Excel is still the biggest and most successful no-code tool. The real problem with no code, is the tools take a lot to learn well and it rarely transfers between roles at the same job.
As far as general "low code" or "no code", I've seen that promise for many decades. They keep reinventing failed ideas under new names. Clickable wizards, rule lists, and drag-able flow-charts can only take you so far. The devil is in the details, and these don't handle details well. And they are hard to factor.
Your Uncle Bob may be able remodel your house for cheap, but what do you do when rain leaks in and the lights stop working? Or worse, catch on fire?
Remember: Total Cost of Ownership.
He started of the talk describing Regex, and how it came to be, and basically showed how that functions have been/could be formally described in papers the same way that Regex is used.
In the same way regex formalized the notation for string parsing, so too will the next language regular function expression formalize. Perhaps computer meta notation is the source of true no-code solutions? Perhaps instead of reading articles about "The Unreasonable Effectiveness of Regex," in the future, we'll be reading about "The Unreasonable Effectiveness of Computer Science Metanotation."
I also understand that what no-code solutions we are typically talking about is a user interface that is graphical, drag-and drop. But who is to say we don't create a tool that does that with CSM?
[1]: https://www.youtube.com/watch?v=qNPlDnX6Mio&ab_channel=Erlan...
Someone once said that D/F makes it easier to express what you intend, but imperative makes it easier to figure out what's actually happening. One improves writing, the other improves reading (or at least debugging).
What's interesting is that many users build quite complex formulas and spreadsheets, yet Excel's formula and VBA syntax is (in my opinion) harder to learn than other programming languages. For example, I'd argue a language like Ruby is simpler in many aspects than using VBA or building complex formulas.
Like Microsoft, Google Sheets also has quite an ugly, clumsy feel to its syntax.
The terrible programmable usability of these products (Excel and Sheets) could only have come from the mind of a developer (harsh I know, but true).
People just kind of default to {do some magic} on [various things] -> (Some result that looks about like I was expecting).
What’s really interesting is Spreadsheets tend to have bugs where the result is incorrect but reasonably close to reality. Aka 38 vs 35 not Null Pointer Exception, this makes them in many ways seductively dangerous in a this looks close therefore it must be easy to fix way.
Yep, people sure write a lot of code in Excel.
(Well, what do you think formulas are?)
=RIGHT(A5, SEARCH("FOO",A5))="BAR"
So, great lock-in capabilities? I can see why money is pouring into these tools.
Edit: I know there are lots of people on the Internet saying you can simulate Turing machines in excel, but they're usually manually unrolling the computation in successive rows, which doesn't really count. For one thing, this style of simulation doesn't even allow one to make an infinite loop, let alone a more complex self-contained loop.
That and Microsoft's insane theory that about 50% of user input is dates.
And eventually require VBScript if you want to really automate anything anyway.