Because a summary is very useful. Imagine you are reading about 5 years of day-in-day-out work by a team of 1500 physicists (eg. CERN, LHC, etc...). Would you

Because a summary is very useful. Imagine you are reading about 5 years of day-in-day-out work by a team of 1500 physicists (eg. CERN, LHC, etc...). Would you rather read the entire as-it-happened stream of research, going back 7500 person-years - or would you prefer to read a summary, in the form of a paper? I agree that ideally the stream should be publically available as well, for those that want to consult it - but having regular summaries, written at intervals is very useful.