Feedback on Coding Challenge Number Three

Personal opinion on Challenge #3

I feel like this is not one but two challenges, being "Generating a snowflake from a seed" and "Displaying a snowflake using ASCII art".

I'm not a huge fan of this aspect because it makes it unclear what is the most important for the challenge: is it the beautiful-ness of the ASCII art, or the accuracy and complexity of the snowflakes? Or equally both?

Other than that, the theme is fine in my opinion, but leaves far less room for originality than Challenge #2. I'm afraid that all "good" snowflake generator will end up looking mostly the same.

I'm also afraid that some people assume ASCII art means drawing with any character in a terminal, even outside the ASCII table itself, as it became so common to see block elements in CLI apps, and to see characters like ò, ó, ... in terminals, so even though the instructions clearly say "only ASCII printable characters", I think it'd be nice to add "no post-ASCII Unicode characters" or something like that. Even Wikipedia has some post-ASCII characters as examples of ASCII art.

I'm also afraid that some people will also use terminal colors, which would also go against the spirit of ASCII art in my opinion.

AI usage instructions are still unclear

Under the Challenge #2 meta post, I have posted this answer, which reads:

Your entry is not permitted to be written, in full or in part, by AI. AI assistance with coding or debugging is permitted if it is disclosed in your entry and the initial code is wholly your own.

I'm not sure I understand: "Your entry is not permitted to be written, in full or in part, by AI." (emphasises mine), meaning essentially that no code in my entry can come from an AI, even if it's just a small part of my codebase.

But, "AI assistance with coding or debugging is permitted" (emphasises mine), so I am allowed to have some of my code written by AI.

Wait, so am I allowed to use AI-written code or not? Don't get me wrong, I'm NEVER gonna use AI to code, but I wouldn't be surprised if I wasn't the only one to be confused by these two sentences that contradict each other.

A staff member replied:

You cannot use AI to write your code but you can use AI assistance in debugging. We tried to make the text much clearer in this challenge than the last but there's always room for improvement!

So AI cannot write code for our entries in the challenges, but it can help for debugging, all right.

So then, why does the Challenge #3 still reads (emphasis mine):

Your entry is not permitted to be written, in full or in part, by AI. AI assistance with coding or debugging is permitted if it is disclosed in your entry and the initial code is wholly your own.

So, let me ask again: is AI-written code allowed? AI not being allowed to write our entry, even in part, goes directly against AI being allowed to assist with coding. Please make it clear, not just as a comment under this answer, but the post itself

Taking previous feedback into account

Most of the answers that have been addressed so far are those that ask for clarification on challenges' instructions.

This leaves very important pieces of feedback completely ignored:

• General Grievance's answer about the post length limit and broken spam filter

• Aemyl's answer about the lack of real-time previews for posts and lack of syntax highlighting

• Daniel T's answer about images being forbidden without prior notices and Efraim Newman's answer about allowing images

• My, Roddy of the Frozen Peas' and burnie's answers which mention problems about the voting system (Roddy's question received a reply which can essentially be reduced to "we didn't want to put effort into it because it's an experiment that might fail", which, in my opinion, is not acceptable and only makes it more likely for the experiment to fail), which is clearly shown by the huge lack of votes in Challenge #2 compared to Challenge #1 even though it received more entries

• YungDeiza's answer about self-undeletion

Those are very significant pieces of feedback, the fact that these haven't been taken into account makes me think that SO doesn't want this experiment to succeed. How can you expect it to succeed if you don't fix its problems when the community tells you about them?

^{Related: Roddy of the Frozen Peas's comment on this question}

Subjectivity of "How do I win?" criterias

We realize this is not the most objective criteria; in the future, we plan to have a more sophisticated evaluation system!

This made sense for Challenge #1. It was acceptable for Challenge #2. Now we're at Challenge #3, but we still have no idea of when the "future" will be. You yourself admit that something about your experiment isn't perfect, and say you will make it better, but you don't, even on the third challenge, which is the penultimate one of the experimentation phase. How can the experiment succeed with such a mindset?