There seems to be a inverse square law of Moore's Law at work between the amount of available processor power and the actual end-user applications neing used on a typical day seem to be inverse perportiant and the actual ratio is apparently a constant over time. I have always wondered why UNIX and FreeBSD in particular (vs. Windows or even the old non-BSD based MacOS) have a constant that is much closer to the end of preferring to idle the processor then have it chase down some low priority daemon that never gets called.
One thing that occured to me in this process is that there is most defently an growth curve over time of the ratio of meta-data/code and the actual engine (local or remote). This force tool developers to err on the side of caution when writing any middle ware and/or build utilities... thus these increase the ratio on favorablely and then the OS and App need to compensate for this caution by a more complex API which then just starts the cycle all over again... I have always marveled about how UNIX and BSD in particular avoided the feature and every possible case handled heavy OS and as a result the upper layers need to conform to a much simpler API (thus dampining the cycle)... comments? _______________________________________________ freebsd-chat@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-chat To unsubscribe, send any mail to "freebsd-chat-unsubscr...@freebsd.org"
