Comparing previous years, they're exactly what I'd expect, to be honest. Only people serious about completion will...well...complete it. Even if they do not know any code, if you pick something well-documented like Python or whatever, it should not be a tremendous challenge so long as you have the drive to finish the event. Code isn't exactly magic, though it does require some problem-solving and dedication. Since this is a self-paced event that does not offer any sort of immediate reward for completion, most people will drop out due to limited bandwidth needing to be devoted to everything else in their lives. That versus, say, a college course where you paid to be there and the grade counts toward your degree; there's simply more at stake when it comes to completing the course.
But, speaking to the original question as to the number of newbies that go all the way, I'd say one cannot expect to increase their skills in anything if one sticks in their comfort zone. It should be hard, and as a newbie who participated in previous years, I can confirm it often is. But I learned new things every time I did it, even if I did not finish.
I have to say, I've read many out-of-touch comments on HN over the years but this is definitely among the most out there, borderline delusional comments I've ever seen!
The idea that anyone who doesn't know any code would:
1) Complete in Advent of Code at all.
2) Complete a single part of a single problem.
let alone, complete the whole thing without it being a "tremendous challenge"...
is so completely laughable it makes me question whether you live on the same planet as the rest of us here.
Getting a person who has never coded to write a basic sort algorithm (i.e. bubble sort) is already basically impossible. I work with highly talented non coder co-workers who all attended tier-1 universities (e.g. Oxford, Harvard, Stanford) but for finance/business related degrees, I cannot get them to write while/foreach loops in Python, and simply using Claude Code is way too much for them.
If you are even fully completing one Advent of Code problem, you are in the top 0.1% of coders, completing all of them puts you in the top 0.001%.
I can't begin to describe how valuable your input has been through this whole thread about something you're quite possessive and passionate about, which surely places you in a position to aggressively dismiss any other possible way of looking at it! Wow, love learning about new perspectives on HN!
Wishing you best of luck in AoC, Life and Love but I imagine someone like you doesn't need it, being a complete toolbox and all.
P.S.: Tell your coworkers I'm sorry they have to put up with you.
You're the person saying Advent of Code is "so easy" that anyone even people with no coding ability at all should find it do-able, which is totally diminishing the difficulty of the problems, and asserting your own genius, i.e. that you found it totally trivial.
I am the person saying that actually, stuff like Advent of Code is incredibly difficult and 99% of active programmers aren't able to complete it, let alone people who don't code.
I am not an elitist at all, unlike yourself, I don't find completing "Advent of Code" easy, in fact, it would take me a long time to complete it, more time than I have available in my busy life in the average December. And I doubt I would be able to complete it 100% without looking up help, getting hints, or using LLMs to help.
You clearly didn't read my whole original comment before mouthing off. Go back and do that, you'll find that I pointed out most do not complete it, that it is supposed to be challenging and I never called it "easy" as you imply ("not tremendously difficult" =/= "easy")
Heck, I even talked about having to be serious about completion, and you could not bother to read the whole comment, then proceed to call me delusional? FFS, I am now praying for your co-workers and I'm not even religious.
Did YOU even read your original comment? You asserted that people who have never coded could complete the event!
Did you realize only roughly 500 people of the > 1M who are registered for advent of code even complete it?
You said "it should not be a tremendous challenge", i.e. not that big of a deal even if you don't know how to code. Which is absolutely diminishing the difficulty of the event, I mean, come on man...
This is why I'm asserting you are quietly oblivious to the abilities of most people. I am asserting that most people who CAN code, cannot complete the event, yet alone non-coders. I am a very active coder (for fun mostly these days, but also sometimes for work), but I could not complete Advent of Code. Maybe if I took all of December off work to dedicate serious time, but even then I wonder if it's possible without looking at hints/LLM-help etc.
