A curiosity question which google has been unable to answer for me:
Why is 'resolv.conf' spelled without the trailing 'e'? I'm sure there's a valid historical reason going back to the early days of BSD DNS resolver library.
Asked
Active
Viewed 668 times
3
1 Answers
0
I'm going to guess that to ensure things could be typed fast (and more importantly, spelt out loud quickly), everything was limited to six characters - as mentioned above. This wouldn't have been a filesystem limitation, just common convention. 'lib' could be added to the beginning of a name to indicate it was a library, and other appendages could also be added to specify its type and so on.
haneefmubarak
- 395
libresolv, but of course that just leads to the question of whylibresolvis calledlibresolv. To me it sounds like one of those legacy 6-character limits — the same reason whystrcmp& co. are all named with 6 characters. – Celada Apr 29 '13 at 22:33libresolvis more than 6-chars, which suggests it's not a filesystem naming limit. Andresolvisn't a C function name likestrcmp. – Flimzy Apr 29 '13 at 22:37.cor.hit's 8 characters, which was the limit. In this case there's aresolv.hin the kernel, so it's probably the same reason – Michael Mrozek Apr 29 '13 at 23:07