Announcement

Collapse
No announcement yet.

Redox: A Rust-Written, Microkernel Open-Source OS

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #11
    Originally posted by pal666 View Post
    isn't it nice when real programs in your 'safe' language consist mostly of unsafe blocks?
    I think you're unfairly conflating writing kernels with writing apps. If you want to do crazy low-level stuff, you can with unsafe, but you should rarely, if ever, need it for applications.


    Originally posted by pal666 View Post
    i.e. it will be as successfull as hurd
    Who cares, right? In a few years' time there's going to be more nextgen OS'es anyway.


    Comment


    • #12
      Originally posted by Shimon View Post
      I think you're unfairly conflating writing kernels with writing apps. If you want to do crazy low-level stuff, you can with unsafe, but you should rarely, if ever, need it for applications.
      i'm pretty sure if language designer is faced with choice 'make your language safe always, but unable to write kernels' or 'add ugly unsafe blocks to your language to be able to write one kernel nobody will use', he will choose first option. but in reality you have to use unsafe blocks in non-trivial apps.

      Comment


      • #13
        Originally posted by duby229 View Post
        http://www.redox-os.org/book/book/in...n/why_mit.html

        I don't like how they associate the word "steal". You can't steal something the license allows, but you can abuse it. Abuse is a far better word to use. Also the GPL is not the only license with a decent copyleft clause. I personally like the GPLv2 only license the best, but that's just preference.
        That page is actually painful to read. They are actually trying to claim that proprietary forks that bleed the project dry are beneficial to end users, when exactly the opposite is true. Violating users software freedoms never gives them a better product.

        Comment


        • #14
          Originally posted by pal666 View Post
          i'm pretty sure if language designer is faced with choice 'make your language safe always, but unable to write kernels' or 'add ugly unsafe blocks to your language to be able to write one kernel nobody will use', he will choose first option. but in reality you have to use unsafe blocks in non-trivial apps.
          As a member of the Redox team and someone who has written non-trivial apps in Rust, I fail to see why you think we need an unsafe block for programming non-trivial applications. We have a lot of complex projects written entirely in safe Rust. We have, for example, Piston's image decoder/encoder which has better performance and security than all existing image decoders for a large variety of image formats -- which may be included in Servo at a future date. As it is written entirely in Rust, there's no need to call unsafe C libraries. It may just be that you are trying to force Rust to do things in a manner which is not idiomatic with safe programming practices. However, the key here is that with languages like C, you are forced to effectively run your entire application and/or library unsafely, whereas Rust will allow you to keep the large majority of a low level project safe.

          Comment


          • #15
            Still, it's very interesting to watch how Rust progresses and what impact it's going to have. If anything, it's a great learning opportunity for experienced C++ programmers

            Comment


            • #16
              I'm curious, whether in some point Linus will allow Rust to find its way into Linux kernel. He weren't friendly toward C++, I hope he would change his opinion toward Rust.

              Comment


              • #17
                If you mean replacing monolithic parts of the kernel with Rust, almost certainly not, but modules are a distant possibility.
                I think it would be nice if llvm could compile the kernel first.

                Comment


                • #18
                  Originally posted by duby229 View Post
                  http://www.redox-os.org/book/book/in...n/why_mit.html

                  I don't like how they associate the word "steal". You can't steal something the license allows, but you can abuse it. Abuse is a far better word to use. Also the GPL is not the only license with a decent copyleft clause. I personally like the GPLv2 only license the best, but that's just preference.
                  They're wrong to as ought be obvious by Linux & Android usage when they say "the GPL forces any code that even remotely uses or links to the GPL'd source, to become GPL'd.", libraries can use LPGL and the kernel makes clear that using system intefaces (normally via glibc) is not considered as a derivative work. You have to wonder if such ignorance is deliberate.

                  Comment


                  • #19
                    Originally posted by duby229 View Post
                    […]another project that gets abused by corporations […]

                    Originally posted by zanny View Post
                    […] [regarding "abuse"] have taken FreeBSD kernels and OSes and used them to make incredible profits, with significant code investment, without ever contributing it back to the original project and while denying the users of their products any semblance of software freedom.[…]

                    It is not abuse. It is not a flaw in the license. It is was the BSD license is for. Don't contribute (in form of code, QA, bugreports or use) if you don't like it. The people of contributing to and creating BSD code have made a conscious choice to allow that.

                    Originally posted by duby229 View Post
                    […]A shame really that people can't see the compromise in a copyleft.
                    100% ack!

                    Comment


                    • #20
                      I can assume that commercialized subliminal messaging indeed works and pays off in spades.

                      Comment

                      Working...
                      X