What is the reasoning for having subscriptions for journals? Does the journal work like some sort of filtering system to weed out the bad articles and only provide the user with good articles?
Unlike math and physics, many disciplines have only recently adopted preprint servers (like bioRxiv) and there's still some bias against posting preprints within the community.
There's been a number of times where I've been researching niche topics and found "peer reviewed" publications where the math was incorrect, the code examples were invalid syntax, and had results/analysis that I couldn't recreate.
Sometimes I wonder if the authors submit to journals until they find one that sticks.
That's not to say "good" journals don't have "bad" publications. Some have various levels of peer review, like full text/abstract only/not reviewed at all. Sometimes you wind up with content that deserves a tech blog post, not a journal article. But different strokes for different folks, it's not hard to filter out mentally once you're familiar with the literature.
See Inside the Fake Science Factory
So what is the reasoning?
Pure institutional inertia.