3 ms·

I had the opposite reaction. Many of the things the OP wants to change are really bothering me about "modern" operating systems. A recipe for coming up with go

I had the opposite reaction. Many of the things the OP wants to change are really bothering me about "modern" operating systems. A recipe for coming up with good ideas is looking at something that has been around for a while and asking "What would it be like if it had been invented now?". And if all else fails, this will be a wonderful learning experience. I know this is a cliche, but before writing comments like this ask yourself if a) you are unnecessarily discouraging somebody for no reason and b) you would have said the same thing about successful projects when they were first started. Finally, the one piece of advice I'd give on this project is that you shouldn't focus too much on the overly ambitious goals and don't worry if you can't accommodate for all of them just yet. The likelihood of achieving the next milestone is somewhat proportional to the number of milestones you have achieved. Get to v0.01 and take it from there.

robinh 14y ago
"I know this is a cliche, but before writing comments like this ask yourself if a) you are unnecessarily discouraging somebody for no reason and b) you would have said the same thing about successful projects when they were first started." a.) No, I don't think so. Considering the number of half-started kernels on OSdev.org, the creators of which incredibly often were convinced that they were going to create the newest awesomest kernel everyone is going to use, trying to get the OP to take a good look at his skillset seems like a proper thing to do. b.) I would have, if the creators seemed completely ignorant about the basics of what they were trying to do. This is not to say that I would tell the OP to give up on OS development - I personally love it. But I think a serious reality check is in order, and because we get people similar to the OP on the OSdev forums an awful lot, sometimes this frustrates me.
- qu4z-2 14y ago
  I don't really expect OP to succeed, but I'm sure it'll be educational. And really, the more enthusiasm for a project the better.
bitwize 14y ago
A recipe for coming up with good ideas is looking at something that has been around for a while Unix and asking "What would it be like if it had been invented now?". creat(2) would have the second 'e'.
- qu4z-2 14y ago
  I'd like to think the "everything is a file" mentality would've been taken a touch farther, and the command line tools would deal with networked resources better (eg cp ssh://example.com/~/somefile ~/localcopy)
  - bitwize 14y ago
    Oh, you mean like Plan 9? :) I would have mentioned it, but I was speaking as if Unix were invented today. Plan 9 is an almost textbook second-system effect; it seems like you have to have invented a Unix first before even starting on something like Plan 9.

robinh 14y ago

"I know this is a cliche, but before writing comments like this ask yourself if a) you are unnecessarily discouraging somebody for no reason and b) you would have said the same thing about successful projects when they were first started." a.) No, I don't think so. Considering the number of half-started kernels on OSdev.org, the creators of which incredibly often were convinced that they were going to create the newest awesomest kernel everyone is going to use, trying to get the OP to take a good look at his skillset seems like a proper thing to do. b.) I would have, if the creators seemed completely ignorant about the basics of what they were trying to do. This is not to say that I would tell the OP to give up on OS development - I personally love it. But I think a serious reality check is in order, and because we get people similar to the OP on the OSdev forums an awful lot, sometimes this frustrates me.

qu4z-2 14y ago
I don't really expect OP to succeed, but I'm sure it'll be educational. And really, the more enthusiasm for a project the better.

bitwize 14y ago

A recipe for coming up with good ideas is looking at something that has been around for a while Unix and asking "What would it be like if it had been invented now?". creat(2) would have the second 'e'.

qu4z-2 14y ago
I'd like to think the "everything is a file" mentality would've been taken a touch farther, and the command line tools would deal with networked resources better (eg cp ssh://example.com/~/somefile ~/localcopy)
- bitwize 14y ago
  Oh, you mean like Plan 9? :) I would have mentioned it, but I was speaking as if Unix were invented today. Plan 9 is an almost textbook second-system effect; it seems like you have to have invented a Unix first before even starting on something like Plan 9.