I see where everyone is coming from in saying that FreeDOS should remain
16-bit. For a long time I was a *firm* believer in the superiority of
16-bit code. Heck, I insisted on making my GUI project 16-bit when *every*
single other one out there was done in 32. But, as time wore on and I had
time to play with the various DOS extenders, I learned 32-bit has its
advantages. That's saying a lot coming from me! lol But I can see both
sides of the argument here.
I agree with Jim that our primary focus should indeed be application
compatibility - which I think we can retain even in moving to a 32-bit
base. I'm simply saying that if we can maintain *full 100% application
compatibility* and still have a 32-bit code base, FreeDOS would suddenly be
the most blazing fast DOS ever conceived. After all, it's not like we left
the sub-386 crowd high and dry. They can still run FreeDOS through version
1.2, which will take care of all the needs of older processors. You may
say, *what about the new userspace utilities released with FreeDOS 2.0?
Users of old hardware won't be able to benefit from that, will they?* In
short, yes they will. Since we're maintaining the common goal of 100%
application compatibility, all the utilities and programs released with any
future edition of FreeDOS (2.0 and up) will still run flawlessly on 16-bit
FreeDOS because they will be traditional 16-bit binaries. I don't believe
that moving the core of the OS to 32 bits will have to break compatibility
for our previous users.
Throughout history every major operating system has upped its system
requirements as it evolved. On the Macintosh platform, the introduction of
System 7.6 meant dropping support for the Mac's original processor, the
venerable Motorola 68000 due to the OS requiring 32-bit clean code. Later,
the release of MacOS 8.0 saw Apple drop support for anything below a 68040.
Yet the OS was still the same MacOS as it had ever been, and even ran the
same applications - it just simply evolved. Users of previous versions
continued to run their computers just as they always had, and were still
able to run new apps which came out because developers didn't make the
mistake of *requiring* the latest and greatest OS version to run their
apps. The same thing has happened with the Amiga OS, Windows and most
likely nearly any other you can think of. The OS evolves, it changes, it
grows. Certain older systems can no longer run new versions of it. In the
cases of most commercial operating systems, this means that users most
likely have to upgrade their hardware to keep up, but our project would do
better than that - we would upgrade the core OS, but still run a 16-bit
userspace so that all DOS users continue to benefit from our progress.
In the end, I'm not telling anyone what to do or beating anyone over the
head to see things my way. As Mike said, I'm just giving my opinion here,
that's all. I think, for a number of reasons, FreeDOS 2.0 will most likely
stay a 16-bit OS - and there's nothing wrong with that. We can add modern
OS features (protected memory and multitasking are still quite doable)
without jumping to 32-bit code. After all, there obviously already is a
32-bit FreeDOS project, and it wouldn't really make sense to have *two*
32-bit versions of FreeDOS.
Just my two cents.
------------------------------------------------------------------------------
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net
_______________________________________________
Freedos-devel mailing list
Freedos-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freedos-devel
