Not everything will be electrified, but electrification will change everything.
-t
David Evans
What a beautiful application of a philosophical tool to cast light on a hype’d up term. The nonsense of the statement immediately obliterates the use of the phrase anywhere else. I’m sure I will be reflecting on the wisdom of this kōan for some time to come, and wondering where else such a technique can be applied. Brilliant!
It’s timesharing on servers – some of us remember that.
Nick Carr
I remember timesharing on a mainframe. As far as I recall, the timesharing networks themselves weren’t connected with one another. That’s kind of a big change.
linnell.myopenid.com
Nick, as you mentioned, timesharing systems usually had private networks, but many were interconnected through “public” packet switching networks such as Telenet http://en.wikipedia.org/wiki/Telenet, which provided both terminal-to-computer and computer-to-computer connections.
A few timesharing systems connected to Arpanet, the predecessor of today’s internet. I recall tinkering with email and file-transfer connections via Arpanet all over the world from MIT’s timesharing systems in 1973 when I was a student.
The electricity industry has changed over the years too, in terms of networks of providers (Enron!) – but we don’t talk of “cloud electricity”. From the user’s point of view, it’s just a remote provider, rather than having one’s own generator.
Note that even with a massively well-built electric grid, we still extensively use small portable sources of electricity (i.e. “batteries”). But, granted, the market for sizeable personal generators is rather minimal (mostly rural areas and campers).
fishtoprecords
Actually, most of the early Arpanet machines were timesharing systems. Typically DEC PDP-10, or Tenex machines, and some Honeywell Multics machines. They were very common in the research community (aka academic) of the time. Typically the mainframes uses something smaller and cheaper, such as a PDP-11 to talk to the network.
What is different today is that the cloud will assume a far more capable end device than the “terminals” of old.
The Greenberger utility model was based on the existing communication grid at the time, the analog PSTS. As Nick has pointed out, the open connectively of the Internet has potentially made the Internet a computing platform in itself. But, ultimately, if you provided storage, bandwidth or CPU cycles over the Internet you have to charge for it which is the basis of a utility model which Greenberger points out that many years ago. When you talk about the “Internet” what you really mean is the public TCP/IP network with its associated protocols/services. (Yes, you skill have to open a socket to do anything on the internet!) Its addressing interface is “logical” construction rather than a physical one, the underlying physical hardware are much more grid like and aren’t too much different that the technology that Greenberger was working with to in 1964.
No “new generation of computing platform” has ever fully replaced the last one… We’re moving from Mainframes to mini-computers to PCs to client server to web to a vast array of devices to cloud… and yet we still have mainframes with us, and will for a long time.
To me what is fundamentally different about the rise of the cloud as a computing platform is that there is inherently a notion of interoperability… apps across all these platforms has to work together! :)
Whether you’re talking apps on iphones, a tvs, cars, picture frames, industrial equipment, PCs, servers or what have you… as classes of apps, each will see examples arise where they are made better by adding cloud capabilities to them.
[Disclosure: I work for Microsoft as an evangelist for cloud computing]
Whether you’re talking apps on iphones, a tvs, cars, picture frames, industrial equipment, PCs, servers or what have you… as classes of apps, each will see examples arise where they are made better by adding cloud capabilities to them.
It feels good to work in this industry each day becomes a challenge. The social media networks are growing that fast and developing new interactive ways to communicate, frequently amazes me in what will be next.
I think the ‘cloud’ potentially creates so many more issues which, although being more solvable because of the fact that the ‘cloud’ is a very powerful network of computers, those problems are ultimately scaled up meaning they are still as much of a problem as they always have been, just bigger ones!
Let’s wait and see if the cloud covers the mountain or the mountain engulfs the cloud
Predeep Kumar
Great morning, I wish to say thank you for an exciting web site about a subject I have had an curiosity in for some time now. I have been looking in and reading the commentary and only wanted to voice my many thanks for giving me some pretty exciting reading material. I look forward to reading more, and getting a more active part in the talks here, whilst picking up some knowledge as well.
Very true.
Don’t expect the Legacy Enterprise Apps to move to cloud. But, do expect that the Legacy App will be a replaced with an off the shelf SaaS service.
There is no cloud. There is only a finger pointing at the cloud. To see the cloud, it is necessary to see beyond one’s finger.
Not everything will be electrified, but electrification will change everything.
-t
What a beautiful application of a philosophical tool to cast light on a hype’d up term. The nonsense of the statement immediately obliterates the use of the phrase anywhere else. I’m sure I will be reflecting on the wisdom of this kōan for some time to come, and wondering where else such a technique can be applied. Brilliant!
