mention office suites

There should be a mention here about online office suites : they being examples of cloud computing Sanjiv swarup (talk) 01:17, 5 March 2008 (UTC)

I would agree with that. The Register

provides a good reference. Lester 21:18, 15 April 2008 (UTC)

ty for your inputs. have done the needful Sanjiv swarup (talk) 04:35, 27 April 2008 (UTC)

Introductory sub section

The foll. line is meaningless.

Common visualizations of a cloud computing approach include, but should not be considered to be limited by, the following:

I request someone to dele this line .

Sanjiv swarup (talk) 09:51, 14 June 2008 (UTC)

deleted Sanjiv swarup (talk) 03:15, 15 June 2008 (UTC)

WTF?

The first sentence of the article conveys no meaning or significance or reason. What?--Shtove (talk) 21:54, 2 March 2008 (UTC)

Unfortunately, the rest of the article doesn't do much to explain the topic, either. I'd like to slap this with an expert-subject tag, because cleanup and clarification cannot take place if the subject matter is unintelligible in the first place. I'd like to remove the spam tag, since the references seem to point to decent sources, and the external links has its own tag. Any objections to this? Wouter de Groot (talk) 13:38, 19 June 2008 (UTC)

Given it is currently a marketing term, why not focus not on defining what it is -that is too vague- but in the growth of the term? SteveLoughran (talk) 21:25, 19 June 2008 (UTC)

100% agreed. Fixed now. samj (talk) 08:44, 1 August 2008 (UTC)

rename to read as "see also"

wrt sub section = Additional Cloud Topics

suggestion = rename to read as "see also"

from Sanjiv swarup (talk) 08:58, 2 July 2008 (UTC)

done. samj (talk) 09:38, 1 August 2008 (UTC)

SETI@home/Folding@home/etc.?

After reading through this article, I feel more confused than before regarding Cloud Computing. Since "Cloud" measn the general internet, so does distributed computing projects like SETI@home and Folding@home belong to the concept of Cloud Computing? As I understand from the article, they are basically raw computing power coming from the cloud to solve a problem, so are they cloud computing or not? —Preceding unsigned comment added by Ufopedia (talk • contribs) 08:44, 10 June 2008 (UTC)

Added per your suggestion - it doesn't matter whether the computing is done on a grid, by the neighbour's screensaver or a room full of monkeys... that abstraction is one of the fundamental features of cloud computing. samj (talk) 09:16, 1 August 2008 (UTC)

Cloud services

The mention of Amazon here seems to be advertising. There are at least a thousand other such Vendors. The author of this para should simply insert a DMOZ link here for a lit of such vendors.

Disagree - I do not believer there are thousands of computing clouds available at this time. See "Google and the Wisdom of Clouds" from BusinessWeek: http://www.businessweek.com/magazine/content/07_52/b4064048925836.htm

Disagree as well, Amazon is one of the archetypes in cloud computing as well as a very early leader. Amazon is regarded as one of the companies (along with Google) to illustrate and execute cloud computing's potential.

Disagree as well. Anyone can expose a web service. It takes security, scalability, support, billing, etc. to make is a real cloud service.

For the record

( section copied from User talk:CliffC )

rv collection of unhelpful edits, sorry if anything good was lost - spam, damaged paragraph, removal of apparently valid links, tag indicating work underway but no work undertaken

How were my edits unhelpful? They were accurate and well referenced, unlike the rest of the article which is a regurgitation of some random paper which differs substantially from what the industry and its analysts have to say.
Why are you apologising on my behalf, or making edits that would result in anything good [being] lost?
What was spam?
What paragraph was damaged (I had added a new paragraph)
What 'apparently valid' link was removed (except for Redundant Array of Inexpensive Servers, a term referenced once by a paper over a decade old which is likely to be deleted soon and which has very little to do with the topic?
How do you determine no work undertaken when the tag had just been added and should not be removed unless there have been no edits for several days per Template:Underconstruction?

samj (talk) 06:07, 28 July 2008 (UTC)

Actually I was just thinking that these edits may indeed seem out of context with the rest of the article. There has been much discussion on this topic recently, and I have spent the last few days working on a consensus definition which you can read about here: The Cloud and Cloud Computing consensus definition?. Maybe you can read through this, and the references at the bottom as well as the Forrester and Gartner reports if you are really interested in this topic.

samj (talk) 06:36, 28 July 2008 (UTC)

Hi Samj. Before I made the reverts with the edit summary you mention above, I reverted two edits by your account (the diff here shows the before and after of my revert). I should have left an edit summary, but my reasoning was that your edits added an apparent promotional link to ebizq.net. Promotional links violate WP:EL. In addition the link is to a blog, also against WP:EL, unless the author is notable.

Next, the collection of edits I reverted with the edit summary you mention (diff here), edits made by you and several other editors, seemed intertwined and not worth reverting separately.

"...spam, damaged paragraph, removal of apparently valid links, tag indicating work underway but no work undertaken"

spam - the misplaced link to gridbus.org has the appearance of spam bacause it was inserted in the wrong place. May not be spam, but the paper linked may not be a WP:RS either. I didn't bother to read it because of the link placement.
damaged paragraph - IMO at that moment replacing the lead sentences

Cloud computing is a style of computing where IT-related capabilities are provided “as a service” using Internet technologies to multiple external customers^[1]. Resources being accessed which are typically owned and operated by a third-party provider on a consolidated basis in datacenter locations

with

Cloud computing refers to computing resources being accessed which are typically owned and operated by a third-party provider on a consolidated basis in Data Center locations.

- appeared to damage the paragraph, that assessment was perhaps incorrect and simply a matter of editing style/taste. Perhaps I had become impatient with the number of edits over the past weeks that seem to come from someone with something to sell. But I did use the phrase "sorry if anything good was lost", and I do apologize for this one.

removal of apparently valid links - links to cloud articles in businessweek.com and infoworld.com
tag indicating work underway but no work undertaken - tag {{underconstruction}} added to article with the summary "article's a mess - needs an overhaul" by an editor with zero edits in the past except for three AfD votes for his favorite singer and musical group. Without a contribution to the article this looks more like vandalism than a constructive edit. I have no objection to the tag (which may be argued against by the article's regular editors), only to its source.

I have too much going on in the non-wiki world to read and comment on the consensus document at your web site, but I encourage you to mention it on the talk page so that you have a chance at a true consensus. --CliffC (talk) 20:07, 28 July 2008 (UTC)

Ok makes more sense now. I have hundreds of edits spanning back to 2005 and a bunch of others with SPAs so maybe I was logged out or you checked while the history was catching up to my recent rename (from an alias to my full name). In any case the existing article was based entirely on a (non-notable?) paper written by a lab working on grid computing and was quite detached from reality. As I have some time on my hands I've spent the last days trying to align the two. samj (talk) 08:43, 1 August 2008 (UTC)

Alright so now the article has been completely overhauled and properly referenced (per complaints made about it) it's pretty clear that my edits weren't spam so I'm archiving this. samj (talk) 08:37, 4 August 2008 (UTC)

Vague and confusing

This article can be written better than it is now.Chmyr (talk) 02:13, 12 April 2008 (UTC

Agreed, this article reads like some boring, nasal nerd who's trying to impress you with jargon instead of getting actual points across. I think many readers are going to walk away from this article still having no clue whatsoever what the general term of "cloud computing" is referring to. Once we cut through the hype, jargon and nerdgos... maybe we can actually have people leave this article with some understanding of it? Is that too much to ask? LOL

When it comes down to it, it's a fairly simply concept (not the implementation and all the derivatives, but the CONCEPT itself). Typically, it's using the power of multiple computers across the internet to work in unison (or you could say in harmony) to accomplish tasks that require a lot of processing power and storage that would otherwise overwhelm the capabilities of an average computer working alone. There, I said it. LOL Cowicide (talk) 00:34, 12 May 2008 (UTC)

I believe the article can be more simpler than what it is. The concept of Cloud computing loses it track at the bottom...

Dhoomady (talk) 11:01, 14 June 2008 (UTC)

I agree with Chmyr and Cowicide This article needs some attention to meet wikipedia standards. Kalivd (talk) 14:14, 14 June 2008 (UTC)

This page makes no sense at all. In fact, the picture of "The Cloud" with the corresponding caption summarizes this article perfectly, much ado about hot air.

