Date: Wed, 29 Mar 2006 21:00:07 +0300 From: "Vladimir Tsvetkov" <npacemo@gmail.com> To: Vaaf <vaaf@broadpark.no>, freebsd-questions@freebsd.org Subject: Re: Why are so many people using 4.x? Message-ID: <666bdb140603291000u354e948fo@mail.gmail.com> In-Reply-To: <7.0.1.0.2.20060329141021.023e9af8@broadpark.no> References: <200603281234.11850.satyam@sklinks.com> <7.0.1.0.2.20060329141021.023e9af8@broadpark.no>
next in thread | previous in thread | raw e-mail | index | archive | help
2006/3/29, Vaaf <vaaf@broadpark.no>: > FreeBSD, and UNIX for that matter, is based off 30-year-old concepts. > Noboy can deny this. That being said, you can compare the development > of FreeBSD to building a skyscraper on shallow grounds. Naturally, the mo= re > you build the more building is likely to collapse. This is now the case w= ith > the old FreeBSD (in which a couple of smart guys decided to savior into > DragonFly) versus the new FreeBSD. I think the same thing is happening > with Windows versus Vista. As OS development progresses, this little > theory of mine will become more and more obvious. If anyone on this list = can > contribute with facts and observations to strenghten this theory, I would > really appreciate it. 30 years of development and continual introduction of new features build on top of existing ones is considered a very good design. And FreeBSD is still extensible and growing, despite of its age. And FreeBSD is not a skyscraper neigther literally, nor metaphorically - it's more like a spaceship - a very robust one - gives you the means and tools to save your life in deep space when a threat to your life appears and there is noone around. Before even starting talking about design, we should give proper definition for this concept. What is good design? How do we measure one design against an alternative one? The widespread notion of good desing is related to the ability to maintain, extent and comprehend easily some complex system. 30 years... You do the math! I'm not sure you're ready to present a new and revolutionary design (you should start a new threat on that). It's more like you're in search of volunteers to your FreeBSD Critisism Project. Revolutionary design means starting from scratch - this would be a huge, tremendous investment of time and efforts(choose a platform, a language, write a compiler for it, start building a kernel, write completely new device drivers - Microsoft have its Singularity Research Project - an operating system written entirely in C#, but they don't share the tools - the C# compiler and linker they use to build that system, neighter the code - you can get just a couple of PowerPoint presentions, an interview, and a short 50 page long paper, about the features that this system will introduce - on the other hand you can get all of the FreeBSD source code, tones and tones of documentation, and hundreds of ready to help you people - FOR FREE). And there's no guarantee that this new design would last even 5 years. At some point in time this will probably happen, but it won't be FreeBSD. FreeBSD is not a vendor - it's an existing and evolving operating system and a commited community of FreeBSD users. The emphasis is on evolving. If we want to stick to FreeBSD, the new design should be evolutionary one, which is pretty different in concept - we would start from a familiar code base and would slowly integrate changes (just like the DragonFly project) into this base, thus creating a new BSD branch of development. Best Regards, Vladimir Tsvetkov
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?666bdb140603291000u354e948fo>