On Apr 27, 2010, at 11:26 AM, Armin Faltl wrote: > >> Remember, the bare hardware without any software at all has the greatest >> potential. Every line of code added to the software system takes away from >> that potential. This is necessary, of course. You have the hardware for >> specific purposes, and the software serves these. But one should not ignore >> the cost in lost capability. >> >> > > Let's assume you got an operating system, that let's an application take > total control of the hardware, like > DOS did. Clearly the OS provides a service by loading the application. Then > the application overwrites the > OS with data, so as to not waste any space before it delivers the result and > after producing the output it > terminates by going into an undefined state, so as not to waste any space for > termination code. > > So how does this OS limit the hardware? - sure it limits the program length > available for the application,
That's one. It may even prevent the use of minimal hardware: the application may require fewer resources than the OS. Another disadvantage of an OS is that booting straight into the application is faster. It's not so much that potential slips away rapidly with added code, but modern programmers are so good at producing vast volumes of code and so poor at producing well-factored code. This seems to be a particular problem for pcb, where there's a constant drumbeat of questions, both here and on the chat, of the form "why doesn't feature x work with feature y?" > but in practice this opposed, orthogonal, features, capabilities is all ... > How do you define a capability of a program with 0 (zero) features? Potential versus capability. Potential is essential for capability. So are features. But features take away potential. This is especially true for features considered in isolation, without consideration of factoring. The bare hardware John Doty Noqsi Aerospace, Ltd. http://www.noqsi.com/ j...@noqsi.com _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
