Becoming the next big OS

[Schol-R-LEA]

Well, that's something at least. I'll shut up about this now, as I doubt I can get through your Great Wall of Ego-Commitment on this topic.

Finally, note that when I came up with the RSVP idea, two decades ago, I played around with the concept for a while before setting it aside. I still thought it was a good idea (though I now can see a lot of flaws in it), but I never pursued it even tot he degree of writing a simulator. Why? Because I didn't know enough about the subject, and knew I didn't. I would like to get back to it someday, either to see whether it could work as it is or see what lessons I could learn from the idea if it can't, but I didn't try to make it reality right then and there.

And this is where you and I differ: you are wearing a lemon-juice mask about SUBLEQ and CPU design in general. You don't know enough about them, and worse, you don't know that you don't.

[Geri]

well i obviously cant be an expert in everything, i dont know much about chip design. if i optimize and develop the dawn os from step by step, and hardware boys do the same to the hardware, any occuring problems will be eventually solved (if they occur even).

[Kazinsal]

Not sure why you would want to implement something like this in an FPGA or even expect it to work worth a damn.

You do know that even x86 has been run as an emulation/translation layer on top of a series of different RISC microarchitectures for 20 years now, right? AMD filed a patent for RISC86 as used in the K6 cores in 1995.

[Geri]

yes. and what is your point

[Brendan]

Hi,

This is purely theoretical. What are the chances at any time of somebody developing the next Windows, macOS, or Linux? Is it an absolutely zero chance?

The chance is not absolutely zero.

For rough estimates, I'd suggest:

• 99% of OS developers don't try, and therefore can't create a successful OS
• 0.9% of OS developers try to create a successful OS, but "real life" (work/career, spouse, kids, health, etc) gets in their way and prevents their OS from reaching "critical mass" (see note)
• 0.09% of OS developers reach "critical mass", but don't succeed for technical reasons (e.g. "not technically superior enough to convince end users to switch from an established OS they already use").
• 0.009% of OS developers reach "critical mass" and create something that is "technically superior enough to succeed", but fail for non-technical reasons (marketing, legal problems, etc)
• 0.001% of OS developers become successful

Note: "critical mass" is the point where an OS is impressive enough to sustain it's continued development if the founder/s abandon it. For example, if Linus Torvalds decided he's not interested in working on Linux anymore then Linux would still continue because Linux has reached "critical mass"; but if I decided I'm not interested in working on my OS anymore then the project would die (become "abandonware") because my OS hasn't reached "critical mass".

How many years will it take for a team of at minimum 5 members? Is that still too small?

This is the wrong way to look at it - the number of people changes over time.

It's best to think of an OS project as having several stages:

• Pre-birth: This is where fundamental pieces are being designed and created (boot code, kernel, a small number of generic drivers, a file system, an initial GUI, etc). For this stage you want a very small team (no more than 3 people, and often a single person) because there's multiple problems caused by having too many people involved (increased project management overhead/coordination, bike-shedding, design by committee, etc).
  Childhood: This is where the OS is becoming increasingly more useful (people improving pieces, creating utilities, creating more drivers, etc). For this stage the number of people involved should grow, starting from the initial tiny team to maybe about 100 volunteers working on different things.
  Adulthood: This is where the OS has become a useful/marketable product, and people start writing applications for the OS (while continuing to improve pieces and writing more drivers, etc). For this stage the number of people (developers) involved should continue to grow.

For the amount of time; I'd estimate about 10 years for "pre-birth" and then another 10 years for "childhood" (but it would depend on far too many things to estimate with any accuracy).

What happened to championing variety? It seems that everything has only a few narrow choices *cough* DIRECTV/Dish Network *cough* Don't get me started on that.

I'm not too sure what you mean; but "different" is hard (much easier to drive along a highway that millions of people have used before you, much harder to fight your way through a jungle creating your own path).


Cheers,

Brendan