So well put and simply stated above by Chmyr and Cowicide . I run an IT Knowledge Solutions company and am still amazed how complicated our profession make concepts sound to the not technical person. Lets get over it and help people move into this new age. Thank you for clearing this up for me. —Preceding unsigned comment added by 61.69.34.229 (talk) 10:59, 27 June 2008 (UTC)

I am planning on archiving this sub section. no action : all talk !! Sanjiv swarup (talk) 09:14, 2 July 2008 (UTC)

There is no point in archiving a discussion just because there has been no action. Please don't. --CliffC (talk) 12:24, 2 July 2008 (UTC)

I agree with previous authors, this article is misleading and has lots of erraneous statements. It's been written by incompetent people. I wish it wouldn't exist at all. I can't give all of my arguments here, because of unusual discussion format (Just a single note: Enomalism is not a cloud, but little piece of python code). We have an extended discussion on the definition of cloud at groups.google.com/group/cloud-computing If you have arguments against, let's do it in proper environemnt. Sapenov (talk) 03:04, 5 July 2008 (UTC)

i think it would be nice if there could be an article or something within Wikipedia that would match the standards of wikipedia.Anoopnair2050 (talk) 15:16, 21 July 2008 (UTC)

Thanks for your comments. I've spent the last days overhauling the article and I hope you are all more satisfied with it now. Feel free to contribute yourself. samj (talk) 08:51, 1 August 2008 (UTC)

Etymology

I assume there is some metaphor behind the use of the word "cloud" for "cloud computing" - does anyone know it? -- 13:52, 9 October 2007 (UTC) I believe the "cloud" refers to the abstract concept of the web or internet. It's used to indicate that you don't really care about the details, about who or what is out there in the cloud, it's just the cloud, the rest of the internet.--Bill.albing 15:59, 9 October 2007 (UTC) I believe that cloud refers to the fact that most tech architects use a drawing of a cloud when discussing the Internet or services available over IP. —Preceding unsigned comment added by 67.188.145.142 (talk) 01:16, 4 May 2008 (UTC)

Surely they draw it as a cloud because of the metaphor, it's not called a cloud simply because it just happens to be drawn as one. User:Jamie Kitson —Preceding comment was added at 08:49, 10 May 2008 (UTC)

See http://www.mercurynews.com/businessheadlines/ci_7124839 Westwind273 17:41, 9 October 2007 (UTC)

No, the cloud came way before 'Cloud Computing'...its the general purpose "some kind of network stuff" symbol used when you did diagrams using a tool like visio; stick the cloud between some boxes to imply the internet was between them. What's changed is that people are now putting storage and computation into that cloud, at least in their slideware. —Preceding unsigned comment added by SteveLoughran (talk • contribs) 09:32, 2 July 2008 (UTC)

Agreed, The Cloud has been around for ages on network diagrams etc. and while it is a metaphor for the Internet, the most important point is the more stuff you can push into the cloud the less you have to take care of yourself (consider the complexity of a WAN vs a VPN... and a diagram of both... usually the latter is the same as the former, only with a dirty big cloud covering over most of the complexity which is outsourced to the ISPs). samj (talk) 08:55, 1 August 2008 (UTC)

Technical jargons

This is all very confusing. I think we need to have a clear difference b/w Cloud computing, grid computing, clustering, multi-tenant, software as a service etc... Some times these terms are self contradictory.?--Mailtoram —Preceding comment was added at 09:28, 3 March 2008 (UTC)

I agree and have spent the last days improving the situation (multitenancy is a key characteristic of cloud computing and SaaS a subset), but there could still afford to be some clarity in terms of differentiating the other terms (esp grid/clustering). samj (talk) 09:10, 1 August 2008 (UTC)

Argh

Argh I cant believe people are using this term!—Preceding unsigned comment added by 58.28.155.207 (talk • contribs)

It seems that this is the last Marketing fad (and Marketing does creates new words especially in the computing world), and as WP is just the image of the real world, I believe we should use this terme too. This not a problem in my opinion, as if the concept settles on an other term we'll just have to do a renaming (a Redirect). This is how WP works --Kompere (talk) 14:53, 18 December 2007 (UTC)

Actually I think cloud computing has a good deal more potential than that as an umbrella term for all the other *aaS guff... I'd certainly much rather see us talking about cloud computing than bolting as a service onto everything. The question now is more about whether the vendors manage to derail it as they did grid computing... samj (talk) 08:53, 1 August 2008 (UTC)

Sounds like an updated version of the 1960's service bureau. I would like to see a comparison. talk (talk) 11:00, 1 August 2008 (UTC) —Preceding unsigned comment added by 68.58.152.238 (talk)

I've added a link to the service bureau article per your feedback. samj (talk) 08:43, 4 August 2008 (UTC)

Dell

Removed content from unreliable blog source:

While the general term "Cloud" had been used prior, the technology firm Dell has been attributed with coining the actual term "Cloud Computing." The term was reportedly first used by Dell in a press release on March 27, 2007.^[2] At this time, Dell had already begun the formal process of trademarking the term. Howeverm, in 2008, as the trademarking process came to a close, Dell found itself being largely criticized by the internet community for trademarking what many believed to be Public Domain term.^[3]

Sorry, I didn't think that a news article from the Wall Street Journal was considered an "unreliable blog source". Nonetheless, I removed the reference and tried to make the statement about Dell's trademark more neutral and factual. Updated reference includes government database entry on the trade mark. The press release by Dell has also been referenced. I do not believe the reference to the pertenent press release consitutes "Conflict of interest" as defined by Wikipedia.—Preceding unsigned comment added by 69.222.228.201 (talk) 13:53, 7 August 2008 (UTC)

The 'WSJ blog' is an opinion piece which has been repeatedly taken apart in the comments. Your unverifiable edits claim that Dell 'popularized' the term and according to IP2Location you are around 10mi from Dell. samj (talk) 19:30, 7 August 2008 (UTC)

Here's the edits as they were at the time of removal. It's not a controversy btw. samj (talk) 19:34, 7 August 2008 (UTC)

== Term Controversy ==
While the general term "Cloud" had been used prior, the technology firm Dell has staked claim to popularizing the actual term "Cloud Computing." Dell filed for a trademark on the term on March 23, 2007 ^[4] and used the term with the indicator "TM" in a press release dated March 27, 2007 ^[5].
In mid-2008, however, objections were raised to Dell's intention to trademark the term. Many in the internet community contended that the term entered the vernacular before Dell filed for the trademark, and that the term should be public domain.
The veracicty of these statements has not been substantiated. Google Trends shows 0 hits on the term "Cloud Computing" prior to 3rd quarter of 2007^[6] indicating minimal use of the term prior to that time. There are however, sporadic news articles containing the term "Cloud Computing" dating back to 2004^[7]
Regardless, the trademark case history shows that the Notice of Allowance furnished to Dell by the United States Patent and Trademark Office on July 8th, 2008 was rescinded on August 5, 2008^[8], and the case was returned to an examiner on August 6th, 2008.^[9]

Warned user 69.222.228.201 (talk) about WP:COI and WP:3RR. samj (talk) 19:35, 7 August 2008 (UTC)

Hey Sam, Anonymous user (John) here. Just wanted to write you a message, though I'm not sure this is the correct place.

You'll have to forgive me, I don't edit Wikipedia that much and the piece on cloud computing was my first real attempt to do so. I just wanted to let you know that the piece I wrote on Dell's attempt to trademark the term Cloud Computing was a good faith effort. I'm not sure about IP2Location, but I tried their "Free Demo" and you are correct, it lists my IP address in Plano, Tx. Why that is, I have no idea. I'm sitting at a desk in Oconomowoc, Wisconsin in the office building of a machining company. I was researching cloud computing for our IT department. I assure you, I do not work for Dell.

I have to admit though, I had laughed pretty hard at the situation, and honestly, I appreciate your zealous protection of Wikipedia's integrity. I understand the internet is a crazy place, and if you don't believe me and you think I'm a malicious employee working for Dell, well, there's not much I can do. Don't worry, I don't blame you.

As for my edit, there were two things I wanted to apologize for.

1.) I was unaware of the general Wikipedia comments with regards to WSJ's opinion section. I'll refrain from referencing it in the future.

2.) I think I worded the article wrong. I state that Dell "staked claim to popularizing the term." Perhaps it would be better to say "Dell has claimed responsibility for popularizing the term" or "Dell has claimed that they were responsible for popularizing the term"

In either case, you should know that I did not mean to imply Dell actually popularized it. Proving that would be impossible, I was merely stating their "claim" is they coined the term, and that they have Trademark rights to ir. I tried to show support for both sides of the argument. Dell's side by providing Google trend results with regards to the search volume of the term measured against a timeline, and the opposing side by providing news sources to the contrary.

As for evidence of the existence an overall controversy, I have to say I disagree with you. Looking across forums and blogs, everything from Slashdot to lowlevel basement bloggers, I have found people arguing back and forth about Dell's attempt to Trademark the term and whether or not it is public domain. I believe that such discourse deserves mention on the wikipedia article, although, perhaps I am going about it the wrong way. Perhaps you could point me in the right direction? Let me know what you think!

PS - I probably won't edit anymore from work. I looked up the history of this IP, and it shows this IP made various velociraptor related vandalism edits (Wtf?). I have no idea. The first edit I ever made was with regards to helicopter anti-torque pedals and now this Cloud Computing article.

I'll message you when I'm on my home machine.

69.222.228.201 (talk) 21:27, 7 August 2008 (UTC)John

Thanks for taking the time to write John. You can (and should) create an account if you want to keep some privacy and build up a reputation (using bare IPs starts you on the wrong foot and you don't even need to give an email if you don't want to - once you have an account people can't see your IP). Apologies for the false accusation too - you must admit that it is very convenient that of the entire planet and 4 billion IP numbers you should have landed on Dell's doorstep!

So basically your writing is good but you need to use reliable sources - the more reliable the better. You can read the policy docs (eg WP:5P) but once you've got the basics down continue to be WP:BOLD in contributing where you can.

The Dell issue, while interesting (you can read more about it here), is essentially a dead duck now so far as I can tell - USPTO will declare it generic and it will be a footnote in history. I don't have a problem with Dell but I don't see that they've given anything to the cloud computing community, or maybe they have and I just haven't found it yet, and in any case it sounds like they will eventually. It is quite clear though that they pulled the term from the public domain and tried to monopolise at least part of it starting back in March last year. They've then let it build up steam where they could have been enforcing it since March as a common law mark - they were already overtly using it!

I would suggest a concise overview belongs in the legal or history section on the cloud computing page and/or the dell and USPTO pages:samj (talk) 23:29, 7 August 2008 (UTC)

Sam - I understand your point. I think its very valid and the more I thought about it, the more sense it made. To be honest I was already thinking similarly after the second revision. I started thinking, "I must really be on the wrong track here."

