From ReactOS Wiki
Jump to: navigation, search

FreeWin95 was the predecessor to ReactOS which was focused on making an open source clone of Windows 95. It never amounted to anything other than being the eventual genesis of ReactOS. The founder and coordinator of FreeWin95 project was Yannick Majoros. Here is project introduction document, written by him in 1996:

                     FreeWin95 project introduction
                         Document revision 0.1
                            Official release
                            English version
This is the introduction text for the FreeWin95 project. This document is not 
meant to be a programmers-only info file, but rather as information for anybody 
interested by the project.

1. What is it ?
The FreeWin95 project is project to make a free system which would be low-level 
compatible with Microsoft Windows 95. The system would be placed under a GNU-like 
license, which would make it so-called "free software". This means that the 
source is released (the original program, under a text-form, readable and editable 
by humans), that nobody may sell the system (it is free!), and that everybody may 
contribuate to the project (this last topic is actually not in the GNU license, 
but the project will be so). Low-level compatible means that it would accept 
system drivers which are designed for use with Windows 95.

2. In which way can I be interested ?
First, as a would-be user. You can not only encourage us, you can also tell us
your suggestions, what you like or don't like in Microsoft Windows 95 and what 
we have to make better. You can help development by spreading informations about
the project (you are encouraged to spread this info file).
Secondly, as a developer. You will naturally understand that programmers are 
necessary for such a project. They are the people who will put the project in 
concrete form.
In the third place (but not less important!) are the people who will handle public 
relations. This goes from information for the public and relations with hardware 
producers for drivers to juridic problems (everything we release has to be legal).
And, at last but not least, we need people who dedicate web pages or bbs'es to 
freewin95. A freewin95 home page would really be interesting. A mailing list would 
be interesting to, and someone has to maintain it. E-mail for freewin95 
developers/helpers would also be handy.

3. Organisation
We need at least a little organisation if we want to achieve a fully-working 
system which meets our main goal : low-level compatibility with Microsoft Windows 
95. Organisation doesn't mean hierarchy. I think everyone has to be treated 
equally in the project.
We first need a coordinator, at least for the first months. I think that the 
project can then goes further without coordinator. His task will be to suggest 
who is going to do what. As I started the project, I am candidate for this and 
will act as if I had this function until there are major complaints from 
potential freewin95 helpers. In this case, we will think about a new organisation 
scheme (without me, naturally).
Then we need someone who will maintain a web site (home page ?) and someone who 
will maintain the mailing list. A newsgroup would be great too.
We also need someone who will write the "official" texts. If after a while we 
see more than one development tendency, there will be more than one distribution.
I don't see anything more which would be necessary in the organisation. If you
do, please tell us.

4. Development information
I didn't discussed this with other people yet, and I think nobody has to decide 
it on his own. We have to discuss it all together. A newsgroup would be great 
for it, and a mailing list is almost necessary, as I said earlier in this text.
I propose (this has to be discussed !) that we do everything in C/C++, so we all 
"speak the same language". I know there is a port of GNU C for Windows 95, and 
I propose we use this for every application we will develop. But this is for 
later, as we first need a kernel. For this, I propose we find a good compiler. 
We could eventually adapt GNU C. We could also do a lot in assembler, but I 
don't know if it is the best solution. It would not only take a lot of time, it 
would also not generate the fastest code. If we have to optimize everything for 
Pentium processors by hand... And who can be sure to make the best code with 
these pipelines, the cache, etc. ? What will almost necessarely have to be in 
assembler is the code which will put the processor in protected mode and manage 
memory. Anyway, ideas are welcome...
I also wanted to say that you don't have to be a great programmer to develop 
something for freewin95. Feel free to try, there will be enough people to correct
you if you're wrong, and it's the best method to learn.

5. How to contact us ?
At the moment, the only mean is by e-mailing to me (project coordinator) :
                       Yannick Majoros, aka Mellery Mutant, aka Melmut
                       internet e-mail :