And speaking of Unix …
I've been using some variation on Unix for almost twenty years now and I'm still coming across new commands that apparently have been in Unix since the Version 6 days (or so it seems). About a month ago I just heard mention of the
seq command, which prints a sequence of numbers (seems pretty silly but it can come in handy when manipulating say, photos from a digital camera from the command line).
And yes, sure enough, it found the command on The Company's servers. But they're running a relatively recent distribution of Linux, I thought. Surely it's some new command.
“No it isn't,” said my over ten year old install of RedHat 5.2, “and stop calling me Shirley.”
I suppose that's expected since I haven't bothered to read all the man pages on any Unix system I've used. But then again, my over ten year old install of RedHat 5.2 has 1,306 possible commmands to run. My even newer CentOS 4 install has 3,117 possible commands to run.
That's a lot of man pages to check.
Assuming every command has a man page to check. There's this
gconfd-2 command that's running, yet there isn't a man page for it, so it's off to Clusty …
This also extends to the various APIs installed on the system. I've been using Linux for easily over 10 years, and just the other day discovered the
ffs() function (which, to tell the truth, I haven't a need for). Big deal—there are a lot of libraries installed that I don't use, but this isn't in some obscure library. It's in the Big Kahuna of libraries—
glibc—the one library that every single program is linked against!
I'm beginning to suspect that program bloat is less about sloppy code and more about duplication of code due to ignorance of what already exists.