Thanks for the pointers. I'll be sure to make an account in the coming days. Have a great night! —Preceding unsigned comment added by 24.223.206.174 (talk) 02:57, 8 August 2008 (UTC)

Introduction

The introduction in this article is poor. When I open this page I want to know what the heck cloud computing is, I'm not interested in some etymological deliberation of the origins of the word. —Preceding unsigned comment added by 83.89.0.118 (talk) 23:19, 21 June 2008 (UTC)

Hopefully you are happier with the introduction now. samj (talk) 09:17, 1 August 2008 (UTC)

The improvements are appreciated, but I think the initial paragraph remains too reliant on technical buzzwords for a general-use encyclopedia. E.g, consider the simpler ZDNet definition [1]. Not only is it more approachable, but also gives the specific example of Google Apps. The casual reader should not have to look up several unfamiliar hotlinked terms to get the initial concept. The technical detail can be retained but the initial paragraph should be restructured similar to the 2nd ZDNet definition. That way casual readers can quickly grasp the concept, yet those with deeper more technical interest can read further. Joema (talk) 16:48, 5 August 2008 (UTC)

Ok I've included the apps example, "in the cloud", etc. but this 'dumbing down' is not necessarily accurate... for example cloud computing is not restricted to applications running 'in or from network servers' (eg peer-to-peer, unless you're calling the other clients 'network servers'), and where is the application running anyway when the vast majority of the action is in javascript sent to the browser (eg Google Apps)? Anyway we're getting there now - it's a lot better than it was. samj (talk) 06:20, 8 August 2008 (UTC)

The very first sentence in this article is hard to read (I also believe it is off-topic):

The term Cloud Computing derives from the common depiction in most technology architecture diagrams, of the Internet or IP availability, using an illustration of a cloud.

The article should begin with an accepted definition of 'Cloud Computing'. As it stands, the rest of the article is as vague as the topic itself. —Preceding unsigned comment added by 89.167.185.18 (talk) 12:36, 25 June 2008 (UTC)

yes!The article should really begin with accepted defination as the defination mentioned in the article is imprecise(indistinct) and hard to read. Anoopnair2050 (talk) 15:19, 28 July 2008 (UTC)

(Hopefully) fixed now. samj (talk) 09:18, 1 August 2008 (UTC)

Similar Meanings

There are getting to be so many terms in this area that have similar meanings. I think it would be nice if there could be an article or something within Wikipedia that would straighten all this out, like a comparison chart. Terms I'm thinking of are cloud computing, grid computing, ASP, thin client, RIA, distributed computing, cluster computing, and time-share (from the 60's). For example, what is the difference between cloud computing and ASP? They seem remarkably similar. Also, I get the feeling there is some difference between Cloud Computing as described in this article, and Cloud Computing as described by Google and IBM in their initiative announcement. See http://www.mercurynews.com/businessheadlines/ci_7124839 and http://www.pcworld.com/businesscenter/article/138195/google_ibm_promote_cloud_computing.html and http://www.itjungle.com/bns/bns100807-story02.html Westwind273 18:17, 9 October 2007 (UTC)

Agreed .

There could be a redirect from articles ( cloud computing, grid computing, ASP, thin client, RIA, distributed computing, cluster computing, and time-share ) to ONE article called xzzee (suggestons welcome) Sanjiv swarup (talk) 09:15, 23 February 2008 (UTC)

Disagreed.

Distributed Computing is the general problem of having more than one box talk to each other. Clustering is something you do on a single site; what this 'cloud computing' story says is that you dont have your own cluster; you have a storage somewhere, CPU elsewhere. Now, Grid Computing and Cloud computing are similarish; though by virtue of superior vagueness, Cloud Computing is a superset of Grid Computing. Grid Computing has taken on specific meanings, often specific architectural meanings (OGSI/GGF grid architectures) which don't hold for Cloud Computing. Which is probalby why press/marketing like the term. I have a presentation on the topic [[2]], which looks at some of the differences. SteveLoughran (talk) 11:48, 23 February 2008 (UTC)

Further...

Grid computing refers to a method for establishing your computing resources. Grid is not necessarily defined to cloud. Someone can build their own grid computer and not make the resource available over IP. Cloud refers to accessing computing resources, whether they are grid or not, over the Internet. Don't overcomplicate it. —Preceding unsigned comment added by 67.188.145.142 (talk) 01:22, 4 May 2008 (UTC)

The New York Times has an article talking about the confusion http://www.nytimes.com/idg/IDG_002570DE00740E180025742400363509.html?ref=technology --Westwind273 (talk) 22:55, 8 April 2008 (UTC)

Having talked to a number of people in industry and academia, the consensus was that cloud computing was congruent to a subclass of grid computing (indeed, it's probably the most interesting subclass where the resources are not bound to a single organization). Thus, in many ways it is possible to view cloud computing as a rebranding of grid computing to sell it to a new group of people without some of the built-up baggage of stuff that people hated about the way grids had become. As it is, in my view the whole term-space is still very new and could do with shaking out for another few years before a truly encyclopædic article on the area can be written. Donal Fellows (talk) 15:43, 21 May 2008 (UTC)

That makes sense. Grid had got bogged down in implementation details, with a lot of focus on standardisation of WS-* based APIs for talking to stuff. Cloud Computing steps back from the details by ignoring them and avoiding the standardisation process. However, I'm not sure about 'subclass'. If you use google mail to keep your inbox, you've moved from a LAN-hosted mail server to the cloud; Now certainly in a utility computing world you could host your stuff 'on a grid', but the main focus for grid work since about 2001 onwards has been batch data processing and computation, with a focus on scientific/engineering apps rather than interactivity. It's Grid Computing that has painted itself into a corner. SteveLoughran (talk) 12:12, 22 May 2008 (UTC)

The Wikipedia articles on Cloud Computing and Application Service Provider give remarkably similar definitions. For someone who knew little about these subjects, I think they would be confused by these two articles. Why have two names for what is essentially the same thing? --Westwind273 (talk) 22:19, 1 July 2008 (UTC)

Agreed, there is a lot of confusion, particularly around grid computing which was originally intended as the computer equivalent of the electricity grid. Unfortunately that vision was never realised and grid became another word for cluster. While there is an unfortunate tendency to focus on virtualization, grids, etc. the real promise is more abstract and it matters not whether the calculations are done by a grid, peer-to-peer, or by an army of monkeys with typewriters, so long as it's fast, cheap and secure. It concerns me greatly that (particularly computer hardware) vendors who are tired of selling rebadged clusters as grids are now interested in selling rebadged grids as 'private clouds', and I hope that the last days I have spent editing the cloud computing article(s) help to clear the situation up a bit. TODO: clear up ambiguities between cloud/grid/saas/asp. samj (talk) 09:03, 1 August 2008 (UTC)

I hope that edits made over the weekend clear things up a bit. samj (talk) 08:33, 4 August 2008 (UTC)

Centralized Computing

I think we should discuss centralized computing... cloud computing is simply a buzzword for a flavor of centralized computing. Basically it's a fancy "futuristic" name that these upstart companies use because it is much easier to market. Technically, the differences are pretty insubstantial. The only difference I can think of is that the network is bigger. Software is stored on a central server, and computers access it via a network. --72.39.35.178 (talk) 16:50, 13 August 2008 (UTC)

Strongly Disagree for reasons that should be obvious, not the least of which is 15,600,000 hits for 'cloud computing', 62,000 for 'centralized computing'. USPTO also just declared cloud computing both descriptive and generic which would suggest that the term (which appears in at least one dictionary already) is here to stay. Furthermore, there are both distributed and centralized aspects to cloud computing, particularly when you start talking about peer-to-peer applications (which have no place at all in centralized computing). I'm surprised we're having this discussion at all but nonetheless since you've suggested (twice now) that cloud computing be merged into centralized computing it belongs on the latter's talk page. samj (talk) 08:19, 15 August 2008 (UTC)

Potential advantages

Doesn't this section should include something about scalability and fail recovery potential advantages?Orimosenzon (talk) 22:08, 21 June 2008 (UTC)

"The Cloud" image

Boy. that's helpful. Took a lot of work to create that image. Very artistic. The caption helps a lot. - Realkyhick (Talk to me) 18:21, 4 August 2008 (UTC)

The image is frivolous and therefore unencyclopedic. It's a Wikimedia Commons image so I suppose it cannot be conveniently put up for IFD. I will or someone else ought to remove it from the article unless its utility and encyclopedic nature can be clearly demonstrated. --AB (talk) 19:39, 5 August 2008 (UTC)

Yeah it took all of about 8 seconds to whip up in omnigraffle and replacing it with something more useful (like a sample network diagram) is on my todo list. Then again if you guys have both got time to complain about it here then presumably you've also got time to suggest (or create) a better replacement? samj (talk) 22:24, 5 August 2008 (UTC)

A sample network graph could be nice. Unlike you, I don't really know much about CC, so I don't know what to put in the diagram. I will remove the cloud image from the article for now, and hope that a nice network diagram magically appears. --AB (talk) 23:46, 5 August 2008 (UTC)

Ok. TODO. samj (talk) 17:41, 6 August 2008 (UTC)

Abstracted layering

The cloud mentality is in fact the idea that what's on the backend of the user interaction is a no-care. With the escalation in virtualization, this methodology becomes more prevelant across all layers of computer environments and user interfaces. A quick example would be that if a user required a database driven backend, it should not matter what operating system the database is actually deployed on as long as it is supported on that OS. One could even argue it shouldn't matter what database manager is installed for the same reasons, as long as dependency chains are met.