A student asked Master Yun-Men “Not even an instance has arisen; is there still a cloud or not?” Master replied, “Amazon!”
With all due respect to our Fearless Leader, Martin Greenberger pointed the coming of the cloud in The Computers of
Tommarrow twenty years before John Gage utter the matra “The Network Is The Computer”.
There is nothing new under the sun.
It’s timesharing on servers – some of us remember that.
I remember timesharing on a mainframe. As far as I recall, the timesharing networks themselves weren’t connected with one another. That’s kind of a big change.
Nick, as you mentioned, timesharing systems usually had private networks, but many were interconnected through “public” packet switching networks such as Telenet http://en.wikipedia.org/wiki/Telenet, which provided both terminal-to-computer and computer-to-computer connections.
A few timesharing systems connected to Arpanet, the predecessor of today’s internet. I recall tinkering with email and file-transfer connections via Arpanet all over the world from MIT’s timesharing systems in 1973 when I was a student.
The electricity industry has changed over the years too, in terms of networks of providers (Enron!) – but we don’t talk of “cloud electricity”. From the user’s point of view, it’s just a remote provider, rather than having one’s own generator.
Note that even with a massively well-built electric grid, we still extensively use small portable sources of electricity (i.e. “batteries”). But, granted, the market for sizeable personal generators is rather minimal (mostly rural areas and campers).
Actually, most of the early Arpanet machines were timesharing systems. Typically DEC PDP-10, or Tenex machines, and some Honeywell Multics machines. They were very common in the research community (aka academic) of the time. Typically the mainframes uses something smaller and cheaper, such as a PDP-11 to talk to the network.
What is different today is that the cloud will assume a far more capable end device than the “terminals” of old.
Conceptually, its the same concept.
The Greenberger utility model was based on the existing communication grid at the time, the analog PSTS. As Nick has pointed out, the open connectively of the Internet has potentially made the Internet a computing platform in itself. But, ultimately, if you provided storage, bandwidth or CPU cycles over the Internet you have to charge for it which is the basis of a utility model which Greenberger points out that many years ago. When you talk about the “Internet” what you really mean is the public TCP/IP network with its associated protocols/services. (Yes, you skill have to open a socket to do anything on the internet!) Its addressing interface is “logical” construction rather than a physical one, the underlying physical hardware are much more grid like and aren’t too much different that the technology that Greenberger was working with to in 1964.
No “new generation of computing platform” has ever fully replaced the last one… We’re moving from Mainframes to mini-computers to PCs to client server to web to a vast array of devices to cloud… and yet we still have mainframes with us, and will for a long time.
To me what is fundamentally different about the rise of the cloud as a computing platform is that there is inherently a notion of interoperability… apps across all these platforms has to work together! :)
Whether you’re talking apps on iphones, a tvs, cars, picture frames, industrial equipment, PCs, servers or what have you… as classes of apps, each will see examples arise where they are made better by adding cloud capabilities to them.
[Disclosure: I work for Microsoft as an evangelist for cloud computing]
“The mathematics is not a science, but any knowledge becomes a science only after the mathematics comes to it” )))
I don’t think that ‘cloud’ is so common and wide generality. Tomorrow we will have new paradigm and what?
Refering to electricity analogy it looks like:
“… the electricity will move into everything” – correct
“… direct currtent will move into everything” – correct with strong limitations
Whether you’re talking apps on iphones, a tvs, cars, picture frames, industrial equipment, PCs, servers or what have you… as classes of apps, each will see examples arise where they are made better by adding cloud capabilities to them.
How much good it will do you to entrust your data to the “cloud” could just be witnessed. Anybody own a T-Mobile Sidekick?
It feels good to work in this industry each day becomes a challenge. The social media networks are growing that fast and developing new interactive ways to communicate, frequently amazes me in what will be next.
I think the ‘cloud’ potentially creates so many more issues which, although being more solvable because of the fact that the ‘cloud’ is a very powerful network of computers, those problems are ultimately scaled up meaning they are still as much of a problem as they always have been, just bigger ones!
Let’s wait and see if the cloud covers the mountain or the mountain engulfs the cloud
Great morning, I wish to say thank you for an exciting web site about a subject I have had an curiosity in for some time now. I have been looking in and reading the commentary and only wanted to voice my many thanks for giving me some pretty exciting reading material. I look forward to reading more, and getting a more active part in the talks here, whilst picking up some knowledge as well.