[LINK] Re: The Next Ten Years

Glen Turner gdt at gdt.id.au
Thu Aug 9 22:42:37 AEST 2007


On Thu, 2007-08-09 at 08:55 +1000, Tom Worthington wrote:
> The next ten years is easy. The Minister wanted to know what would 
> happen in the next 30 years.  ;-)

The year is 1977. Could you have anticipated:
 - the Internet
     probably not.
 - the spreadsheet
     unlikely, there was a fierce academic debate about
     computers and "knowledge workers" and the big guns
     were on the side of their complete replacement by
     artificial intelligence rather than augmenting them
     with productivity-focused computing support.
 - the ubiquity of digitisation
     probably not. Imagine arguing that converting things
     to 0 and 1s would catch on because this was convenient,
     cheaper and smaller than simple analog processing.
 - miniaturisation
     almost everyone got this. many people thought it
     would be a communications device too, although
     they would probably have been accused of too much
     science fiction.

It's also interesting what hasn't changed. Transport changed
immensely between 1947 and 1977, especially with containerisation.
But it hardly changed at all 1977 to 2007.  So it's possible
that we might go through a similar era slower technical advance
and of consolidation.

We're at the limits in a lot of communications technolgoies
now. Battery lifetimes now improve in single digit percentages,
we can't build faster CPUs without water cooling, hard disk
drive have only doubled in speed in the past ten years.

> I have arranged for Kevin Miller, a green architect,  to talk on how 
> to build environmentally friendly offices in Canberra 
> <http://education.acs.org.au/mod/resource/view.php?id=3635>. I have 
> asked him, if IT people replace desktop PCs and phones with low power 
> thin clients 
> <http://www.tomw.net.au/blog/2007/05/thin-linux-workstations-to-save-planet.html>, 
> how will that change office design. One option is to eliminate mains 
> power to office desks and use Power Over Ethernet instead for 
> computers and task lighting. Using a low voltage does not itself save 
> power (it is less efficient than mains power), but it would stop 
> office workers using power hungry devices.

POE is a tad ambitious. That gives you 18W per socket. That's
currently not enough for a CPU and LED-backlit screen, but it's
tantalisingly close.

The "less efficient that mains power" depends were you get your
power from. -48VDC is a natural match to battery systems, and
thus to solar systems. If you run solar power through an inverter
and then a rectifier (in the PC, remember the aim is to get 12VDC
and 5VDC for the motherboard) you are wasting a about 10% of the
power on needless conversion to AC.

Also, AC power supplies are very inefficient. There's a move in
the US called "80plus" to get PC rectifier performance above 80%.

I suppose my question is -- why do you think there will still be
an office as we know it today?  With better communications
technologies and higher travel expenses and time why won't those
how can do work from home.

We've already seen the move by a lot of workers to establish a
second office at home. I'm suggesting that this will become the
primary office for a lot of people.




More information about the Link mailing list