Further then that, not only the what of the install, but now, the where of the install becomes a "clouded" entity. Centralized computer farms could be established as shared resource pools, and as long as performance criteria are met, the users environment could now be deployed at a selection of physical locations.

Cloud computing is a suitible. name —Preceding unsigned comment added by 74.67.121.215 (talk) 00:20, 1 February 2008 (UTC)

--Esolution (talk) 14:05, 23 January 2008 (UTC)

Agreed. Cloud computing is the mother of all abstractions. samj (talk) 09:15, 1 August 2008 (UTC)

Cloud services vs. 'cloud computing'

I want to suggest that 'cloud services' are the proper term and 'cloud computing' just creates confusion as it's jargon-based. Cloud services are akin to web services, but outside the firewall. One poster asked about the difference between ASP and Cloud computing, mentioning that they seemed similar. They aren't. The ASP model is one of a complete application run in a hosted manner. Cloud services are discrete point services (storage, compute, payment processing, etc.). You put together a number of cloud services to build a complete application. This is EXACTLY the same as Web services. The major differences lie in the fact that 'web service' was mostly something produced and consumed behind the corporate firewall. Cloud services differ by being out in the cloud, providing access to everyone for that web service, and using some kind of innovative charge model (usually the 'utility' model) where anyone can afford it, even for a short period of time. I tried to outline this more here: http://neotactics.com/blog/technology/short-sighted-about-cloud-computing

My main thrust being that 'cloud service' == 'web service' might help clarify the discussion. 'Cloud computing' doesn't really relate to anything.

And yes, 'web service' is a terrible term, but it's somewhat established at this point. :D

Disagree Cloud services are a subset of cloud computing but they need not be web services n the traditional sense (eg REST/SOAP)... XMPP is used to push mail to iPhones for example... and I think we'll see increasing diversity going forward. samj (talk) 09:08, 1 August 2008 (UTC)

Ok cloud service ~= web service, but cloud computing stays. Archiving. samj (talk) 10:59, 21 August 2008 (UTC)

Grid computing

No fair removing opinion pieces about cloud not being ground breaking while leaving opinion pieces about how revolutionary cloud is. Gotta be fair here and present both sides of that debate Sam. --Rw2 (talk) 19:38, 25 August 2008 (UTC)

Where is the debate? I'm yet to find anyone else claiming "a cloud is a resource on the grid" etc. samj (talk) 21:09, 1 September 2008 (UTC)

if this conversation is complete, I may archive it . Sanjiv swarup (talk) 06:58, 10 September 2008 (UTC)

Cloud standards

Looking through the latter parts of this article it seemed like just about every web-technology is being now label as "Cloud-foo". DO we really need to list every web browser, every web language, every web standard in the article and create duplicate categories for all of these? Its all beginning to look a lot like original research unless citations can be found for all these products asserting something specific to cloud computing. --Salix alba (talk) 18:16, 12 September 2008 (UTC)

Agreed we need references for this; most of the standards are by induction from Web 2.0, SaaS, etc. and some new ones are being discussed specifically in this context around storage, virtual machines etc. There will certainly be some standardisation of things like cloud databases (eg SimpleDB et al) in the future too. samj (talk) 10:19, 14 September 2008 (UTC)
Regarding listing 'every browser' etc. I'd be happy enough to trim this back to ones that have targeted cloud computing environments; Chrome, maybe IE 8, probably FF4 etc. samj (talk) 10:20, 14 September 2008 (UTC)

Introduction opaque

The introduction is opaque as hell, and gets worse and worse after the first sentence. Might someone perhaps provide a new lede that concisely says what the heck "cloud computing" actually refers to? --jpgordon^∇∆∇∆ 21:29, 1 August 2008 (UTC)

Mass Archive November 2008

Changes Made

Moved the text from the Introduction into a heading called "Brief". AdityaTandon (talk) 07:57, 8 October 2008 (UTC) Aditya Tandon

Apple

I think Apple, Inc. should be added to the list of companies heading the "cloud movement". Since it has released it's new "MobileMe" service, it now has users doing a lot of cloud computing. —Preceding unsigned comment added by Robo56 (talk • contribs) 14:54, 13 September 2008 (UTC)

I'm unconvinced; the point is to illustrate what cloud computing is and Apple is well known to be a hardware company... your average joe thinks of ipods, iphones and macs. Conversely when one thinks Google you thinks Internet/cloud. That's not to say Apple doesn't get credit elsewhere for their efforts, just that they're not the best example. samj (talk) 08:09, 17 September 2008 (UTC)
- Yeah, same for SAP. 84.72.91.139 (talk) 08:32, 12 November 2008 (UTC)

Cons

The article made no mention of the downside of cloud computing particularly the privacy issues with this technology. 78.86.217.191 (talk) 09:56, 19 September 2008 (UTC)

This article has a "Potential advantages" page. Wouldn't it make sense to add a "Possible Disadvantages" page for the sake of continuity? Wikipedia tries to be as non-objectionable as possible, so it seems like a good idea. --Jnorm (talk) 13:58, 27 June 2008 (UTC)

Its a good point i think disadvantages also should be added. Kalivd (talk) 06:23, 26 July 2008 (UTC)

its a great idea.disadvantages should added so as to make people understand abt the article very well.Anoopnair2050 (talk) 15:11, 28 July 2008 (UTC)

Cons about cloud computing any argument against adding the same to the article page

It’s not secure.
It can’t be logged. Tied closely to fears of security are fears that putting certain data in the cloud makes it hard to log for compliance purposes.
It’s not platform agnostic. Most clouds force participants to rely on a single platform or host only one type of product.
If you need to support multiple platforms, as most enterprises do, then you’re looking at multiple clouds. That can be a nightmare to manage.
Reliability is still an issue.
Portability isn’t seamless. As all-encompassing as it may seem, the so-called “cloud” is in fact made of up several clouds, and getting your data from one to another isn’t as easy as IT managers would like.
This ties to platform issues, which can leave data in a format that few or no other cloud accepts, and also reflects the bandwidth costs associated with moving data from one cloud to another.
It’s not environmentally sustainable.
Cloud computing still has to exist on physical servers. As nebulous as cloud computing seems, the data still resides on servers around the world, and the physical location of those servers is important under many nation’s laws.
The need for speed still reigns at some firms. Putting data in the cloud means accepting the latency inherent in transmitting data across the country and the wait as corporate users ping the cloud and wait for a response.

Kalivd (talk) 07:17, 1 August 2008 (UTC)

'Advantages' and 'Disadvantages' sections tend to read like datasheets and some 'advantages' are also 'disadvantages' (eg security can go both ways). I hope you are all satisfied with the 'key characteristics' section which should be unbiased statements of fact.

Regarding your specific points:

FUD. cloud computing is often more secure (particularly in terms of availability and integrity) than legacy systems.
FUD. there is no reason cloud offerings cant be logged, and most are. getting access to the logs can be difficult though so your comment is partially accurate.
if you have ever tried to deploy linux software in a microsoft shop or vice versa you will in fact appreciate the flexibility that not having to buy infrastructure affords you
again, there's nothing stopping you from using multiple products/platforms/providers (in fact it's easier as you don't need to invest)
Reliability is usually better than legacy systems
Portability is typically addressed via APIs which are settling down - yes this is a valid concern but no moreso than proprietary file formats
See above
FUD. You really think it's better to have every man and his dog running servers? I guess you would also argue that we should shut down the power stations and run diesel gensets too? Cloud is Green.
Cloud computing need not necessarily exist on physical servers, but the jurisdiction problem is real. Providers like Amazon are making 'local' datacenters available for europeans (others will follow), but harmonisation of regulations will become increasingly important going forward and is just a side effect of globalisation.
Bandwidth and latency requirements are highly application dependent; fast pipes are prevalent nowdays and solutions like gears help where they are not available. In any case moving infrastructure outside almost always helps distributed enterprises (as most are these days, with road warriors, work from home, decentralisation, etc.)

I agree that there are outstanding issues that need to be addressed, but I also believe that they will in due course. Conversation (like this) is the first step. samj (talk) 09:38, 1 August 2008 (UTC)

Alright, can you collectively list out the disadvantages of cloud computing so that i can get a clear picture of the cons against the usage of cloud computing. Kalivd (talk) 06:00, 2 August 2008 (UTC)

As I said before: Advantages' and 'Disadvantages' sections tend to read like datasheets and some 'advantages' are also 'disadvantages' (eg security can go both ways).. A great example is Gartner claiming that the very opaqueness that makes cloud computing attractive to many (that is, that you don't need to see or care about what goes on inside) is in fact one of its greatest dangers^[10]. Anyway I'll run through and make sure the points are balanced.

I would appreciate the improvement.. Thank you. Kalivd (talk) 07:10, 2 August 2008 (UTC)

HTML? JavaScript? Ajax?

These have very little to do with cloud computing. I am removing them until someone gives me a reason they are there. — ^Fatal Error 07:52, 1 August 2008 (UTC)

Thanks for the feedback. Ajax enables processing to be done on the client side. Existing cloud based software (eg Salesforce, Facebook, Google Apps) uses Ajax extensively and indeed could not exist without it, so I would argue that it is a critical component of cloud computing. OTOH I agree with you that HTML and Javascript by themselves aren't interesting in the context of could computing, in the same way that TCP/IP is not really relevant. Let's leave Ajax and drop the others. samj (talk) 08:34, 1 August 2008 (UTC)

Cloud Computing page suggested/request edits

Sam

The Cloud Computing page should be updated to include a few items. You appear to be active in this section and I cannot edit it due to a business relationship with one of the companies involved. Particularly, this sentence:

The cloud computing "revolution" is being driven by companies like Google, Red Hat[9], Salesforce and Yahoo! as well as traditional vendors including Hewlett Packard, IBM and Microsoft[10] and adopted by individuals through large enterprises including General Electric, L'Oréal and Valeo[11][12].

