat is. I don't feel compelled enough to throw money at them.
It's pretty, it supports PostgreSQL and it's not pgAdmin.
Other than that, it's full of fail. If pgAdmin worked reliably, I'd actually rather used it.
Can't wait to replace both these turds with JetBrains DataGrip, if it pans out.
-
Can't wait to replace both these turds with JetBrains DataGrip, if it pans out.
That's how they called it, finally?
-
Yep.
Just hit 1.0 a little while ago IIRC.
I'm impressed
-
That's how they called it, finally?
DataGrip: The Cross-Platform IDE for Databases & SQL by JetBrains
A powerful IDE from JetBrains for SQL on macOS, Windows, and Linux.
-
Yep.
Just hit 1.0 a little while ago IIRC.
I'm impressed
My one complaint: there's no easily accessible DDL overview of the objects you're working with. You can right click a node, copy DDL into clipboard and paste it into an SQL editor, but this is IMO clumsy. They should add a window where you can glance and see the structure with which you're working. pgAdmin actually does a good job here.
-
It's pretty, it supports PostgreSQL and it's not pgAdmin.
Meh.After spending some time with pgAdmin, I took time to learn enough information_schema to inspect stuff without training wheels. I suspect pgAdmin, being Qt4-based, only works well enough on Windows which I don't use.
https://www.jetbrains.com/datagrip/
I'm starting to feel pretty much compelled to subscribe to their "Full pack" (I already have PyCharm and PHPStorm). Probably in spring.
-
My one complaint: there's no easily accessible DDL overview of the objects you're working with. You can right click a node, copy DDL into clipboard and paste it into an SQL editor, but this is IMO clumsy. They should add a window where you can glance and see the structure with which you're working. pgAdmin actually does a good job here.
Strange, I thought it was there. That, or my memory is starting to fail me (I'm now doing databases mostly for my own use, my job is far from that).
-
I suspect pgAdmin, being Qt4-based, only works well enough on Windows which I don't use.
I'm using it on Debian.
It's the same turd everywhere.
-
On a Mac, it didn't give a fuck about Mac-native shortcuts, forcing me to use PC shortcuts. I have no problems with either platform, but having to use the paradigm from one platform on another just annoys me.
-
Wait. It's like you're working with all that shit, seeing it pop up on your screen into your face all the time, typing it, and forgetting it the very next second? Are you a goldfish or what?
What's the point of memorizing things when there are tools that make it unnecessary?
-
Caching, ever heard of it?
-
Say I do have goldfish memory; does that disqualify me from being able to use a computer?
I would say if you did have a goldfish memory it should disqualify you from some uses of a computer. I wouldn't trust someone to run a payroll who couldn't remember how to do it from week to week. (Payroll software isn't exactly easy to use once you get beyond what Quickbooks will let you do.)
I'm sure you won't be willing to appreciate the nuance between what you wrote and what I did, though.
-
Most newcomers would probably want "MSSQL Management, but for PostgreSQL", and that's wrong IMO, because the two are too different. (On the other hand, ALL requests to "make a clone of tool X for vastly different Y" are wrong and should be discarded.)
I concur; GUI projects should start from a blank slate.
That's another gripe I have about open source, even those open source projects with GUIs, the GUIs are usually copied wholesale from a closed-source competitor. (See: LibreOffice.)
(Of course that's not to say not copying wholesale produces a better result. See: GIMP.)
To be fair, that would probably disqualify you from being employed as a database programmer or admin.
But SHOULD it?
And you don't even work on open-source stuff to make up for being an arrogant dick.
How would filling the world with buggy substandard awful software make up for being an arrogant dick?
They are both relational database management systems, with similar feature sets and language supports. They are competing in pretty much the same market and targeting the same user base.
Right; but that doesn't mean Microsoft came up with the best of all possible GUIs for that type of tool.
PERCENTILE_DISChttps://msdn.microsoft.com/en-us/library/hh231327.aspx
Slick, but it's brand-new. Back when we needed to write our own
MEDIAN()like 4 times a day, we were on SQL Server 2008. If I still worked there, I'd push for upgrading for that alone.
-
It's like you're working with all that shit, seeing it pop up on your screen into your face all the time, typing it, and forgetting it the very next second?
If the thing is a four-digit number, yeah, it's pretty much a crap shoot as soon as I close the popup. I'll probably remember four digits. They might even be the right ones. If I'm reaaaallly lucky, they'll be in the right places. But I can't count on that. Hi, welcome to learning disabilities.
-
the GUIs are usually copied wholesale from a closed-source competitor. (See: LibreOffice.)
To be fair, LibreOffice started at Sun as StarOffice and was closed-source, open-sourced later. Their designers probably thought copycatting Office 97 (or even 95) interface was a neat idea (because they want to be competitive and learning curve to be smooth, all that), so it stuck (OTOH, I'm using it when I must and feel far better about it than the ribbon; guess I'm getting old).
How would filling the world with buggy substandard awful software make up for being an arrogant dick?
