If there is "a winner for every language" in a code golf challenge, does competitiveness matter?

Question

The consistent attitude with regard to the golf language situation is that for every challenge, each language is a category, and subsequently there is a 'winner' for each language. This is also a very very popular opinion with regard to the new user experience.

There is a problem with this idea. With the exception of a very small percentage of languages, Most challenges have only one answer for each language. You might see where I'm going with this. If there is one winner for each language, and also one answer for each language, then by extension almost everyone is a winner. Anyone with a half-decent answer is never challenged.

By this logic, why is competitiveness important to the PPCG stack?
Does the reason expand beyond "The rules say so"? If not, then why are the rules discouraging competition?

After some confusion of definitions, a formal definition of competitiveness:

Possession of a strong desire to be more successful than others.

source

Answer 1

PPCG code-golf should not be considered an individual competition.

Because PPCG solutions are publicly displayed, many users peek at solutions by others in their language. Suppose Alice, a Jelly programmer, writes a 20-byte answer to some challenge. Bob glances at the byte count, thinks he can outgolf her, and finds an 18-byte solution. Finding it too similar to Alice's and being generous, he leaves it as a comment. Later, Carol studies Alice's answer as well as others' in different languages, and finds an efficient approach that solves it in 13 bytes.

All three users essentially had different playing fields and cannot compete fairly. Due to these asymmetries resulting from the structure of our site, PPCG code-golf is flawed as a competition (fair test of individual ability) and should not be thought of as such.

However, competitiveness (trying to outdo others) is very much in the spirit of PPCG. A competitive mindset is not by any means a requirement to contribute to PPCG, but it's an important motive for PPCG contributors. (Other motives include theoretical interest, improving one's code-golf skills, community, etc.)

Reasons why challenges often have only one answer in a given language

• They don't. If a question is popular and easily approachable, it likely has multiple answers in common production languages. For example Oreoorererereoo has 4 Python and 3 JS answers.

• The author of the first solution is likely to be fluent (to write it so fast) and very active (therefore experienced, therefore skilled), so a refinement of the first solution has a good chance of being the best.

• Each esolang has a small community. In Jelly (currently the best and most popular golfing language), 3-4 golfers combine to write over 80% of Jelly answers. In 05AB1E, Kevin Cruijssen writes the majority of answers, and in a specialized language like Canvas dzaima writes 80% of answers.

• In golfing languages, the solution is sometimes a few builtins. It's hard to golf down a 6-byte answer, and even if you do, it often feels like you haven't contributed enough to post the 4-byte answer as your own.

(I'm not sure of OP's exact question yet, but I tried to give my best answer).

Answer 2

why is competitiveness important?

It's not, really. We need some kind of scoring criterion, and going by byte count is simple, objective, and easy to measure in (almost) every language, but is still enough of a departure from the way code is normally written that it poses a unique challenge.

That most languages don't get more than one answer on a given question really doesn't matter. Trying to get a program as short as possible (or, in many cases, trying to simply write it in the first place) is entertaining regardless of whether there are other people for you to compete with.

If you're using a language which neither (1) has enough competition for your taste, nor (2) is inherently fun to golf in even without competition, might I suggest switching to a different one?

Answer 3

Competitiveness does matter

Anyone with a half-decent answer is never challenged.

Incorrect. If it's decent but not at the level of "very good", generally people will comment improvements, or (rarely) submit their own if they use a significantly different approach. Just because they don't have their own answer does not mean they aren't competing.

then by extension almost everyone is a winner

Incorrect, see above. Anyone that contributes to the best solution, or even comes up with an identical solution, is a winner. It's just that the Stack Exchange system is not the best indicator of this.

and also one answer for each language

For easier challenges, there is generally only one approach. For harder challenges, very, very few people attempt it because of the sheer amount of effort involved. Also keep in mind that certain people enjoy certain types of questions - kolmogorov-complexity and number challenges are generally very different, and generally require different tools/methods/a different way of thinking to golf.

Furthermore, consider the sheer number of languages of the submissions. PPCG is a relatively (very) small Stack Exchange community, and I'd say the number of active users is only several times larger than the number of active languages. Considering the fact that not everyone will attempt every challenge, I doubt there are more than a few people who golf in a certain language that look at any given question.

Answer 4

The consistent attitude with regard to the golf language situation is that for every challenge, each language is a category, and subsequently there is a 'winner' for each language. ... If there is one winner for each language, and also one answer for each language, then by extension almost everyone is a winner.

I agree with you that this interpretation is a problem, but there are more useful interpretations.

1. If you trawl through the early days of meta, I think you'll find various discussions about language handicaps, but all proposals for them were abandoned as unworkable. The concept that languages don't compete against each other may in part be a reaction against these proposals: handicaps are unnecessary because direct comparisons are unnecessary.

2. We require entries to make a serious attempt to optimise for the winning criterion. This is sometimes phrased as "entries must be competitive", and the concept that languages are incomparable helps to clarify that this doesn't automatically disqualify more verbose languages from code golf. (I suspect that you're thinking of suggesting a change to this rule, but I won't preempt).

3. I think it may also be used from time to time to justify not selecting an accepted answer. When the shortest answer is a one-byte built-in then IMO the question is poor, but some people have different attitudes or don't realise what built-ins exist and yet don't want to reward a 'no-effort" answer over one which took a lot of effort.