This sentence lists Google, Red Hat, Salesforce and Yahoo as leaders/drivers in cloud computing. It should begin with Amazon, http://en.wikipedia.org/wiki/Amazon and 3tera http://en.wikipedia.org/wiki/3tera, who are pioneers in the efforts to offer commercial cloud services.

Among the large enterprises, you should also include British Telecom, http://en.wikipedia.org/wiki/British_telecom, who is running on of the largest 3tera clouds in the world.

Additional, I suggest expanding this sentence into one or several paragraphs: Commercial offerings need to meet the quality of service requirements of customers and typically offer service level agreements[6]. Open standards and open source software are also critical to the growth of cloud computing[7].

This sentence is very accurate and important. Many of the commercial offerings fail to offer service level agreements or open standards, let alone suppor open source software (Google, Amazon and Yahoo, as far as I understand, do not but I am not an expert in this area:

Jonahstein 17:10, 5 August 2008 (UTC)

Cloud vs Grid

Cloud computing is often confused with grid computing. Grid is a larger concept that allows access to many kinds of resources, including clouds. As such, a cloud is a resource on the grid. This confusion comes from the improper use of grid as a synonym for cluster and from the expansion of the term cloud (which originated as a term for compute access) to include storage (thus making it closer to a synonym to than to its utility computing origins). The majority of cloud computing infrastructure currently consists of reliable services delivered through next-generation data centers that are built on compute and storage virtualization technologies. The services are accessible anywhere in the world, with The Cloud appearing as a single point of access for all the computing needs of consumers. Commercial offerings need to meet the quality of service requirements of customers and typically offer service level agreements[4]. Open standards and open source software are also critical to the growth of cloud computing[5].

Reverted unreferenced, uncommented, controversial edits by numbered user. samj (talk) 08:44, 9 August 2008 (UTC)

More edits have been reverted and I had an offline discussion with the editor (Rw2) which included:

Anyway discussion is good but changing large swathes of an article with contentious views based on opinion pieces, blogs and an ancient reports without first obtaining consensus via the talk page is not on. Usually ancient isn't such a bad thing but the report you cited predates cloud computing by half a dozen years and fails to account for subsequent shifts in the grid area.

Furthermore it takes an otherwise quite clear definition of 'a "super and virtual computer" is composed of a cluster of networked, loosely-coupled computers, acting in concert to perform very large tasks' and replaces it with meaningless drivel that could apply to many different types of computing: 'creating a hardware and software infrastructure that provides dependable, consistent, pervasive, and inexpensive access to high-end computational capabilities'. You could toss this on cloud, mainframes, distributed computing, centralized computing and no doubt hundreds of other computing articles and it would be impossible to differentiate between them.

Anyway, grid and cloud are accepted to be different things; if that weren't the case a new moniker would not have been invented. Yes there are similarities if you drill down to virtual machine providers like GoGrid, Amazon and Sun Grid (it's no wonder then that some of these providers have 'grid' in their name), but the similarities end there. Like it or not, in the eyes of the public grid is about high performance computing, batch jobs and coordination between large clusters run by different administrative domains. Most of the articles talking about cloud these days are actually talking about saas providers like google apps for example.

The point is that wikipedia readers should derive clarity rather than confusion from our articles, and it's our job to impart the consensus view on a subject while citing relevant, verifiable, notable references rather than anything we can find which reflects our views.

samj (talk) 06:05, 26 August 2008 (UTC)

Centralized Computing

I think we should discuss centralized computing... cloud computing is simply a buzzword for a flavor of centralized computing. Basically it's a fancy "futuristic" name that these upstart companies use because it is much easier to market. Technically, the differences are pretty insubstantial. The only difference I can think of is that the network is bigger. Software is stored on a central server, and computers access it via a network. --72.39.35.178 (talk) 16:50, 13 August 2008 (UTC)

Strongly Disagree for reasons that should be obvious, not the least of which is 15,600,000 hits for 'cloud computing', 62,000 for 'centralized computing'. USPTO also just declared cloud computing both descriptive and generic which would suggest that the term (which appears in at least one dictionary already) is here to stay. Furthermore, there are both distributed and centralized aspects to cloud computing, particularly when you start talking about peer-to-peer applications (which have no place at all in centralized computing). I'm surprised we're having this discussion at all but nonetheless since you've suggested (twice now) that cloud computing be merged into centralized computing it belongs on the latter's talk page. samj (talk) 08:19, 15 August 2008 (UTC)

Sam Johnston, you are not the only person I want to read my comment. Don't go and "archive" it just because *you* don't like my comment or want to discuss it. --72.39.35.178 (talk) 06:01, 7 September 2008 (UTC)

I disagree too. There is an implicit message in "Cloud Computing" that the data and computation is "someone else's problem". A workgroup email server on a remote site is not on the cloud. Furthermore, cloud application architecture is all about having many, many more servers than "a central server". There is one more datacentres, each with a number (possibly a few thousand) servers, with disks and interconnected by high speed networking. The disks fail, the servers crash, yet the cloud keeps working. All the old application designs that worked well on single server, even small cluster systems no longer apply. SteveLoughran (talk) 21:06, 7 September 2008 (UTC)

I am not talking about email. I am talking about things like Citrix XenApp. How is "cloud computing" any different other than simply having a backup server? --72.39.35.178 (talk) 18:33, 8 September 2008 (UTC)

Citrix looks like it is doing something that makes the clients look like thin terminals on a remote machine. But you know what? That's only one design, and its not how everyone else works. Look at Hadoop and read the MapReduce paper to see what other people are doing in their datacentres. This isn't the stuff old mainframes did; by embracing farms of commodity (and unreliable) computers, those people who are building the datacentres have suddenly taken a leap in what they can do. If you look at my slides [[3]] you can see that the jump from a cluster to a datacentre with a farm of computers and a high speed network fabric changes a lot of the assumptions. XenApp is an attempt to host existing apps on such an infrastructure. That doesn't mean its the right thing to do, just a stop-gap measure while the new architecture evolves and people outside google, Microsoft and yahoo start coding for it. SteveLoughran (talk) 21:26, 8 September 2008 (UTC)

Could you give me something more substantial than agriculture analogies and motivational speeches about evolving assumptions and leaps and whether or not "it's the right thing to do?" Networks are not made of cloth and I do not grow corn in my computer. Keep your paradigms in your pants and just explain how it is all hooked up--I'm way more confused by jargon than a highly-complex but straightforward technical explanation. --72.39.35.178 (talk) 18:36, 9 September 2008 (UTC)

I've revised the page's history and it started off saying it was just another "form" of grid or distributed computing; far more accurate in my opinion than the current version (despite it's many deficiencies). In my opinion, the term "Cloud Computing" is just another fine example of marketing and buzzword spinning than of anything tangible or specific. It's vague and it means many different things to many different people; all the hallmarks of good marketing. I think wikipedia is being used as an effective "marketing" tool for tech companies to peddle their vaporware terminology... along the lines that if Wikipedia says XYZ then there must be something to it. At the very least, this page should be flagged - it does not represent a true and factual account of anything and I question it's neutrality. CarlosLozanoDiez (talk) 17:02, 28 October 2008 (UTC)

The article's history is irrelevant - I invested a significant amount of time and energy into rewriting it from scratch as the previous version did not reflect reality. Cloud computing is not a marketing buzzword and most of the leading vendors (notably, including Microsoft, traditionally a staunch opponent) have changed direction to focus on it. -- samj _in^out 06:21, 3 November 2008 (UTC)

Introduction paragraph

The introduction para is very very long, It just looks like a sea of blue links rather than a definition to me. As i go on reading, the very first para itself confuses me to such an extent that i really care a less about reading the whole article, thought of adding a suitable template for the cleanup purpose but before doing the same would like to discuss about this on the talk page. Any comments appreciated. Kalivd (talk) 14:39, 8 September 2008 (UTC)

someone just added to it earlier today, and yes, it does need taking a sharp knife to it by someone. Go for it! —Preceding unsigned comment added by SteveLoughran (talk • contribs) 16:24, 8 September 2008 (UTC)

I just did a few changes Sanjiv swarup (talk) 02:41, 9 September 2008 (UTC)

I feel Except for the first para rest all be put in a seperate para which speaks more about cloud computing. As on of the Wikipedian has mentioned above about the confusion it creates at the first instant. So putting it in proper paras might just help.

Dhoomady (talk) 07:18, 11 September 2008 (UTC)

It appears that the length issue has been addressed, so I removed the tag. momoricks ^talk 03:13, 10 October 2008 (UTC)

Thanks. -- samj _in^out 06:28, 3 November 2008 (UTC)

Google Chrome

Can this article please be more neutral? Google is not the center of the universe. Citing Chrome, for example, was a bit too much for me. More traditional browsers, such as IE and FF deserve the merit much more than that piece of failure called Chrome. —Preceding unsigned comment added by 24.16.37.118 (talk) 08:20, 20 September 2008 (UTC)

I'm on OS X so I don't use Chrome (yet), but we need more from our browsers in terms of performance, scalability and security (I've lost count of the number of times I've lost a bunch of tabs to Firefox crashing and I haven't had a Windows box to use IE on since... what... 2005). IE8 might be a contendor, and possibly FF3, but the point was that cloud computing demands a new breed of browser. -- samj _in^out 23:16, 20 September 2008 (UTC)

Easy to understand Cloud Computing

Cloud Computing is a type of New World Wide Web Infrastructure you might want to call it, whether or not the infrastructure will pertain to only mobile devices, desktop or even hybrid is not really clear.