I often try and help my co-workers who are working on AI based side-projects for fun, so I have a strong insight into the abilities of non-coding smart people, and the reality is that yes, they get very turned off as soon as you get anything more complex than for-loops and if-statements. This isn't me being mean to co-workers, this is the reality of things I have experienced. It's not a brains thing, they can understand more complex stuff, but they don't want to, they find it annoying, boring, not worth the time/effort etc. So the idea of them learning dynamic programming, DFS/BFS, more complex data structures etc, is well, just not going to happen.
My point is that you are effectively saying, "oh just about anyone can do Advent of Code if they want to", is totally not grounded in any sort of reality.
The amount of injected implication you are imposing on everything I said...this is some seriously unhinged gaslighting in effort to obfuscate the fact that you came out of the gate calling someone delusional over a comment you barely understood. We're wasting each other's time, so I'm out.
Right, but a counter point is the etherium fork. Only a handful of people stayed on the “classic” chain after that first DAO turned out to have a massive extraction bug in it.
It’s a privately owned company. This leads to an entirely different relationship between employees and the top layer of management.
You have to be very misguided to believe that the c suite in most companies is not engaged in n adversarial relationship with its employees, whether those employees are unionized or not.
I’m sorry but there is no way you can demonstrate a universal aesthetic. Your opinion of other people’s tastes does not reflect on their taste — it reflects on yours.
> I’m sorry but there is no way you can demonstrate a universal aesthetic.
What do you mean by "aesthetic", because I've already made the distinction between objective beauty and subjective taste. If my explanation is true, then it follows that there is an objective ordering of beauty (of at least two kinds: with respect to the same form/end, and between forms and ends). Then, there's the question of how competent someone is at recognizing this order. And finally, there are contingent factors that will affect expressed volitional preference as a function of factors like attainability or character flaws or whatever.
Making beauty a matter of purely subjective response makes it more mysterious and nonsensical, not less.
> Your opinion of other people’s tastes does not reflect on their taste — it reflects on yours.
How do you know this? You haven't demonstrated this claim. I've at least explained the basis for mine.
I claim that on the contrary, yes I can. I can claim that someone who thinks rape or murder are beautiful has objectively deranged tastes, because these acts are intrinsically ugly.
Definitely there has been a shift towards video content.
But I think there is a discoverability dynamic at play as well. Finding a blog post that isn’t garbage can be harder than finding a decent video on any given topic. This is clearly a feedback loop towards video but I think partly this is because up until now you couldn’t just create a spam video channel with the same ease that you could a spam blog.
My intuition is that if there’s no 2 dollar coin to go with it, there’s no way for it to gain practical traction.
In the US, change is already an annoying factor because sales tax is rarely included in eg, 4.99. So no one is jumping up and down to go from five slices of paper to five rattling coins.
Python is more widely known and a very popular tool for “LLM engineering”, so I’m curious what would be the reason to choose Raku in this case and wondering how the feature benefits of Raku outweigh the general incentive to use more popular tools.
Mostly, because Python is not a good a "discovery" and prototyping language. It is like that by design -- Guido Van Rossum decided that TMTOWTDI is counter-productive.
Another point, which could have mentioned in my previous response -- Raku has more elegant and easy to use asynchronous computations framework.
IMO, Python's introspection matches that Raku's introspection.
Some argue that Python's LLM packages are more and better than Raku's. I agree on the "more" part. I am not sure about the "better" part:
- Generally speaking, different people prefer decomposing computations in a different way.
- When few years ago I re-implemented Raku's LLM packages in Python, Python did not have equally convenient packages.
Ah - sorry. I plead guilty to trying to add some controversy to this thread to try and get a debate going - and thus maybe achieve the HN front page listing. I thought that Anton would be OK with this level of spice. OTOH I do see that it was rather unkind, so in hindsight I should have stuck to the motto "be kind".
For me, it is a mystery how programmers "decide" as a group that a new tool or language is better than the established & familiar ones. But I would love to see more folks open to try new tools like Raku that could make their lives easier and more fun.
reply