You could make a difference and make open-source software that sets the bar higher. After all, it's not like all superior software suddenly becomes substandard just by virtue of a free license slapped onto it.
-
You could make a difference and make open-source software that sets the bar higher.
Why?
Seriously. What's in it for me? What's my incentive?
You're asking me to do something I currently charge six digits a year for for free. You're going to have to have a pretty damned good reason.
-
If the thing is a four-digit number, yeah, it's pretty much a crap shoot as soon as I close the popup.
We are talking database schemas here, which means, basically, names for tables, columns, functions. Even in the worst databases I've worked with so far, these were quite descriptive, and in the second-worst, they were even predictable.
As for the numbers, I remember working some reporting software that involved construction projects (like roads, streets, stadiums — big shit). I've had so many of those "rounded sum of items' values/sum of rounded items' values" kinds of errors that after two weeks into it, I easily spotted whenever a 8-digit amount was off by a few dollars, or by a few cents. We had different customers who did their bookkeeping very differently; and for some, it was that their subcontractors handled that inconsistently. Yet the customers demanded a perfect match of the reports, no matter how fucked up.
Guess my memory must be good. But then again, what good a memory like this is when you want to forget these numbers and can't, and they haunt you in your sleep?
-
You're asking me to do something I currently charge six digits a year for for free. You're going to have to have a pretty damned good reason.
You sound like the kind of person who doesn't donate to charities because you aren't poor enough to get aid from a charity, completely missing the point.
-
You're asking me to do something I currently charge six digits a year for for free. You're going to have to have a pretty damned good reason.
The shit having its source in the open doesn't mean you need to do it for free. There are business models around that, like implementing solutions around OSS tools. If you open-source utilities you use for your own projects (loosely related), you gain if you have a contributing user base (protip: your utility should be useful to people other than yourself).I wouldn't equate "open source" with "shit" for the sole reason that Oracle and SAP exist. They cost a fortune and are utter shit nevertheless (like, they can sink your business very easily and you still end up owing them). I guess shittiness doesn't have anything to do with the development model.
-
To be fair, that would probably disqualify you from being employed as a database programmer or admin.
But SHOULD it?
If you can't do the job, then yes.
Being able to perform a cushy intellectual job isn't an inalienable human right.
-
The shit having its source in the open doesn't mean you need to do it for free.
Good point:
Red Hat:
Net income: US$ 178.29 million (2014)
Hmm...
-
He sounds like the kind of liar who says he makes 6 figures but barely makes 5.
-
-
We are talking database schemas here, which means, basically, names for tables, columns, functions.
Basically, I guess. But realistically also constraints and indices and the source behind functions and views and materialized views and probably other stuff that I don't usually deal with but maybe others do.
-
We are talking database schemas here, which means, basically, names for tables, columns, functions. Even in the worst databases I've worked with so far, these were quite descriptive, and in the second-worst, they were even predictable.
I once had to work with a university database in Oracle. I think the database was originally part of a third-party school management system. That thing had nearly 10,000 tables, all named various indecipherable things like
TBLCHEI3OorPK88LYMS. For the record, at the time the university had between 7,000 and 8,000 students. There's a reason I decline to interview every time they call me up and say "We still have a position available!"
-
Basically, I guess. But realistically also constraints and indices and the source behind functions and views and materialized views and probably other stuff that I don't usually deal with but maybe others do.
While we are talking about different management tools, psql displays them all if you \d table_name. You get to see all indices and related tables.
-
You sound like the kind of person who doesn't donate to charities because you aren't poor enough to get aid from a charity, completely missing the point.
No; I don't see how participating in open source makes the world better in any way. Most charities actually attempt that.
The shit having its source in the open doesn't mean you need to do it for free.
Ok; then you pay me for it.
I wouldn't equate "open source" with "shit" for the sole reason that Oracle and SAP exist. They cost a fortune and are utter shit nevertheless (like, they can sink your business very easily and you still end up owing them). I guess shittiness doesn't have anything to do with the development model.
Shit is stuff like:
- Release early, release often (i.e. expose end-users to your buggy untested shit)
- Do one thing and one thing well (i.e. don't actually solve any problems your users have, just solve little bitty subsets of those problems)
- "PRs welcome" (i.e. no contributions accepted from people who aren't able/willing to write code)
Sure a closed source project could do all of those things. It'd probably end up shit. Or it could end up shit for an entirely different reason, who knows.
But all open source projects do them. And they're actually proud! of some of them. Like release early, release often.
-
Shit is stuff like:* Release early, release often (i.e. expose end-users to your buggy untested shit)* Do one thing and one thing well (i.e. don't actually solve any problems your users have, just solve little bitty subsets of those problems)* "PRs welcome" (i.e. no contributions accepted from people who aren't able/willing to write code)
Oracle and SAP do neither. Still, they are.No, rather like that: they are SHIT.
Nope. I'd put it like this:
They are UTTER SHIT.
Now that's more like it.
