SQL now a dirty word for Oracle, at least in cloudy data warehouses Oracle has updated its cloudy data warehouse and made structured query language harder to find, in the name of having database administrators spend less time working with the product. If that sounds like a bizzarro move for a software giant literally built on SQL, and which has for years courted DBAs, know that Oracle’s …
Wide Area No Knowledge Access (W. A. N. K. A.) where massed middle managers play with data all day to try and improve their carrer prospects. I would imagine that DBAs will be fully employed tydying up assorted tangled messes whilst crazy "non query" queries drag everything into the gutter.
Look at the mess that is PowerBI. Data "analysis" without understanding the processes that create the data, statistics, statistical relationships, basic data visualisation concepts...
This oracle product sounds very similar, and is not really welcome. Instead of us finding meaningful relationships and changes in the data, we now have to tell mgmt that they have no clue and their analysis is plain wrong. Thank you oh so bloody much!
Not that anybody will listen to people who have a clue. Oh no, their results are so convincing (and the statistics are... not supporting their message, and neither is there a process that can cause any causality they think they discovered)
I need a drink...
Akchley, yes, one important lesson was learned. The middle manager of doom is stupid precisely because they believe that their tinkering is valuable instead of costly.. The cloud is running other people's workloads at profit. If the other people are stupidly inefficient, so much the better. You can kid them that the volume discounts they get for their inefficiency will save them.
Now the middle manager of stupid will run his spreadsheet queries on an unlimited serverless cloud, paying per instruction instead of his hardware limited over-specc'd laptop...
If only it were just middle managers.
Oracle is promoting low-code (soon to be followed by no-code I fear) to everyone. They phrase it as 'line-of-business' people it's destined for anybody with a bright' idea that they think would make their job easier. Right until it doesn't work and they complain to IT.
I'm a Project Manager and I hate these products as much as most DBA's.
The naive DBA's who relish receiving fewer calls from marketing will get a brief respite before everything blows up in their faces.
What will come out of this will be sets of dashboards and reports that the business rely on to make strategic and tactical decisions/
As these become more complex they'll take up more and more CPU cycles and take longer and longer to run until a DBA is asked to 'just tune them'.
Once the DBA looks at the half dozen queries generated under the covers by the BI product and optimized them what will come out will be a completely different set of results which will then need to be explained. Of course the person using the BI product will swear that their results are corrupt and the 20 year experienced DBA has got it wrong.
This will be followed by 3 months of strife with meeting after meeting being called to try and find a way to provide the original results before finally someone is fired/ moved / demoted
The victim will vary depending of the size of the organisation / seniority of both partners and the impact on the share price of publicizing the fact that the sales forecasts for next year are accidentally inflated by 130%
This sounds so familiar. At one employer I fixed several reporting systems that had been producing incorrect statistics, and explained to the sales and management why the data had been incorrect. But the incorrect data had suggested much healthier figures that were reported to existing and potential clients. Management bonuses were also based on the false figures. So I was ordered to break the system so sales could continue lying to the clients and managers could get their undeserved bonuses. Suffice to say that was a reflection of company culture in general, so I no longer work there.
"... the incorrect data had suggested much healthier figures"
This is the prevalent position in risk management (where 'assessment' is done by pure guesswork), and increasingly in science, where researchers use statistical tools to apply methods they don't understand to their data and publish the results on trust. In both cases, a lot of superficially convincing complete nonsense results. Clearly, it's a cultural problem, not one limited to "low code".
"I have the same complaint regarding all these "easy to get started" languages"
It's clear to me over the 30+ years I've been developing software that successive new languages and dev systems have made it progressively easier for people who don't understand what they're doing to produce executables they don't understand either.
Awareness and application of essential first principles underpin all other engineering disciplines, but in this one the dominant state is lack of awareness that first principles even exist. Of course they do, and their application would assist in improving the parlous standard of most software
Sales Manager explains the report he wants, all the inputs and the data set to work on. I go and write some real SQL to run the queries required. "No, that's not the answer I want" he says. After a few more goes at this he decides is easier to be vague about the inputs and the data, and just ask for the result he wants. I stuck it out for a couple more months before leaving, somehow they stayed in business for nearly a couple more years.
From "easy to read,write and understand" code in COBOL to SQL itself, history has shown the manglement and PHB classes are completely incapable of writing useful anything and worse, regard such mundane simple things are for the peasants, not important process drones like themselves.
Remember all those tools that sprouted decades ago that were supposed to make programming redundant? Didn't happen did it. This one is almost guaranteed to go the same way, largely because it's a Larry lucre levitator rather than an attempt at a genuine business productivity tool.
Barrier? What barrier? Low-barrier programming actually means "any idiot who can both edit text and invoke a compiler."
Right now I am working with the result of what looks like a CLIP+BigGAN AI wrote the code. However, it is 100% human generated. To produce a "working" program, all you need is time. And then somebody has to clean up.
These high level BI tools are too generic, presuming that all businesses can fit their data to the tool's way of seeing things. Unfortunately most businesses have a good percentage of their activity that simply can't be analysed or visualised using them. So they always reach a point of failure where the manager wanting info says 'but why can't it do xxx...'
This:
"whilst crazy "non query" queries drag everything into the gutter."
and
"As these become more complex they'll take up more and more CPU cycles and take longer and longer to run ..."
Honestly, I may be cynical (OK, yes I am), but I'm assuming this is part of the end goal of discouraging direct use of SQL -- unoptimized and downright ineffcient queries drag down database performance; more database capacity needed! So then they'll either buy more cloud capacity (if it's the cloudy style) or if it's on-premises, they'll either be installing a few extra DB servers (additional copies of Oracle), or Oracle has been installing per-core fees for years so replacing a DB server with a beefier one will also mean additional fees to Oracle.
"unoptimized and downright ineffcient queries drag down database performance; more database capacity needed!"
It isn't just the poor quality of queries that will drive sales, the sheer quantity when every single middle manager is encouraged to crank out a wall of dashboards will drive up demand.
Did you actually mean to say the opaque and propreitary SQL needed by DBA's to tune the system has been hidden away behind the scenes, or did you really mean to say access to SQL in general has been hidden? If so it might have been helpful to be more clear whats replaced it, because from reading the article I dont have a clue.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
