There's the lightweight approach of just deleting all your bookmarks and trusting you will be able to answer any future question for yourself. This is similar to choosing to live without a long-term todo list, trusting that anything that needs attention will present itself to your attention.
The latter has never really worked for me, so the heavyweight approach is to have a why-justified system. If you're going to save a bookmark, it had better be referred to in some checklist or documentation attached to a long-term project that you are actively working on, and that you've sufficiently justified as being a necessary part of your long term goals or values. In other words, any bookmark worth saving should be able to be attached to some sort of actionable tactic.
And as part of that, all of these projects (and bookmarks) need to be actively reviewed. If you aren't regularly reviewing them, then just delete them.
The latter has never really worked for me, so the heavyweight approach is to have a why-justified system. If you're going to save a bookmark, it had better be referred to in some checklist or documentation attached to a long-term project that you are actively working on, and that you've sufficiently justified as being a necessary part of your long term goals or values. In other words, any bookmark worth saving should be able to be attached to some sort of actionable tactic.
And as part of that, all of these projects (and bookmarks) need to be actively reviewed. If you aren't regularly reviewing them, then just delete them.