I already see lots of boutique consultancies popping up specifically saying they will fix tour slop code bases. Developers and doctors will never run out of work.
They will have to trash a bunch of it and rewrite entire modules and components from scratch.
I’ve seen as AI keeps iterating over enhancements and bug fixes, the spaghetti code gets worse and worse over time. At some point only the AI understands its own code, as it is too much garbled nonsense for a human to trace through. Eventually that becomes too much garbage for the ai to trace through. And then that’s where you end up.
The idea that this can be some kind of non-deterministic abstraction layer between pseudo code and actual code is absurd.
The tech debt always gets paid one way or another. Either preemptively, as part of ongoing maintenance; or after a data breach in the subsequent lawsuit and settlement; or in the slow, inexorable trickle of increased infrastructure costs and lost business from slow dependencies and ineffectual bandaid solutions.
Of course but the cost can be mitigated. I’ve seen an AI built project and the outcome is that you pretty much have to continue using AI to continue development because it’s borderline incomprehensible for humans. When AI fails to continue development you scrap the entire project and start from scratch.
I imagine all companies that allowed AI to go wild in their codebase have a lot of components that need to be rebuilt from the ground up.
In theory, completely rewriting legacy software with AI could be a great thing - doing so every 3 or 4 years, allowing newer and better AI to tackle the task, optimizing for every development made in technology or standards.
However, this still requires human minds to define, guide, and verify these projects. AI opens up possibilities for better development, but executives seem determined to fritter away such potential by axing the human element.
Just wait until all the technical debt has to be paid as well.
I already see lots of boutique consultancies popping up specifically saying they will fix tour slop code bases. Developers and doctors will never run out of work.
I’m not sure fixing it will be possible.
They will have to trash a bunch of it and rewrite entire modules and components from scratch.
I’ve seen as AI keeps iterating over enhancements and bug fixes, the spaghetti code gets worse and worse over time. At some point only the AI understands its own code, as it is too much garbled nonsense for a human to trace through. Eventually that becomes too much garbage for the ai to trace through. And then that’s where you end up.
The idea that this can be some kind of non-deterministic abstraction layer between pseudo code and actual code is absurd.
I know what you mean, but the tech debt problem will never get resolved.
The tech debt always gets paid one way or another. Either preemptively, as part of ongoing maintenance; or after a data breach in the subsequent lawsuit and settlement; or in the slow, inexorable trickle of increased infrastructure costs and lost business from slow dependencies and ineffectual bandaid solutions.
Of course but the cost can be mitigated. I’ve seen an AI built project and the outcome is that you pretty much have to continue using AI to continue development because it’s borderline incomprehensible for humans. When AI fails to continue development you scrap the entire project and start from scratch.
I imagine all companies that allowed AI to go wild in their codebase have a lot of components that need to be rebuilt from the ground up.
In theory, completely rewriting legacy software with AI could be a great thing - doing so every 3 or 4 years, allowing newer and better AI to tackle the task, optimizing for every development made in technology or standards.
However, this still requires human minds to define, guide, and verify these projects. AI opens up possibilities for better development, but executives seem determined to fritter away such potential by axing the human element.
It is like a certain story…
You forgot that AI cannot “rewrite legacy software”.
Otherwise it’s a nice thing to fantasize about 😁
And people realize they don’t understand the output anymore or the architecture. AI isn’t creating anything new either.