Since foreigners can copy the Unix kernel and make a Linux, then why can't we absorb the best parts of DOS, Unix, and OS/2 to develop an OS of our own?!
Some people say our Chinese technical level still isn't enough. I don't think so. It's not that Chinese people aren't good enough at programming, but that our development thinking may be a bit lacking. Many people may think that to absorb the ideas of Unix, Linux, and OS/2, we have to completely analyze their kernels. Actually that's not necessarily so. There are already plenty of books analyzing these kernels. We can just read them and absorb the ideas in them, without having to fully understand their source code.
DOS with available source code is a very good model. Its scale isn't too big, and everyone is familiar with it. We could completely analyze the DOS source code and on that basis rewrite a completely localized 32-bit, multitasking, multi-user DOS-like operating system, while also keeping compatibility with the DOS command line. Then later add an X-Wndows to it—isn't that exactly the successful approach Linux took?
Some people may say that this kind of thing already exists abroad, and is already being done. But that's still somebody else's. We still have to do Chinese localization and so on, and after all that back and forth, the effort spent won't be that much less.
What we most need to do right now is carry out “DOS source code analysis.” As a model we can choose some version of any DOS with public source code, including Ms－dos6.22 (its source code has also been made public). We only need to analyze the kernel part. The external commands can be left alone for now, otherwise the scope will be too large.
Some people may say that we might as well analyze the Linux kernel, study Linux, and do Linux Chinese localization. But the Linux kernel is too huge. Even with so much reference material, it still takes a long time to fully understand its kernel. With that much time, if we had reference materials, DOS would already have been thoroughly analyzed. Linux Chinese localization in China doesn't seem to have been very successful either. After all, it's still someone else's thing.
Some people may say that DOS source code has comments, but does having comments make it easy to understand? Is everyone's English really that good? We are doing “DOS source code analysis” so the road ahead will be easier, and so people won't have to waste time struggling through English.
Finally, let me emphasize my point: what we want is to rewrite a localized OS, not to Chinese-localize DOS!!!!
Some people say our Chinese technical level still isn't enough. I don't think so. It's not that Chinese people aren't good enough at programming, but that our development thinking may be a bit lacking. Many people may think that to absorb the ideas of Unix, Linux, and OS/2, we have to completely analyze their kernels. Actually that's not necessarily so. There are already plenty of books analyzing these kernels. We can just read them and absorb the ideas in them, without having to fully understand their source code.
DOS with available source code is a very good model. Its scale isn't too big, and everyone is familiar with it. We could completely analyze the DOS source code and on that basis rewrite a completely localized 32-bit, multitasking, multi-user DOS-like operating system, while also keeping compatibility with the DOS command line. Then later add an X-Wndows to it—isn't that exactly the successful approach Linux took?
Some people may say that this kind of thing already exists abroad, and is already being done. But that's still somebody else's. We still have to do Chinese localization and so on, and after all that back and forth, the effort spent won't be that much less.
What we most need to do right now is carry out “DOS source code analysis.” As a model we can choose some version of any DOS with public source code, including Ms－dos6.22 (its source code has also been made public). We only need to analyze the kernel part. The external commands can be left alone for now, otherwise the scope will be too large.
Some people may say that we might as well analyze the Linux kernel, study Linux, and do Linux Chinese localization. But the Linux kernel is too huge. Even with so much reference material, it still takes a long time to fully understand its kernel. With that much time, if we had reference materials, DOS would already have been thoroughly analyzed. Linux Chinese localization in China doesn't seem to have been very successful either. After all, it's still someone else's thing.
Some people may say that DOS source code has comments, but does having comments make it easy to understand? Is everyone's English really that good? We are doing “DOS source code analysis” so the road ahead will be easier, and so people won't have to waste time struggling through English.
Finally, let me emphasize my point: what we want is to rewrite a localized OS, not to Chinese-localize DOS!!!!
QQ：13688252
DOS 是我的初恋，ASM 是我的母语
DOS 是我的初恋，ASM 是我的母语
