Well, probably lots of people have thought of this, but since I was remembering thinking about it, I thought I’d post it.

Compression has been described to me as the art of removing redundant information or as the art of finding patterns and using them to reduce redundancy.   Since I’ve been archiving some information lately,  I wondered if the compression could be influenced by gathering up similar files while compressing groups of them.

For instance, if all the Word documents in the folder being compressed were taken together, would the compression algorithms achieve better results than if the files were presented in some random order with other file types?

Well, long-time compression utilities like ZIP, GZip, and StuffIt  probably already do this. But its always good to keep looking for a better way. (Maybe I’ll call it  “looking for n – 1” or “n -= 1” ).

Compile!  (Just struck me as sounding like Klingon “Qapla'”. )