I know Unix filesystem sometimes has things that appear to be files but are not actually files. The files in
dev actually point to devices. Its also possible to created named pipes which also appear to be files but are actually fifo message queues of sorts. The
proc directory has files based on processes, memory, hardware, etc.
I am trying to better to understand these abstractions to decide which I might be able to utilize. How many ways does Unix feign files and which can be readily implemented/used by admins? (i.e. named pipes and symlinks are easy enough to use.) If I knew the right terms for the concepts I might be able to find the details online.