Backlog, scope, and deadlines.
- Backlog:
>When I think about things they are all sounds interesting to me, especially if they are connected to technology.
Writing ideas down with a timestamp "frees" up your head, reduces overwhelm, and makes ideas that feel grandiose in the fog of your mind look puny. Writing down ideas reveals the adjacent possible, the most urgent, and the most actionable.
I use taskwarrior with a "+musing" tag. Many times some ideas will morph and get combined with others, and make their way into product.
Writing ideas down allows you to be lazily productive and gives you perspective.
- Scope:
>I have a lot of books and articles to read, topics to explore, projects to finish but every time when I'm trying to dig into it I end up feeling tired and resigned.
You are overwhelmed. Depending on context, some times call for having one book or one article to read, one topic to explore, one project to finish.
Other times benefit from having five books open and cross referencing/checking/complementing.
It appears you need at least one win and so having one book to read one page at a time may help. The scope is one page, then another, then another. That said, I wouldn't feel guilty about dropping a book like a hot rock. I will not waste life on a lousy book I started reading for the sake of discipline. Not all books are equal, that's why I curate. I get references to content from content I respect and like, not from a ready-made best selling list. This increases the odds of reading books I start.
- Deadlines:
>I've tried ti blocked time-wasting pages (like 9gag/reddit/etc.)
Given two processes competing for the same reagents or resources, making one of them efficient can make it consume resources faster than the other. One resource is time, one process is productive, the other is unproductive; both compete for time.
Deadlines increase focus to the point you will not need to force yourself to avoid distractions, you will simply not have enough time because you spent it on more useful things.
Sometimes I'll say: this hour, I'll read this article and will do this only. I'll refactor this sub-system by end-of day. I'll increase code coverage by X% in one day.
I find it good to set agressive deadlines. Whenever I say "it'll probably take X" I usually halve it and feel a "mini-panic" that shows I'm not confident I can do it by that deadline. Not feeling a mini-panic means I'm comfortable with the allocated time, which means I'm either clueless and can't estimate, or that I'm getting complacent.