ToAruOS is a hobby kernel and supporting userspace, built mostly from scratch, in development since December of 2010.
It was originally developed at the University of Illinois at Urbana-Champaign. For a period of time, it was the development focus of the university’s SIGOps chapter.
There are also instructions for testing and building.
It’s neat what people manage to accomplish as hobbies.
I wish the blog had more details, I’m curious what differentiates an OS like this under the hood.
It’s a bog standard hobby os. That is, init cpu, configure pic and interrupts, setup vga or vesa, import freetype and a graphics composer (easily ported) and a driver for vmware (easy, it’s just a command buffer)
With uefi booting and acpi, hobbyist osdev is dirt simple these days, except the amount of work required in drivers. That’s what kills these projects (including mine, in which I lost interest after writing the sixth network driver from bad specs…)
Can you be anymore discouraging? What happened to doing something because it is fun, or a learning experience?
If you aren’t interested, fine, but keep it to yourself, this is a hobby OS, and we are on OSNews.
I think he’s jealous.
Not really, since I’ve already done everything I wanted with my kernel project (it was further ahead than the bog standard kernel mentioned in the article, but lost interest when I came to the inevitable let’s-make-a-million-drivers stage)
I think this is the point. A lot of hobby/research OS’s are done for fun or study. A few of them reach the point where the basic things work. Then it is time to make it work on other hardware and do something useful with it. The useful things sometimes get done by making a browser run on it or (porting) GNU tools. The driver part simply doesn’t get done because it is WAY too much work and not fun at all.
That is why there are really only a few OS-ses that matter in the world: Hardware support