How to deal with large inseparable stories in SCRUM?

Question

Time Management

Good afternoon!
I am trying to understand and, if possible, implement "SCRUM for one" for my projects, which are not always informational. And the question arose: how can you write a story in the Product backlog if the results are achieved in a time exceeding the duration of the sprint and achieved through the regularity of execution, and not the total time spent on the solution? For example, learning to play the guitar, achieving some success in sports (lose weight, get yourself in optimal shape). Allocating an hour a day or two, all this is achieved, let's say, in three or four months. But how can this be reflected in the SCRUM methodology?
If we translate this into an informational plane, then there will be something like "get a history of user requests for four months of the application's operation." And how to work with such a story? How to correctly calculate the amount of work in story points for her?

Answer 1

It is necessary to somehow divide into subtasks that fit into the iteration. If you don’t know exactly how to do this, then you need to dive deeper into the subject area and figure out how it works in detail, or abandon SCRUM.
"Learning to play the guitar" - 1) train your fingers 2) learn basic chords 3) learn a specific melody. Well, or quite banal: lesson 1, lesson 2, lesson 3 (if there is a textbook, then it has already been done for you).
"Lose weight" is the same. 1) Drop 1.5kg 2) Drop 1.5kg 3) Drop 1.5kg. If there are stages in the method of losing weight, then use them.
As mentioned above, you can opt out of SCRUM. I hope you understand that it cannot be shoved everywhere indiscriminately. There are cases where SCRUM works very poorly, and another method goes with a bang. The same training and fasting examples do not require any scaram.
Establish saving the history of user requests in the database. (1 day)
Or
Download and analyze the history of user requests for the last 4 months. (1 day)

Answer 2

You are confusing tasks with processes and results.
This is not a design and development task. It's just some kind of repetitive process that works, the tasks will be "make it possible to record the request history" "set up recording on the production server" "check in a week that everything is ok" "set up a backup of the request log" "make a history viewer" and so on.
Name me any of them that will last all four months?
Always ask the question "what needs to be done for this?" and you will have smaller things.
Even if you can't think of - they are.
Like "pull the same string for two hours" today. Tomorrow will be the same task, but different.
Everything breaks down to the level of "press A" then "press B"