On Thu, 6 Jan 2000 David Schwartz wrote:
>
> Sure, but those tasks will be done by 64-bit chips even though they're
> suitable for 32-bit chips. Everything will. It's the same reason I'm using a
> P3-500 to write this email even though a '386 would suffice for the task. We
> redefine the tasks to fit the available hardware.
I don't think anybody today can even guess at what computers in the year
2038 will be like.
For that reason I find this whole argument now to be just a tremendous
waste of bandwidth.
64 bits? Hell, that will be ancient; the whole concept of bus structure
will probably have changed radically by then. I expect we'll see things like
memory totally integrated with the processor, processors that can actually
reconfigure their logic circuitry on demand or as they "learn", etc, most
likely making this entire argument moot.
Maybe in the year 2028 it will be worth starting to think about it, by
then we might be able to take at least a guess as to what computers might be
like ten years down the road. By 2033, maybe it will make sense to actually
start working on the implimentation of a fix if the problem hasn't become moot
by then.
As someone pointed out earlier, we can extend the current clock another
68 years by merely going from a signed to an unsigned 32 bit interger.
My first computer I owned was an 8-bit 2mhz machine with a whopping 48
kilobytes of RAM and about 140kb of floppy disk space. Graphics were 48x128
block graphics, monochrome, on/off no gray scale. Text was 16 lines of 64
characters.
This was only 18 years ago! How much has changed since then? Now we're
on the verge of 1ghz CPU's, that was a 14 million dollar Cray back then. With
all that change in the last 18 years, and the rate of change appears to be
accellerating, how much do you think computers 38 years from now will resemble
todays computers? Not very much I think.
So why don't we table this discussion for another 28 years or so when we
might have a very very vague idea of what computers will be like in 2038.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jan 07 2000 - 21:00:08 EST