> headers in HTTP/2 are compressed using HPACK which negates the issue you mention You are right, my statement is not correct. It certainly does not completely

> headers in HTTP/2 are compressed using HPACK which negates the issue you mention You are right, my statement is not correct. It certainly does not completely negate the overhead of additional headers. It does greatly reduce it though. So far, I have only heard of compression inefficiency being a real problem when not concatenating. Do you have some examples of header overhead actually being a real issue? Would be happy to update the post.