Traditional World Wide Web work like the following

Database (Server)
Purpose-based Server (server that perform Services / Application, Browsing (Web), Gaming, Multimedia Broadcasting...etc.)

via

Internet Cloud (2 parts: Infrastruture)

Hardware Infrastructure (DNS, Service Provider (ISP)...etc.) and Algorithm

Abstraction Infrastructure - They are thousand of Cloud in the entire internet, where each cloud represent a field of study (when you try to find stuff, such as on Google, Amazon, Yahoo, or any type of search engine) the Engine will use an Algorithm (or a Logical flow chart that try to find the most appropriate information for your accordingly).

The Semantic Web has developed by MIT define a clear set of method using programming language for categorizing each field of studies which are stored as prototype or netrual data.

Your Computer

Cloud Computing

Database (traditional Database, e.g. Data Center)

Purpose-based Server (virtualize)

Internet Cloud

Service Provider (there is a catch, see Below)

Your Computer (application are virtualize)

Note that Cloud Computing only accelerate the speed of processing, because application, services...etc. So the definition of Cloud Computing means that you can be anywhere in the world however, the Hardware Infrastructure of the internet cloud doesn't require any DNS, NetBIOS, LMHOST file so basically you are kind of connecting P2P. However, 90% of the people on the market right now is getting one information extremely incorrectly. That is Virtualization is a very object-orientated based meaning that the objects are more like container and the container can only obtain a certain type (or range) of data and they usually require API to communicate each other while API are written by IDL (Interface Definition Language) and several other Schema (definition language or languages that teaches computer how the language works), they are not flexible to semantic constraints and are doesn't have the capability to many variations.

In order for application to perform really well they need to be written in a topological structure where each stages and levels of coherency and relavance references are written in a clear way that it doesn't affect the traffic engineering of chipset, doesn't effect the physical and software level of parallelism, caches are allocated correctly through methods of file system, partition, or other structure methods of implementation.

So how does this translate ultimately? Well this can affect simple data-orientated presentation, such as presenting a simple graph, a SVG picture, high quality graphics, sharing, middleware communications. Faster method of protocols will also be hard to implant which can dramatically increase the ability to implant security algorithms, data are not easy translated. As time progress with each type of new concept of programming language such as event-driven programming (like application that are firmly concern with things like scheduling), these will decrease the performance, since data values are physical, where object can't really define values clearly such as Zero in one type of math translating to the defintion of Zero in other type of Math. In each program they can, but transfering through a fast standard of The Semantic categorizing fast growing infrastructure, it won't stand a chance.

--Ramu50 (talk) 04:04, 1 October 2008 (UTC)

I'm more confused now. -- samj _in^out 17:24, 20 October 2008 (UTC)

Add a "Criticism of cloud computing" section?

In The Guardian recently it's been reported that two significant people in the computer industry - Richard Stallman and Larry Ellison - have criticised the concept of cloud computing. Stallman called it "marketing hype", and Ellison called it "fashion-driven" and "complete gibberish" - see here. Should such criticism be included, and if so, how? I would go ahead and just add a "Criticism of..." section myself, but I see the article's protected so maybe the whole concept's a sensitive subject. What do others think?--82.148.54.195 (talk) 11:47, 4 October 2008 (UTC)

I'm in favour of criticism; I think the protection is related to WP:EL abuse rather than anything else. But, I'm not sure that rms's faults are valid. Also, if Oracle are so negative about it, why the recent launch of Oracle-on-EC2, combining Oracle 11g (g for grid) on top of Oracle's RedHat linux derivative, Unbreakable Linux.

Some criticism points are: too vaguely defined to be meaningful. Not a solution to many problems. Creates single points of failure in the infrastructure. For the latter, the S3 outage is a good example; a lot of sites broke when S3 stopped working. However, we really need citable criticisms, instead of inserting our own analysis. SteveLoughran (talk) 14:08, 4 October 2008 (UTC)

I totally disagree what Richard Stallman says, that is a direct-unintelligent comment that is similar to what Pat Gelsinger said about the future of Graphics processing. Yet few month later Microsoft and many other reviews have already debated of that topics. ref

--Off topic: From your link, "Gelsinger also added that, unfortunately, the software industry still doesn't take advantage of hundreds of threads and cores." This was a big mistake, or bad idea, by the CPU industry. If I send instructions to the CPU, it should send them through to its cores by itself. It is one CPU, it should act as one CPU, but only faster because it is being processed in parallel. With multiple GPU's do I need the game to be programmed for it? No, the drivers for the hardware take care of that. Think about cars with gas and electric engines, do I have 2 pedals in my car for the two engines? No, the car (it's computer) takes care of it for me. So, why do we accept these sub-standard features from the CPU industry now-a-days?--deewhite

Moore's Law have been debated as some RAM companies hypothesize transistors might end by 2020 ref
Intel Atom SCH, shows Intel lack of understanding of how multiplexing device work at all.

Also base on the past reputation of IDF and Intel inability to support bracket USB as a basic knowledge of comoputing, why should anyone believe in that buffoon.

Reading from that topic, I can already see that Richard don't even understand how Cloud Computing can change AJAX to a whole new level and implant parallel computing processor design which can work in sync with Cloud Computing and can dramatically increase the relational database processing. Likewise I think he is like the rest of the amateur who is still getting confused with UML and XML + OO objects and the previous things I just said. The proof is obvious, the article didn't comment anything that shows his understanding of the topic at all. It sounds more like he has an anger / frustration management issue than knowing what he is talking about.

Also the architecture of how a CPU-GPU has already been known to one of the company that make server and workstation CPU which I do not wish to reveal. With that being said, I think a lot of Intel and Microsoft are statements are totally immature and I do not trust their fact at all.

Actually there already been good YouTube videos that explain cloud computing quite well, the only thing is that too many companies right now is sometimes confused and almost reveal their secrets and the idea of cloud computing, so you can't really tell who is going in the right direction.

The fact is too many people right now in the industry is judging things by what is obvious and therefore is conditioned. It is similar what many reviews debate on Solid State products and Hard Drive, yet the only knowledge they use is power and speed. SSD (NAND, SLC or MLC) vs Hard Drive, CPU-GPU +/- programmable codec has, Cloud Computing are all products that haven't fully matured, so even if we are on the right track that doesn't mean you are correct. As other science has proven the just, because you are more "advance" doesn't mean you are the best. As recently the debate of evolution hypothesize that all animals may be all capable of doing the same thing, such as the ability to fly, it is just a matter of how efficiently they can fly. However, just because eagle can fly the best by using soaring, doesn't mean they are a good flyer, if you place them at an enviroment where there is nearly no wind, the feather basically is more heavy. You may say I am wrong, that not everything can perform "anything" they desire, but I am going to telly you that is where you are wrong. Data utilization can achieve that, unforunately very few people that I talk to before know how to utilize it at all due to insufficient knowledge.

More examples are For the past 30 some odd years, speed has already proven us that it is not a future, the ability to efficiently utilize data mangement knowledgeability towards a design is whats important as seen in Celeron, Pentium and Core processors all having speed such as 2.8GHz can have a dramatical difference in the workload. The same as Overclocking the highest speed doesn't gurantee stability in performance.

The reason why I don't want to reveal it, because Intel would go ahead and copy it straightaway, this is not the first time Intel have done it and I totally don't trusted the idoit and unethical practice of Intel as they always do one thing and say the other. As far I am concern, before they say their C2Q is better than Phenom and now they copy it, wtf. For Christ sake, over 15 ads I seen on TV about computers less than 3 ads advertise C2Q while they only advertise Core 2 Duo, yeah so much for the claims that society like to make. Load of crap. --Ramu50 (talk) 22:39, 6 October 2008 (UTC)

Breaking news: threatened vendors critical of next generation tech. Film at 11. A criticism section is a slippery slope so valid, specific weaknesses have been rolled into the 'key characteristics' section (one man's weakness is another man's feature). -- samj _in^out 10:11, 20 October 2008 (UTC)

-- I definitely agree that there needs to be a criticism section: cloud computing is a marketing term and not a real computing term. As stated in the introductory section: "It is a paradigm in which information is permanently stored in servers on the Internet and cached temporarily on clients that include desktops, entertainment centers, table computers, notebooks, wall computers, handhelds, etc." THIS IS HOW THE INTERNET HAS ALWAYS WORKED! "Cloud" is the internet...well ya know what? Every server, desktop, TV set, handheld, etc. that has an IP addy IS the internet...it's always been that way, nothing has changed. When I connect to a site, I d/l a page and it is temporarily stored on my computer to display it. This is how it's always been.

"Software as a service" is a much better and more precise name for (some of) the same thing(s): it means that instead of buying software (like MS Word), I can go to a website and perform my word processing via a web portal. Even so, tell me, what has changed? Nothing. Is this revolutionary? No. Is Facebook a cloud computing service? No, it's a website, run off a server, that has an IP address, and is connected to the internet. Until Facebook is run off everyone's computer and moved around as people go on and offline, it will never be true "cloud computing". Think of P2P, that is almost cloud computing, but it is not a purely web-based "service" as it requires software installed on your computer to access the data, but even so, everyone needs a web browser to use Facebook/Gmail. I agree with the dissenters, cloud computing is nothing more than marketing hype for new services that people did not want to call Web2.0; so, they made up a new name. Wake up people! Now, if I am completely wrong here, then someone needs to clarify this entire page, because to me it all sounds like old crap being repackaged with a new crappy name. If I have a website on one server and it uses a database on another server, that doesn't make it cloud computing, it makes it the internet people!

