There's a specific way of writing that shows up over and over in LinkedIn posts.
It's easy to recognize. Text broken into small fragments. A lot of empty space. A cadence that guides the reader's eye more than the argument itself. People adopt it because it seems to work. It's visible, it stands out, and the platform appears to reward it. The result is a stream of posts that start to feel interchangeable. Emojis often become part of the same pattern, used less as expression and more as visual anchors.
At some point, the form becomes more noticeable than whatever is being said. After seeing it a few times, your eye starts skipping it almost automatically. You don't really read the post anymore. You register the shape and move on. It's a familiar kind of filtering, similar to how many of us learned to ignore the right-hand ads on Google back when ads were still visually separated from organic results.
This shows up in odd ways. You can find people warning against automated writing while using the same generic structure that most generated text defaults to. The issue isn't whether a tool was involved. It's the uniformity. Once everything fits the same mold, it all starts to sound the same.
The feedback tools on the platform don't really affect this layer of the experience. You can express preferences, mute topics, or hide individual posts, but that feedback mostly serves the platform's own internal goals, not any meaningful improvement in the user experience. The dominant posting formats and behaviors remain the same, driven by imitation and fear of missing out rather than by any proven effectiveness.
So I decided to change the experience locally.
Nobuco is a small Chrome extension that filters posts by their formatting patterns. It doesn't inspect meaning or intent. It removes posts that match certain structural cues: heavy use of line breaks, very short paragraphs, spacing meant to manufacture emphasis, polls, and the overuse of emojis as structural markers.
I'm using it myself and it does what I built it to do. If LinkedIn adjusts how posts are rendered, the rules will need to be updated. I wrote this to make my own feed easier to read. It's rough, but it works.
It doesn't make the feed better in any general sense. It carries a cleaner signal-to-noise ratio at the level of presentation. Less repetitive in form. That alone changes how the content lands. Good and bad content are still there, but some of it is visually synthetic, shaped by cookie-cutter formats. That's what gets filtered out.
References: