3 ms·

>is anywhere close to settled enough Some things are pretty well settled. 40 year old Unix software (K&R C,IPC/network sockets) would work on a modern linux wi

>is anywhere close to settled enough Some things are pretty well settled. 40 year old Unix software (K&R C,IPC/network sockets) would work on a modern linux with trivial or zero changes, and it doesn't look like it's gonna change soon. That was the age of not room sized, but half a fridge sized "minicomputers". Old software that people rely on doesn't necessarily need new features or improved capabilities that new standards and interfaces provide. Security, the constant change of the environment(network, protocols), and hardware and software changing in tandem are the only reasons people have to migrate or rewrite some pieces of well established software. Some of these changes in the environment feel as if these changes are for the changes' sake, to create programmer jobs.

AnimalMuppet 3y ago
> 40 year old Unix software (K&R C,IPC/network sockets) would work on a modern linux with trivial or zero changes, and it doesn't look like it's gonna change soon. Is that true? Will K&R C (that is, C with old-style function declarations, not prototypes) compile on a modern compiler? I thought that gcc, for instance, no longer supported that declaration style. (And if you just have to add a flag to the makefile, I will agree that it's a trivial change.) On the sockets, I agree with you.
- nullifidian 3y ago
  >I thought that gcc, for instance, no longer supported that declaration style. It's g++ that doesn't support this, gcc should.