One more note on SaaS: anyone remember "free" ad-based ISP's? Or, "free" ad-based computer programs? Yeah, they died because they sucked. So, why are we now accepting them just because they are on the internet and not installed on our computers? Gmail isn't free, you "pay" for it by using Google more and seeing more of their ads. Nothing changed except the location of the software and we are being bought right into it all over again. It ended badly the first time around, why is this time going to be different?

Furthermore, in the distributed work-load sense of the phrase, isn't this just a fancy word for virtual servers? Apache has had the ability to host numerous sites on one computer for awhile now... From the "Characteristics" section: "By sharing perishable and intangible computing power between multiple tenants, utilization rates can be improved (as servers are not left idle) which can reduce costs significantly while increasing the speed of application development." Lets ignore the falsity that "the speed of application development" has anything to do with whether my webserver is being used to serve web surfers (how does my server serving content make me program slower?), this quote sounds exactly like Apache's virtual servers...am I wrong?

Does anyone remember internet computers (where you boot up from an internet-based OS)? If so, I bet only vaguely. Why? Because it was crap! What happens if your internet goes out? You can't boot your computer. IDIOTIC! Cheaper? Maybe, but still stupid overall.

From the previously linked guardian article: "pushing forward their plans to deliver information and software over the net." Hasn't information always been delivered over the net? Isn't that what is was invented for? Again, what has changed enough to deem an entire new word/phrase? "The interesting thing about cloud computing is that we've redefined cloud computing to include everything that we already do," this is exactly correct!

Finally, I want to talk about security. Is it more secure to store my e-mails on Gmail, or to download and store them on my computer? That way when I am off-line, my data is off-line and inaccessible. How many Gmail accounts have been hacked while the user is off-line? How many have been tried to be hacked? If it's on my computer and I am off-line, IT IS IMPOSSIBLE FOR IT TO GET HACKED. Less time on-line equates to safer data. That's why the golden rule for keeping sensitive data safe is, and has always been: Don't connect the computer to the internet. Why are we now trying to throw everything online and giving it to a corporation only interested in making money? Take AOL blogs, for example, they are shutting down their service on October 31st and providing no way for their customers to backup their blogs other than manually copy/pasting the text and "Save As"-ing the images. This is because they stored the data in a proprietary system and really don't care about our data. They only care that the service isn't bringing in money and are now leaving the users out to dry, without providing a way to convert their proprietary blog system into something open like WordPress. Why would they help their users, their users didn't bring in the ad money, screw their data. Is this really the path the internet should be taking?

This entire article needs removed and replaced with "See Web2.0" or, more accurately, "See Internet" links.

Disclaimer: I'm not saying that "cloud computing" is not cool, has no benefits (for some, mainly the large corporations), or should be dissolved all-together. I'm just saying that the term is too broad & idiotic and that Stallman is right in saying that it reduces our (the peoples, the customers) ability to control OUR data. --deewhite

This has a strong focus on offline issues and makes various assertions that I don't think are valid, like that internet computers are IDIOTIC because Internet is unreliable - maybe in the US it is but at least here in France it's rock solid and on mobile devices where most of the action is these days it's virtually bulletproof (eg android, iphone). You also assert that keeping your mail on your computer makes it IMPOSSIBLE FOR IT TO GET HACKED, which is simply delusional.

I agree that the article needs to be balanced, but I have resisted Pro and Con sections because they create a free for all and are often incomprehensible (one person's con, like most of the things you cite above, is another's pro). -- samj _in^out 06:27, 3 November 2008 (UTC)

Cloud-based version of wikipedia

Is there something like that yet? Nodes would communicate with each other a lot like file sharing applications do now, with articles being the items that were cached locally or searched for. Updates would be an interesting problem, but there is still a decentralization advantage here that could mean decentralizing and open sourcing the very last piece of the pie: the server hardware itself. —Preceding unsigned comment added by Zaphraud (talk • contribs) 00:43, 5 October 2008 (UTC)

I think UML can utilize RPC to achieve that and in some virtualization. Theoretically it is very possible to achieve it, you just need to manage the protocol traffic engineering very well or else leakage would cause havoc that might lead to Data Duplication, faults...etc. --Ramu50 (talk) 22:53, 6 October 2008 (UTC)

Sounds something like Freenet. Interesting idea. Not relevant here. -- samj _in^out 10:12, 20 October 2008 (UTC)

Private and Public

I’d like to suggest we add a definition of private versus public clouds to this page. The industry is embracing these terms and even starting to talk about “hybrid clouds” that are a combination of both public and private.

Private Clouds can be defined as: Private clouds use the public cloud architectures and methodologies but are deployed by a single organization inside the firewall. Resources are typically not shared with outside parties and full control is retained by the creating organization.

An example of a private storage cloud is: Private Cloud storage is typically a loosely coupled architecture, where the nodes don’t need to talk to each other to facilitate writing in parallel to a single file spread across multiple nodes. Instead meta-data operations are centralized enabling the data nodes to focus on delivering data to applications or users. Examples include ParaScale, Hadoop and mogilefs. —Preceding unsigned comment added by Mwmaxey (talk • contribs) 20:08, 13 October 2008 (UTC)

Note: User:Mwmaxey appears to be Mike Maxey, Parascale's Director of Product Management. The Parascale article has been deleted as G11 (blatant advertising) and user warned about conflicts of interest. -- samj _in^out 17:22, 20 October 2008 (UTC)

There is ample confusion already (not to mention significant dissent) where the purpose of the article is to give clarity. While it is quite possible that 'private cloud' purveyors like ParaScale will be succeed in co-opting the cloud computing moniker to mean something different, the vast majority of cloud computing discussion is about Internet based solutions; we'll review this when and if that changes. -- samj _in^out 17:22, 20 October 2008 (UTC)

I agree; additionally, private/public is quite self-explanatory and there are many other resources/articles to explain what a private network is. There is no need to spoon-feed the readers with this.--deewhite —Preceding unsigned comment added by 76.18.230.145 (talk) 07:26, 23 October 2008 (UTC)

definition

Is ad-hoc network cloud computing? Is brainstorming a model of cloud computing? --Ramu50 (talk) 02:54, 12 September 2008 (UTC)

Any definition that starts with 'means' and goes on as this article begins is a really poor definition. The definition includes the word 'cloud' several times - in itself a poor move - which itself is never further defined. —Preceding unsigned comment added by 81.50.46.9 (talk) 11:13, 15 September 2008 (UTC)

Thanks for the feedback. Fixed (hopefully). -- samj _in^out 23:16, 20 September 2008 (UTC)

＠＠ what do you mean by the word 'mean'. Though I think this video does a quite clearn and structure. Somehow for the past few days, I think cloud computing is just simplifying the structure of database into object-based orientation (I am not talking about ORM or OODBMS) which is a faster way of accessing data. Traditional RDBMS and ORMs seems like it is more suitable for HPC, Supercomputer and other high performance application, thus not suitable for desktop. The object-based I am talking about is the UML model. I think mobile device is doing so well, because the engine that runs the OS is built by a framework, which is drastically different from desktop kernel which is so complex, that is why things like Driver will take so long to write. However, when the engine is built by a framework, everything is so structured by concepts like CRUD, ACID...etc that implanation are offloaded thus you retrieve thing so easily. Model that use it are ActiveX, ADO, OLE, Silverlight, Flash, C#...for more info see the stuff I write before [Template talk:Databases Click Here]. --Ramu50 (talk) 03:31, 28 September 2008 (UTC)

Can anybody name me a normal desktop computer software which uses internet and could not be considered "cloud computing". And of course explanation why? --82.203.173.213 (talk) 09:44, 12 October 2008 (UTC)

Yeah, sure... FTP clients for example usually just connect to a single server for a point-to-point transfer... use it to access a cloud based CDN on the other hand and you've got 'cloud computing'. 84.72.91.139 (talk) 08:34, 12 November 2008 (UTC)

What is Cloud Computing (simplified version)

DO NOT go back to the Archive and read

Easy to understand Cloud Computing

That is a horrible example I made before and it is way too technical.

New Explanations (Digrams)

Cloud Computing for me I think is probably a new sets of technology that work in an infrastructure, known as The Cloud. First you should know how traditional internet works. For most programmers I think the first chart is very obvious.

Diagram 1

Database

(e.g. SAN [storage area network], VTL [virtual tape library], holographic storage])

Purpose-based server

(refer to application server, game server, web server…etc.)

ISP

mainly control the network traffic

ATM & SONET was placed there just to show that other types of network infrastructure also exist.

MAN (Metropolitan Area Network) WAN (Wide Area Network) Home Business (SOHO, Enterprise)

Diagram 2

AI (Artificial Intelligence) SAN (Storage Area Network)

So to implant cloud computing there are two method. Implant technologies in purpose-based server. Implant virtualization in your purpose-based server. Since webpage scripting should be build using JSON, XML (interchange formats) and in the future Aspect-Orientation (traffic control) Programming. Each virtualized application act as a (communication portals) and thus it can connect to multiple devices mediums (e.g. mobile devices, desktop devices, game consoles).

Another way to connect to other mediums it to build web applications that can support synchronization and integration technologies such as Office Live.

(Read the Virtualization Classification)

Implant technologies in your Databases -for the past decade bioinformatics and IT have been trying to work together to develop a technologies that can make database information act as DNA so it can behave artificial intelligently -Well for a start you can implant Storage Virtualization and The Semantic Web (efficiently classifying and managing the data) and than implant an AI. Since middleware itself is an API, you can implant it to communicate with other devices mediums. Except our technologies is so little right now that probably the only service middleware can provide is XMPP phone calls. Or while you are skiing, you take a couple of photos and use the middleware to gather a bunch of friends from other place and share the picture (temporarily).

