zlacker

[parent] [thread] 2 comments
1. IgorPa+(OP)[view] [source] 2026-01-04 18:58:26
The s in sbin stood for static initially. Of course nowadays this is not enforced.
replies(1): >>buzer+g6
2. buzer+g6[view] [source] 2026-01-04 19:37:37
>>IgorPa+(OP)
What is the source for that? Some of the oldest references to sbin I can find are 4.3BSD Net/2 man pages (https://man.freebsd.org/cgi/man.cgi?query=hier&apropos=0&sek...) and Filesystem Standard v1.0 (https://www.ibiblio.org/pub/Linux/docs/fsstnd/old/fsstnd-1.0...). Former doesn't mention anything about static binaries, latter only mentions that static ln (and even mentions sln being static version of ln) and sync can be useful.
replies(1): >>JdeBP+y55
◧◩
3. JdeBP+y55[view] [source] [discussion] 2026-01-06 10:03:57
>>buzer+g6
There are no contemporary sources for that because it is, as it was called here on Hacker News some years ago, an 'ahistoric retcon'.

* >>31338349

The 's' is for 'static' version of the explanation of the name of sbin is not actually supported by any 20th century Unix doco. The books on AT&T Unix System 5 (before which, things were in /etc) that actually give an explanation for sbin all say system binaries, or system administration commands; and none of them says anything about linkage.

The 'static link' story came from Linux people years afterwards. Here's Ian McCloghrie correcting this misconception in a Linux discussion back in 1993:

* https://groups.google.com/g/comp.os.linux.development/c/EKzL...

I pointed out the origins of sbin some years ago.

* https://lists.debian.org/debian-user/2019/02/msg00041.html

[go to top]