Incoming links: Novel result
People will be more interested if they see how the stuff they are learning is useful.
Useful 99% of the time means you can make money with it.
Achieving novel results for science, or charitable goals (e.g. creating novel tutorials) are also equaly valid. Note that those also imply you being able to make a living out of something, just that you will be getting donations and not become infinitey rich. and that is fine.
Projects don't need of course to reach the level of novel result. But they must at least aim at moving towards that.
This is one of the greatest challenges of education, since a huge part of the useful information is locked under enterprise or military secrecy, or even open academic incomprehensibility, making it nearly to impossible for the front-line educators to actually find and teach real use cases.
The projects you do must always aim to achieving some novel result.
You don't have to necessarily reach it. But you must aim for it.
Novel result can be taken broadly.
E.g., a new tutorial that explains something in a way never done before is novel.
But there must be something to your project that has never been done before.
You can start by reproducing other's work.
E.g. International Mathematical Olympiad, International Physics Olympiad, competitive programming, etc.
Events that trick young kids into thinking that they are making progress, but only serve to distract them from what really matters, which is to dominate a state of the art as fast as possible, contact researches in the area, and publish truly novel results.
Financially backed by high schools trying to make ads showing how they will turn your kids into geniuses (but also passionate teachers who fell into this hellish system), or companies who hire machines rather than entrepreneurs.
The most triggering thing possible is when programming competitions don't release their benchmarks as open source software afterwards: at least like that they might help someone to solve their real world problems. Maybe.
On a related note, hackathons are also mostly useless. Instead of announcing a hackathon, just announce a web forum where people with similar interests can talk to one another instead, and let them code it out on GitHub if they want to. Restricting intensive development to a few days tends to produce crappy code and not reach real goals.
Some irrelevant people highlight that knowledge Olympiads can have good effects, because they are "an opportunity to meet university teachers and their research organizations". Ciro's argument is just that there are much more efficient ways to achieve those goals.
As an alternative way to get into university, this is not 100% horrible however, e.g. the University of São Paulo accepted students from olympiads in 2019 and then again 2023: jornal.usp.br/institucional/usp-oferece-200-vagas-em-mais-de-100-cursos-de-graduacao-para-alunos-participantes-de-olimpiadas-do-conhecimento/?a
The cool think about Ramanujan is how British mathematicians heard about him and then just went crazy that someone they had never heard of before had come up with so many novel results. It is as if God had come down from the clouds and handed them those results. Without proof. But in that field of research, the statements are not easy to come up with, so much so that G. H. Hardy commented that:
they must be true, because, if they were not true, no one would have the imagination to invent them
