Thanks Jim for your encouragement and support. What your suggesting is an 
interesting idea, like Pat Villani's original idea of implementing FreeBSD 
drivers into DOS. This idea intrigues me alot, i am a big BSD fan especially 
NetBSD and DragonflyBSD, thus implementing a Hybrid DOS-BSD system would be a 
rather neat feature.

I agree with you for the most part Rugxulo that focusing on smaller bite size 
peices and planning it before you start i.e. the userspace utilities first is 
the best way to go. :D

 I don't have alot of knowledge on alot of the scripting language which is my 
limitation for knowing what things are possible. I would agree that porting 
from Minix would provide cleaner code than Linux, but while i say that i 
wouldn't be porting from Linux code but most likely BSD code. BSD code is 
generally scrutinised more and far more stable than Linux code, also BSD code 
would be easier to port than Linux code, as mentioned above. NetBSD in 
particular has alot of source that is compatible and has been used by alot of 
other OS's thus using some of their code wouldn't be a stupid idea.

 The biggest issue is that the project would take years for me to do alone and 
i don't know enough about all the different areas, scripting language being one 
of them to implement the idea in the cleanest and most efficient way.

 I think your question Mike is valid, but my belief and core idea is not to 
remove the INT 0x21 programming interface or the current software compatibility 
infact that was one of the key motivations of my design, was to keep all those 
features. It was the core idea of implementing the current FDOS kernel and 
running it along side a compatible DOS kernel that implements alot of the 
functionality of a current SMP capable kernel but at the same time can allow 
the DOS kernel to run the system indirectly or directly.

 Although I don't believe how DOS memory paging works or the Direct hardware 
access is what makes DOS, DOS. That was just the way of designing the system at 
the time, direct hardware access undoubtly has its benefits but in the grand 
scheme of things does the end user ever really see the difference? would it 
ultimately change what DOS can on a computer?

 The idea is similar to how MS implemented Windows 95 but the key difference 
NOT placing DOS into a virtual machine that is just the bootloader essentially. 
The only other OS i know that implements a similar feature is DragonflyBSD. It 
uses a virtual kernel in userspace but the key difference is that the virtual 
kernel provides instructions amongst other things to the kernel.

 Whether the creation of a hybrid system of a DOS kernel alongside a BSD kernel 
is possible (i.e. Converting between the different scripting languages)? or 
wheather you would have to write/ re-write a new kernel i am not sure?

 Running DOS like DOSEMU etc ontop of a Linux kernel would be pointless because 
the core interface is still Linux command and only becomes DOS when you enter 
the DOSEMU. Thus, to update the Linux kernel or to fix problems related to 
Linux (i.e. bugs, security, etc) you still need to use terminal. Unless the 
DOSEMU is able to implement a Linux command feature within or emulate it. Its a 
neat idea and would be kool, but at the end of the day its just another Linux 
system, with a different spin.

 I think DOSEMU demonstrates that the idea of putting the DOS kernel alongside 
another (non-DOS) kernel is infact possible. Strip DOSEMU down to the bare 
minimum and what you have is a Linux programming converted into DOS language. 
Just like how Wine is used in ReactOS.

 Martin
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Freedos-devel mailing list
Freedos-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freedos-devel

Reply via email to