> The one thing you will find almost as a theme with Linux design is that there is minimum of "future" it. That applies to most of the Linux user space as well. This is a good thing and a bad thing.
Maybe because it is mostly about the "past", bringing enterprise UNIX and mainframe designs into Linux?
That applies for the core kernel. But drivers/hardware support on the other hand is pure Linux way of doing - don't think any mainframes or enterprise UNIXes had to bother with multi-GPU systems, so what you see in that area of Linux design is pure, quick to market hackery aided by delay-design-until-it-no-longer-works-without-it.
It's depressing in a way but that's the best we got when it comes to a hackable Open Source OS that you can somehow run on your hardware and even patch things you don't like.
SGI, Sun, HP, even NeXT supported multiple displays, frame buffers, and GPUs.
You could put 2 or 3 NeXTdimension cards in a NeXT Turbo Cube, for example - each with its own i860 running Display PostScript and something like 32MB of RAM - and use a display attached to each.
And the other workstation vendors had even fancier graphics subsystems and options. You could get real time stereo (LCD shutter) 3D with full Z-buffer and overlay planes for UI use (requiring a few dozen MB of VRAM) for the cost of a sports car. And with some systems use several at once.
Find some old copies of UNIX World or Personal Workstation or BYTE from the era...
May be GPUs are a bad example - for USB, FireWire and most of the other driver code and supporting infrastructure that constitutes the Linux kernel, the point (that its design is mostly uninfluenced by older UNIX) still stands. Besides the old UNIX vendors did not seem to have contributed much in terms of Linux GPU stack, maybe because that's not where the money was/is. SGI contributions for e.g. are in the area of FS and scalability predominantly. (I am aware that some places use Linux workstations for Graphics - but Nvidia essentially ships there own Xorg replacement stack along with highly kludgy properietary driver - nothing remotely related to great design there.)
Filesystems, TCP/IP stack, Multi core scalability etc are the areas that most benefited from old UNIX design. (Questionable in case of TCP/IP - Alan Cox rewrote it and I am not sure if he was influenced by any older UNIX implementation or not.)
Yes, but still, how much of IRIX's GPU stack did SGI contribute to Linux? How much of the current DRI/DRM/GPU Driver infrastructure is inspired from old UNIX designs? The point I am trying to make is that core kernel code is influenced by old, tried and trusted UNIX designs even if it wasn't up to the snuff in the scalability area for a long time. But the rest (incl. GPUs, USB stack, V4L2, driver support code (kobject) etc.) is a different story altogether - where old designs were either inaccessible or largely inapplicable due to different requirements and that's not an area of the Linux kernel anybody is proud of.
Maybe because it is mostly about the "past", bringing enterprise UNIX and mainframe designs into Linux?