I had discovered FreeDOS about 2 months ago. Today, I LiveCD-tried ReactOS, an open-source initiative aiming at reproducing the Windows NT environment.
Although I’m not an expert, I found the software to be pretty stable. However, the system crashed when I installed good old “Alone in the Dark” (those who were playing PC games back in the middle of the 1990s certainly know what I’m talking about). The ReactOS team had warned me though: the current version (0.3.0) is an alpha stage release and is not recommended for everyday use.
Looking at the history page of the ReactOS community website, it appears that the project “grew out of a dissatisfaction with Microsoft’s monopoly over the operating system market” (stated in context).
Jason Filby, and all guys belonging to the team: you’re doing an amazing job. But how the heck can you claim to escape from a so-called monopoly by reproducing its flagship product?
You’re definitely all excellent at hacking quality code, and you wanted something new, a different OS. By the same token, it was obvious you’d never reach Windows NT’s quality and functional scope. So why didn’t you guys teamed up to start a start-up company? Why didn’t you help yourselves fullfill your dream and come up with a brand new, disruptive operating system?
I’m guessing you’re sitting on the fence with Microsoft. It’s a soft of love-hate relationship. If I were a painter and hated Picasso for any reason (which is far from being the case), I would want to copy anybody but him.
Before I finished writing this post, I discovered on Alex Ionescu’s blog, one of the main recent contributors of the ReactOS community, that Alex actually is a Microsoft Student Ambassador.
Eventually, it looks like my intuitions weren’t far from the reality…a love-hate relationship.
everyone who wants to make a new OS has to deal with this problem: everyone has Windows software and drivers and is used to the way it works. An OS like this could be the first step, allowing people to try it out as a OS they can use and just make it better over time.
Comment by arielb — December 7, 2006 @ 02:46
Hi Jeremy,
Our website is over 10 years old and the original reason why ReactOS was started is quite different from our motivation today. Since the server is back up, I invite you to read the about pages to read a more up-to-date mission statement. It’s not about defeating a monopoly anymore.
Some of us are doing it for fun, others are doing it to learn NT internals and driver development (Which can land you a nice job), and others are doing it to get generic OS writing experience. Others yet are learning how to cope with projects with 5+ million lines of code, with testing, frameworks, emulators, compilers, and all the intricacies involved with producing an OS. More then anything, this experience has taught me a lot.
Sometimes I hear people ask Microsoft why it takes them 2 weeks to come out with a patch that’s 5 lines long. After working 3 years on ReactOS, I personally want to bash those people in the head. Repeatedly. You don’t have a choice. And we don’t support nearly all the languages, service packs and special hotfixes that MS needs to worry about when making a fix.
Just managing our SVN repository is a gargantuan task, and getting builds to always work (thank God for BuildBot!) when compiled on so many machines. Hopefully VMWare 6.0 APIs will help with regression testing.
I disgress, but I hope you get my point
.
Best regards,
Alex Ionescu
Comment by Alex Ionescu — December 29, 2006 @ 08:01
Thanks Alex, I couldn’t access the most up-to-date pages before indeed.
Your input provides a lot of insights about the developers’ motivations investing time to get:
i) new skills (NT nucleus or gigantic software projects: 5m+ lines);
ii) OS-writing experience;
Regression management is indeed key, as you point out: it’s nice to fix a little bug, but less sort of nice if your patch impacts another part of the software. Sometimes, complexity makes it better to live with a small bug than fix the bug and end up with a major problem.
Of course I’m getting your point, thanks for your very valuable input. Your project is extremely interesting, and you’ve done a great job devising an OS from scratch – congrats to you and your team for achieving such a thing. Furthermore, motivational aspects are much clearer as far as I’m concerned.
Take care,
Jeremy
Comment by Jeremy Fain — December 29, 2006 @ 12:25
You can try DisBox, a DOs emultor
It’s work well on my Macintosh
I used it to compile old C++ et Pascal sources abd lauch i386 C++ and pascal appications
http://desechalliers.ldeweb.net/entreprise/index.php?2006/06/12/565-1991-revival-ou-comment-ressortir-ses-vieilels-productions-du-placard
Comment by Laurent Desechalliers — January 2, 2007 @ 00:46
Many thanks for the hint Laurent, I’ll try that soon.
Comment by Jeremy Fain — January 2, 2007 @ 01:10
It’s DosBox and not DisBox
Sorry
Comment by Laurent Desechalliers — January 2, 2007 @ 01:20