Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

There is one point at which your analogy to pointers breaks down. After having studied pointers, you will never be in a position to disagree. But you could come away from a class on feminism 101 disagreeing with many things.


I've seen plenty of people disagree on whether pointers are a good idea.

Even better, should pointers be nullable?


You should be able to get through OS design either way.

EDIT: addendum. I guess one could imagine a situation where you only needed to have knowledge of feminism 101, but not be required to agree with the arguments. This does not seem to be the case though.


You can get through the class either way, but if you refuse to use pointers, say negative things about them in code reviews, etc., you're eventually going to get everyone frustrated at you. If you can suspend your beliefs enough to use pointers, then you can absolutely get through it, and maybe once you graduate you can figure out how to write a kernel in Python. (I'm not being sarcastic here; there's a python.efi that one hardware vendor is using in production.)

Maybe a better example is monolithic kernels and microkernels. You can believe either opinion, but if you're wading into Linux kernel development now and keep complaining about the lack of microkernel sensibilities, you're going to neither turn Linux into a microkernel nor get people to keep listening to you. If you really, truly believe that Linux's monolithic design is bad for the world, start your own kernel. (Andy Tanenbaum, for instance, had this argument right when Linus Torvalds announced Linux, and since then has been working on MINIX, not showing up on LKML.)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: