Talk:Cloud computing/Archive 3: Difference between revisions

Revision as of 03:53, 10 April 2016

Revision as of 03:20, 25 July 2016

Line 826:

This statement is incorrect. RJE was never referred to as time-sharing. Time-sharing was always an interactive experience involving a TTY terminal, a 2741 Selectric terminal, or something like that. In contrast, Remote Batch involved a large piece of hardware that was essentially a remote cardreader / printer device. They were called RJE terminals. I forget the IBM designation for them. Control Data also had them and I presume Univac and Burroughs did too. [[User:WithGLEE|WithGLEE]] ([[User talk:WithGLEE|talk]]) 22:27, 5 December 2014 (UTC)

:Yes RJE was not time-sharing. As for IBM's name for RJE, I remember IBM's [[Job Control Language]] having support for RJE jobs but Google cannot find anything about it. [[User:Sam Tomato|Sam Tomato]] ([[User talk:Sam Tomato|talk]]) 18:43, 30 January 2015 (UTC)

== Origins of "Cloud" ==

The cloud goes back farther than this article suggested by rAJNI KANTH

1985 source, book on ISDN:

https://books.google.com/books?id=UYcoAQAAMAAJ&q=packet+switched+diagram+cloud&dq=packet+switched+diagram+cloud&hl=en&sa=X&ei=729TVej7HdTZsAT984C4DA&ved=0CCoQ6AEwAQ

Make extensive use and reference to clouds in network diagrams.

Here's a 1988 source on the subject:

https://books.google.com/books?id=9M4SAQAAMAAJ&q=packet+diagram+cloud&dq=packet+diagram+cloud&hl=en&sa=X&ei=AWxTVfn8LuSOsQTJ04G4Dg&ved=0CD4Q6AEwAw

"Packet switching networks are universally represented on network schematic diagrams as a cloud. Presumably, data enter the cloud at one end and find their way miraculously through the fog to their proper destination at the other end."

:[hmm... looks like the date on this article might be wrong - [[User:Battling McGook|Battling McGook]] ([[User talk:Battling McGook|talk]]) 18:03, 13 May 2015 (UTC)]

Another 1988 source:

https://books.google.com/books?id=MtNrAAAAIAAJ&q=packet+switched+network+cloud&dq=packet+switched+network+cloud&hl=en&sa=X&ei=cG5TVYGzH4O_sQT4poCwBw&ved=0CDAQ6AEwAA

"When a packet enters the network 'cloud' from an end-point device, it must be routed to the destination end-point device by the packet switches forming the network."

And from 1989:

https://books.google.com/books?id=2GpPAAAAMAAJ&q=packet+switched+network+cloud&dq=packet+switched+network+cloud&hl=en&sa=X&ei=cG5TVYGzH4O_sQT4poCwBw&ved=0CEIQ6AEwBA

"Packet-switching Packet-Switched Data Networks (PSDN) are usually represented as a cloud with the legend X.25 inscribed on it, implying that it doesn 't matter what goes on inside the network as long as the data arrives."

To summarize, the cloud was useful in diagramming end-to-end communications in a very large network, where you really didn't care what was happening in the very complicated middle part of the diagram/ Otherwise network diagrams of complex networks would have to needlessly show a vast number of internal links that were utterly beside the point when discussing end-to-end communications.

This is also why the "cluster of servers" notion that the article talks about is completely ridiculous. First, the use of clouds to represent large sections of networks predates the modern version of the Internet. These early networks were unlikely to even have clusters of servers. Further, it's obvious that the cloud represents a broad chunk of network, while any cluster of servers would have been one single spot on a network diagram.

The reason the "Cloud" became synonymous with the Internet was because the Internet is one big giant packet-switched network. And the very idea of putting things in "the cloud" so you don't have to worry about them comes directly from this original usage of the cloud to represent a bunch of stuff out there who's function you didn't have to worry about.

The sources are clear, and that part needs to be updated. My opinions on how this turned into the modern usage of "Cloud" are my own, and unless and until someone says it out in the real world, you can't really put it in the article, but hopefully you can at least see how obvious it is that this is the real origin. [[User:Battling McGook|Battling McGook]] ([[User talk:Battling McGook|talk]]) 15:47, 13 May 2015 (UTC)

== Removing the 1950s section ==

I have BIG BIG problems with the section on the 1950s, and I plan to remove it completely.

In the 1950s, there was only one model of computing. There was no network and no cloud. People used big centralized mainframes for their computing because that was the only model in existence. Dumb terminals were just "terminals", because there were no smart terminals. There were no servers, and hence no clients, let alone no thin clients. Thin clients are actual computers that rely on network store. Dumb terminals are simply interfaces into a mainframe. The concept of time sharing was basically the polar opposite of cloud computing. Time-sharing arose because there was more demand than there was computing time. Cloud computing arose (in part) because there was excess computing time being wasted.

This entire section is basically a fractured fairy tale that never happened.

I'll let this comment sit for a week or so before taking action. [[User:Battling McGook|Battling McGook]] ([[User talk:Battling McGook|talk]]) 15:50, 6 July 2015 (UTC)