Haiku will never die

Even if it takes 100 years,one day we will use Haiku os on our computers,but we need to find financing,how about kickstarting,I told earlier,money is the key to fasttrack development.

Hey hydragon,

luckily I don’t have to wait 100 years to run Haiku; I’ve been doing that for many years now… :slight_smile:
WRT to kickstarter, do a search for “kickstarter”. It has been discussed many times. The gist as I see it: nobody currently active in the Haiku project is sufficiantly convinced a kickstarter campaign would be successful to spend time and effort on it at this time.
Of course, anyone that is can go ahead and just do it. If it proves successful after all, I bet the patches/commits of all the hired developers will be very much appreciated by everyone.


Haiku already accepts donations so the only point of something like a kickstarter campaign would be to raise awareness.

IMO A better, more genuine way to raise awareness would be for users to do interesting things with Haiku and document it (the more pictures and thoughtful layout, the better) - write it up on their social media.

This isn’t exactly a low-hanging fruit idea but if Haiku suddenly became bootable on an RPi I think that would earn the project a lot of attention. On that topic, can Haiku stop short of loading its graphical desktop? Does it have a text mode? Would that be a more achievable goal for the ARM port?

We have a kind of text mode used for the kernel debugger. This uses the same framebuffer as everything else and draws the text with simple font rendering code. The good news is that the framebuffer driver is one of the few already running for our ARM port (currently on the BeagleBone xM, because Haiku on the Pi will be like anything else on the Pi: too slow for practical use - especially without video acceleration). We already have the splash screen showing on the BeagleBone xM.

Our current task for the ARM port is getting either USB or SD mass storage running, so the kernel can load drivers and userland. Once that's working, we can see what's the state of the userland and which apps already run (probably none at first), and fix the bugs.

Thank you for your informative reply. While I’m aware of other reasons that make RPi a challenge to develop for (such as a GPU that’s geared toward limiting its functionality), surely the RPi2 is at least as fast as some of the low end systems Haiku can be used on? It’s a test (and testament) of an efficient OS to see just how usable it is on such systems. ( … fond memories of hacking away on an Amiga 600 'til 2004 … )

Yes, the Pi2 is a better target mainly because it has an ARMv7 CPU and not ARMv6. A lot of ARM boards are using ARMv7 and this version of the ARM architectures standardizes a lot of things (MMU, etc) which means we need to write our code only once for all these things.

The reason we have started our ARM port (well, the most recent attempt) targetting the BeagleBone xM is that there is support for it in QEMU. At early stages of development this is quite helpful to debug things. Once we get the kernel booting on one board, porting it to other boards is just a matter of writing drivers (for the USB, ethernet, wifi, etc as usual). The hardware is still a bit more diverse than in the PC world, which means a lot of drivers will be needed. But, some boards have "everything over USB" which means our existing USB drivers could be reused. Once we get the USB stack running, that is...

Guys if some shitty linux version can collect 1.8 million on kickstarter we can do it also,my God,would you allow me to set up something?

Sure go ahead!
Don’t let anyone stop you!

ok I am working on it,nothing can stop us now,give me a week.