I can perform a completely independent experiments in my house.

And I can scream into the abyss, it's just as relevant. The absolute majority of actually useful and relevant science is performed socially for social purposes.

I make a hypothesis that my stove can boil 1L of water in 10 minutes.

You aren't even supposed to do a scientific experiment in the way you have just described. Or rather, there is neither a universally agreed upon scientific method, nor would your described experiment hold up to any standards.

An actual scientific experiment into water boiling would involve at the minimum

1. A model predicting the speed of boiling based on relevant variables
2. A collection of many data, and preferably corroborated by independent sources
3. Statistical analysis of the data (there are many methods to choose from) to gauge confidence in the model.
4. Publishing or proofreading of the results.

However, at each of these steps, you have a choice of how to approach the problem. And this depends on what you are trying to do, and what the best standards in the industry are. The process has also changed over time.

And this reveals the problem of many people's metaphysical approach to science. They treat it as if it were a platonic ideal, or floating constant in the human minds pace. In reality, "science" is an industry with its ever-changing standards, culture, interaction with the rest of society, and a million other complexities.