Bobby, yes the events of the entire last decade have been bullish for NC/terminals/and "the way it ever was".
Important... power to the masses corporate PC on every desk is realized to be a wasteful joke. No wonder they call it _MS_NBC. MS is an entertainment company! I get it now... see how I am laughing as I reinstall for the umpteenth time?
10 years of experiment. Of ever fatter clients pulling ever greater loads off servers ever farther away is over. No gov't decree, no "Tomorrow morning at 6AM we are switching to the metric system." It will just slowly, then quickly, happen.
If you cut me you will find packets and cells, they are decidedly un-blue. But you gotta figure there is a reason that IBM became the greatest computer company of all time, AND is experiencing a re-awakening.
The very, very smart people that invented the original multi-client computing architecture, actually did get it right. Their version wasn't scalable or flexible enough(they worked with what they had right) but the basic idea was/is the only answer for business computing.
I find that when people have a method (in this case centralized computing), then become enamored with a new method(PC/server), but return to the first within the working lifetime (memory) of the people using/choosing the method, that returned method becomes un-displaceable.
"Oh that! Yeah we tried different ways but 'old faithful' here, this is how we do it.". It is called experience. There is such a thing as "group" and even "corporate" experience and memory. Highly distasteful experiences (such as PCs on company desktops for non-content creators) are usually permanent memories.
The reason I opt for three-tier is that I want (1) device talking to my backend, representing many endusers. Without getting into what is an extremely interesting discussion but quite lengthy and boring for non-geeks; basically a lot of problems go away or are turned into manageable ones in this scheme. One or two new ones are introduced but the really ugly ones go away and the new ones are tiny or at least modelable in a way that allows _reasonable_ suppositions about the future to be applied.
In the current popular two-tier scheme many of the problems are of scalability and cannot be solved within the constraints of that scheme. How did Kirk solve the "Kobiyashi Maru"? He changed the parameters.
FWIW Dept: Business is FINISHED with networks and computer as neat toys. You yank the net in any US company over 50 people and you might as well tell the folks to go home, you're going out of business. The "gee this is fun to play with" technoids are being run out of MIS or they have become mature and conservative, having crashed Finance once or twice... ;( (I'll never tell...)
Later days... |
