From your formulations later in this thread about your unease and unhappiness of the development of the discussion I formulate in a handful points below what I say at community writing events to hopefully give you a better feeling for the next thread.
Even if you yourself are not unsure about what you want to achieve by starting a new thread, consider:
* Start by formulating it offline, to give you ample time to revise
* Focus on formulating one single line of thought as clearly as possible
* Don’t speak for others, at all
* Assume even less, especially about others’ intentions
* Avoid using more strong words than necessary, rather remove one full sentence instead of keeping an uncertain formulation
* Be fair on citing
Then, if the title is not easy to formulate from what you written ask someone what a good title would be from reading your text.
You can disagree with attah, but personal attacks are not acceptable. This isn’t the first time you’ve resorted to insults, and this kind of behavior has no place in a constructive discussion. Keep the discussion focused on the argument, not the person.
As for your claim, it appears you misunderstood what was actually said in your Sail Recorder thread. Nowhere do I see attah labeling you a “hacker”. His criticism regarding the undisclosed use of AI and the lack of proper source code was also valid.
What exactly are those “past words”? Please quote the specific statements you are referring to. So far, all I’ve seen are vague accusations.
With this kind of attitude, you are contributing to the very toxicity you are accusing others of creating. If your goal is to address toxic behavior within the community, resorting to inflammatory language and broad accusations is hardly helping your case.
One thing that bothers me about this vibecoding phenomenon and the discussions around it is that many don’t seem to care as long as an app "does the thing”.
To use an analogy, this is a bit like saying that a house is fine as long as it has walls and a roof and you can live in it. The thing is, I wouldn’t live in a house that may or may not be structurally sound.
This narrow way of looking a piece of software brings to mind Thinking Fast and Slow by Danny Kahneman:
“To explain overconfidence, Kahneman introduces the concept he terms What You See Is All There Is (WYSIATI). This theory states that when the mind makes decisions, it deals primarily with Known Knowns, phenomena it has observed already. It rarely considers Known Unknowns, phenomena that it knows to be relevant but about which it does not have information. Finally it appears oblivious to the possibility of Unknown Unknowns, unknown phenomena of unknown relevance.” ( https://en.wikipedia.org/wiki/Thinking,\_Fast_and_Slow#Optimism_and_loss_aversion )
I’m not a developer, but I use software. I’m not an architect but I live in a house. I know that in both cases I have to trust the maker, their motivation and their skills. Vibecoding has killed this trust system by allowing anyone to make things that may look fine right now, but there’s no guarantee that it’s been made with care to be secure now and in the future. This is a “Known Unknown” if literally no-one has reviewed the code. For this reason, I would never use a vibecoded browser, for instance.
Of course there’s no guarantee for non-vibecoded software either, but previously the piece of software wouldn’t have come about at all if there’s no dev. Open source does allow those who know coding languages to go in and see how the code is, but for something like a browser that’s a huge task.
Personally I’m very worried of all the slop that’s being put out there and I think a good middle ground would be a AI disclosure policy so that everyone can make the choice to use or not to use an app for themselves. This has been talked about before and I think there’s something in the works that’d allow disclosures in chum at least, which is a great step!
Let me rephrase, because being too abrasive does not fit the pseudo-cozy nature of this “community”
I was called a moron by two people, also accused of not contributing anything to this community by three more. Apparently you can call people names if the mods like you and you have “friends” to back you up - please stop lying then.
Edit: another brigade I’m starting to love this. It seems facts are something some people really dislike.
I’m honestly sorry that I tried to bring some perspective to this. Like I said, when a new developer creates an app with the help of AI and then publishes the code, it’s great when experienced developers contribute to the code. That’s how collaboration works in a community. But this thread no longer has anything to do with the original topic. All we’re doing here is arguing—just like kids in a sandbox fighting over a shovel. So I’ll leave it at that. It’s pointless. As far as I’m concerned, the post can be deleted. The facts and the message have simply been overshadowed by misplaced egos.
I could add that not only the vibe code is untested, it is also probably “bad” in ways we cannot find easily.
For a house, you have a process to check if its sound before buying. For a vibe-coded black box, we are not ready yet to follow a simple check list for quality control.
We need clear sand boxing, and we need a trusted way to check the code.
Until then it should be labelled accordingly so that users know to be careful.
You are not a farmer, but you eat store-purchased food. You are not a doctor or herbalist, but take vitamins, painkillers, use herbal teas, spice, whatever.
Since AI does not exist (LLM, a stochastic machine is not AI), there is nothing to disclose But if I assume by “AI” you mean “LLM”: do I have to disclose I use IntelliSense for code suggestions in Visual Studio? Do I have to disclose I use code blocks suggestions in IntelliJ? Do I have to disclose that I used “Refactor” command that uses language model in its implementation? Do I have to disclose that I prompted the generation of my app description? Do I have to disclose that I told ChatGPT or Claude “make me an app”? Those are distinct things.
Regarding “proper software authorship”, if the requirement is that software must be “100000% professional” and unquestionably safe to use, then surely answers the following should be available:
Where is the formal Software Requirements Specification?
Where are the documented functional and non-functional requirements?
Where is the architecture specification and system architecture diagram?
Where is the complete test plan?
Where are the test cases, test steps, expected results and actual results?
Where are the test execution logs?
Where are the regression and compatibility test reports?
Where are the automated test results and code coverage reports?
Where is the threat model and formal security risk assessment?
Where are the penetration-test and vulnerability-scan results?
Where is the Software Bill of Materials listing every dependency?
Where are the static-analysis and dynamic-analysis reports?
Where is the formal code-review record, including reviewers and resolved findings?
Where are the documented qualifications, diplomas and certifications of the developers?
Where are the documented qualifications of the people performing the testing and security review?
Where is the independent verification and validation report?
Where is the documented release, rollback and incident-response procedure?
Where is the reproducible-build evidence and cryptographic verification of release artifacts?
Where is the formal risk register, including mitigation and residual-risk acceptance?
Where is the independent third-party certification confirming that all of the above is correct?
Because naturally, if a hobbyist application shared voluntarily with users who are entirely free not to install it is going to be judged by “professional software” standards, then I assume we are applying those standards properly rather than selectively invoking whichever requirements are convenient at the time.
A house analogy is stupid, see my other post. I don’t trust apps released by some random people more than I trust software written by LLM. For my own apps, at least I know whats not working for me
But if I assume by “AI” you mean “LLM”: do I have to disclose I use IntelliSense for code suggestions in Visual Studio? Do I have to disclose I use code blocks suggestions in IntelliJ? Do I have to disclose that I used “Refactor” command that uses language model in its implementation? Do I have to disclose that I prompted the generation of my app description? Do I have to disclose that I told ChatGPT or Claude “make me an app”? Those are distinct things.