The blog post misunderstand that many people that do that are simply using "hello" as a "ping" to see if someone's really there for immediate synchronous communication. They want to type just 1 word first ... and then wait for a human-style-TCP-ACKnowledgement ... before typing a bunch of extra words of the real question they had in mind. For receivers, it's annoying but I tried to explain in the previous thread that many deliberately avoid indeterminant async communication if possible: https://news.ycombinator.com/item?id=25883095
The whole issue is that with "hi", the intent is entirely unknown to the recipient. The sender should signal their intent such that anyone responding knows roughly what they're getting into. It can be as simple as "hi, free to chat for ~N mins about X?" but anything more that just a greeting is better than none. That's the "educational" value of the post as far as I see it, and I totally agree with it.
It takes the sender a lot of time to type the question, so the communication is not immediate synchronous. What actually happens is that I go elsewhere after 5 seconds of waiting, then in 20 more seconds arrives the question, but I already switched to a different task and don't bother what another stupid hello I received there. The site provides a chat log with timestamps to better understand delays in communication.
