On 04/16/2014 11:25 PM, Linda Walsh wrote:
Anton Aylward wrote:
On 04/16/2014 04:45 AM, Dirk Gently wrote:
In contrast, SystemD tries to do EVERYTHING, and does NONE of it well -- which is straight out of the Microsoft/Windows way of doing things.
There could be a perfectly good explanation why the design follows MS's serviced and why even the config files look like microsoft setup file (besides under the table payments).
There could also be a perfectly reasonable explanation why Microsoft (and others) have followed UNIX/Linux in so many ways, including, eventually as shell rather than a gui. After all, Linux is open source. The differentiator in my mind is that Windows has the Directory, well aren't there two of them?, while Linux sticks to text files for configuration. Or is MS not following Linux in that and giving up on the directory as a Bad Idea? We _can_ make the root FS of Linux RO, but the way the directory works you can't do that with desktop Windows, you have to do other gyrations to use Windows as an appliance with a RO FS. Its just another way that Linux can be strapped down to be resilient in the face of a malicious environment. Ultimately Microsoft copied UNIX is that UNIX pioneered writing the OS in a HLL - 'C'. Back when I was working on BSD4.1 and 4.2 I recall the race between Bill Joy and David Cuttler. All Cuttler's efforts to speed up VMS by carefully tuning the assembler were matched by Joy in minutes coding in C. It must have been frustrating for Cuttler as he eventually stated he would never write another OS in assembler. Hence WNT (which is to VMS the way HAL of 2001 is to IBM, so demonstrating the 'if you can't create then copy' principle) was written in C. The point isn't that systemd follows some other design. In fact its very specific to Linux. From what I see it would take quite some re-working to incorporate it into some of the Big Iron version of UNIX, but one never knows. The point is that the sysvinit technique had reached the limits of its capability. I'm a great advocate of shell and CLI and laughed out loud when I heard of Window's "Power Shell" -- it proved the GUI approach was totally unsuited to the kind of 'automation' that system administration, especially of large numbers of possibly virtual machines demands. But while the shell is a great Swiss army knife, it is just that. For specific tasks it can be far from the best solution. Its not that you can't build a dispatcher with the shell; I have. But after a point the shell model gets inefficient. Prime among then was that the sysvinit model was sequential and was most certainly not a dispatcher in the sense of monitoring the tasks asynchronously. UNIX/Linux is not set in stone. If it was we'd still be running something out of the 1970s or 1980s, V6, V7 or possibly an early BSD if you insisted on having networking for email and netnews, though that networking might be UUCP rather then TCP. -- When languishing for solutions, don't ask "Have I got the correct answer?" The correct question is "Have I got the correct question?" -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org