[ros-kernel] Different Hardware Architectures

[Jasper van de Gronde]

Waldo Alvarez Cañizares wrote:
> As for putting MS to shame, I would only try to do it if that would increase some quality in their products because that affects many of us directly or indirectly even if you don't use their software but just for hate to MS makes no sense, I do not hate MS I hate what they do or try to do sometimes. In fact I have reached to the conclusion that even to have an enemy  is a great thing because it usually leads to development.
> 
> If it is for sports, well be welcome it won't hurt anyone, however there are more interesting acomplishments like just making ROS run every win32/win16/Dos application out there. Don't you think? I would be very happy if i could see just that and I'm sure that many will be. Or what about about making it more efficient, currently ROS in graphics mode is a turtle, too much repainting without need. What about finishing the network code. There's a company here in my country that build cash registers, there are very low end machines that use custom software that access a central database to store/retrieve data using some network card. That software runs on top of Win95. I was thinking about talking with someone there, I'm sure they will be interested in ROS and probably would invest some money in developement for ROS. I just can't talk with them in the actual stage of ROS, what I would show, some tests and that it runs on 3 different archs? Bahh they will not hear me.
> 
> As for applications ... I'm sure there's a few that would port over. If
> you're pointing out the low return on investment, then yes, I think you
> are right. However, it only takes one or two persons that ... for
> whatever reason ... very much want or need ReactOS to run on some
> alternative platform and are willing to put forth the effort.
> 
> -rick
> 
> The only reason I see to port it is if you make something like a decompiler + compiler to make those existing programs run on top of the new arch. But keep in mind that such a thing would bring a lot of compatibility problems, decompilation is a non-computable problem and can only be done using some inexact methods. I don't think that the few that would port as you say will be enought to satisfy the requierements of many. I'm not a lover of x86 and I think that the PowerPC design is more clever/simple/scalable simply because it is newer and was based on the experince of Motorola (builder of CPUs) IBM(builder of CPUs OS and compiler) and Apple (builder of OS and computers and compilers) and that the design also was made to last. However I ask myself what's the need to put ReactOS to run on an alterative plataform. The only reason I see is to use ReactOS as a server or a high performance cluster but then there are plenty of free OSes out there already with enought applicatio
ns mature enought to be used as servers or clusters. In my opinion the only hole for free OSes lies just where Windows + Intel x86 lies at least in the next years to come.

Another reason why it could be useful to port ReactOS would be that 
porting a program is a reasonably effective way of debugging certain 
aspects of the program, as some careful thought will have to put into 
things that might otherwise go unnoticed.