Since purpose-based and database in the future might be standalone, and do not require each other, they itself will act as a service provider and servicing desktop web services thus making the entire web acting like a P2P and P2P just looks like a cloud.

Virtualization is usually classified into the following

Hardware based

Hardware-assisted (Virtual Machine, codes are complied or interpreted e.g. JVM)

benefits: Cross-platform

Emulator

OS level virtualization

the most common form of virtualization that uses a Hypervisor to virtual a Virtual Machine.

Network Virtualization

programming the entire network using network devices (such as Routers) to create a VLANs

Storage Virtualization

basically they use a different set of (Network) File System interface technologies, e.g. LDOMs (LUN, SMB, CIFS)

They are other types of Virutalization that many coorporation claim to be a type of virtualization, but some of them I am still trying to understand them, while the other I totally disagree with their concept.

Mobile devices usually use a bytecode or runtime compiler to run the mobile framework and hosted Virtual Machine or Browsers to communicate with applications. Since webpage can be entire build with XML (static interaction), JSON (dynamics interaction) and AOP the entire browsing will totally speed up, because (runtime engine---via---MVC).

The mobile framework is acting as a framework.

Commentary

I REALLY suggest you learn and implant JSON and XML. The currently world believing in JavaScript must be needed or must not be needed is entirely stupid, aspects, objects and data are the fundamentals elements of programming computer science, you can’t omit them, you have to organize them at “EVERY” single level, MVC level, coherency, traffic flow, behavior….etc. --Ramu50 (talk) 00:54, 26 November 2008 (UTC)

Summary

Server (XML, JSON, AOP) interchange, synchronization, integration Virtualization layer

Purpose: to act as a service provider

In Diagram 2, I draw the ISP as the background, because I think in the future ISP will be integrated into the purpose-base server, but that will only happen if the industry decide to implant the TCP/IP OSI model + GSM specifications. Because I think the industry is dumb not to develop a management systems for managing traffic control.

Of course, they are other implementations that most industry haven't notice at all, but I am not going to reveal it, because they are just going to steal my information.

Reference: Database (template)
Other brainstorming topics: Visual Thesaururs --Ramu50 (talk) 23:16, 26 November 2008 (UTC)

Wow, you're obviously putting your heart and soul into this... thanks for your efforts and apologies for archiving something that may have still been current. To be honest I'm still a little confused by these diagrams (mostly by the relevance of some of the components) but it's clear that you are viewing the cloud from a different perspective - from 'underneath' perhaps while I am looking at it from a user's point of view. Anyway, interested to see what comes next. -- samj _in^out 11:33, 26 November 2008 (UTC)

Suggestion

Thanks for your efforts. I was thinking you should be able to represent cloud computing with the actual physical things involved: computers. The Internet is fundamentally three different kinds of computers: servers, clients, and let's call the in-between computers routers. So could you draw a diagram using these three basic elements, and then show how the information flows through the system. I think this will help simplify it more than protocols and acronyms and artificial intelligence, because that's about as illuminating to most people as drawing a cloud on a whiteboard.--72.39.35.178 (talk) 03:29, 3 December 2008 (UTC)

Ramu50 has been blocked indefinitely due to a long history of tendentious editing, editing against consensus, and NPA violations. I haven't looked at what he's written (or drawn) here in much depth, but based on the quality of many of his other contributions, the number of times he's been reverted, the zeal with which he attacks anyone who disagrees with him, and the repeated calls for him to "please stop", I wouldn't count on his material here to provide any new insights into how to best describe cloud computing. His way of looking at things seems to be unique and is almost never in agreement with either reliable sources or consensus. I would advise simply moving the section he's written here to an archive and forgetting about it. Jeh (talk) 05:22, 3 December 2008 (UTC)

Already did that once, and he's back. Maybe you're right. 93.3.248.168 (talk) 00:08, 5 December 2008 (UTC)

If you don't know anything about Cloud Computing mine as well stop bragging. --70.79.65.227 (talk) 08:26, 5 December 2008 (UTC)

External Links Diagram

Regarding about external links diagrams. Currently they are a lot of company out there that have a diagram of the infrastructure. Many of them claim to be cloud, but if you wish to place a diagram there, try to choose a diagram that includes most of the technologies. Not a diagram that only focus on a topics, the information is too forefont, but not everybody may agree with your opinion, so its best to place them to each article accordingly.

Because the number of technologies that Cloud can use is nearly infinite. Some believe in the following

("as a "service")
{ Web Application (+/-) Web Services (+/-) WebOS } Buisness Processing
middleware tunneling

platform computing
edge computing
elastic computing
green computing
virtualization

and many more...etc.

Note: Many of the types of computing does exist in Wikipedia, but are only recognized or introduced by some companies we shouldn't eliminate them. The Cloud Computing has even reach a pre-mature stage, so its not for us to judge who is correct since the industry haven't agree on a common standards or a common concept. Therefore try to be openminded.

The (+/-) means "and / or"

e.g WS (+/-) Web Application.

Actual meaning: Web Service and/or Web Application

I choose Enomaly ECP diagram, because it include the following: VPN, middleware, API, WS and little of Platform Computing prototype ideas. Hope you can find a better one.

Suggestions

Another great way to organize Cloud Computing external links, I would suggest is whenever an example or company is placed. Record down every one of them, and organize them by type of services / companies.

e.g. Are they the following

DotComs
Server Products (software)
Services
Solutions
Platforms
Mobile Application

I think this will give everybody a better idea of what cloud computing actually is, instead of arguing what technologies is cloud computing. Thus it will clear up of what is "as a service", platform computing...etc (<--the aforementioned topics)

Though I guess in the future the Cloud Computing article will mostly likely be explaining the types of implementations, history, practices...etc. and probably things like traditional approaches, methodologies. --70.79.65.227 (talk) 04:33, 6 December 2008 (UTC)

Links to examples or companies usually are not appropriate. See WP:EL, WP:SPAM, and WP:NOTLINK --Ronz (talk) 05:15, 6 December 2008 (UTC)

The aforementioned subjects I mentioned clearly support that this isn't spam, and this clearly isn't an example, it is a reference diagram intended for users understanding how Cloud Computing works. Since the infrastructure section doesn't have any structure paragraph develop on this subjects, yet. I think this diagram would serve a well guide as it encompass more than just one topics. Users can then find out each of the technologies if they wish to learn on the technicality of the subjects. Though this may not be the best one, but we should try to find more, and choose the best one accordingly.

If the diagram only deals with one subjects that is spam (advertising), but in this case it isn't. --70.79.65.227 (talk) 05:43, 6 December 2008 (UTC)

While it is indeed advertising, it's clearly promotional. Further, we're here to write encyclopedia articles, rather than to link to other sources of information. If you want to dispute this further, WP:THIRD would be a good place to do so, or anything else recommended in WP:DR. --Ronz (talk) 15:43, 6 December 2008 (UTC)

I am placing it back on, as you are unable to provide any evidence that it is advertising. The links does meet any of the with WP:EL, WP:SPAM, and WP:NOTLINK policy violation. Also before I asked should be just link the image, you ignore my questions and insists of removing it and end of story without any evidence, that is violating WP:CONS.

Edit summary:

http://en.wikipedia.org/w/index.php?title=Cloud_computing&diff=256171873&oldid=256160287

((rv) This isn't advertisement, if the company allow Wikipedia to use the 
diagram copyright I wouldn't have link it.)

--70.79.65.227 (talk) 20:27, 6 December 2008 (UTC)

Sorry, but the burden of evidence is with you. See WP:PROVEIT. --Ronz (talk) 02:42, 8 December 2008 (UTC)

I already prove the diagram doesn't violate the policy. --70.79.65.227 (talk) 10:35, 9 December 2008 (UTC)

Looks like I'm not the only one that disagrees. Thanks for not reverting it back yet again. --Ronz (talk) 16:25, 9 December 2008 (UTC)

Thanks Ronz. Turns out this was Ramu50 again (as a sock), even though Enomaly have a very long history of spamming Wikipedia. -- samj _in^out 18:14, 14 December 2008 (UTC)

Advertising Tone

This whole article needs to be rewritten to deal with its advertising tone. It is full of hype and it is blatantly promoting its subject as a product. --Nogburt (talk) 04:53, 3 November 2008 (UTC)

Unsurprisingly, I disagree. This 'hype' has been thoroughly validated by changes in focus by virtually all significant vendors (even those like Oracle and Microsoft who have traditionally bashed their cloud based counterparts) and the article is balanced, referring to the pros and cons in the key characteristics section. Wording like 'reliable services delivered through next-generation data centers' sounds positive, but it's accurate... the services are reliable and data-centers next generation. I'm removing the blanket tag, but encourage you to identify passages that are inappropriately worded, so as we can reference, remove or rewrite them. Thanks. -- samj _in^out 06:38, 3 November 2008 (UTC)

Here are some areas:

-The Political Issues section starts off with "The Cloud spans many borders and "may be the ultimate form of globalisation"". This sentence just isn't helpful. -The list of Companies in the Brief section isn't neutral. It is irrelevant as to whether or not these companies do or don't do whatever here. This listing is, though true, advertising (name dropping). I'm considering cutting it entirely. -The Key characteristics section is listed as if it was from a marketing presentation. Again, true or not, it needs to be presented in a neutral manner. The "Sustainability" point specifically is improper. A lot of things are "more sustainable" that doesn't mean that it is relevant or proper to them all to be noted as "sustainable". --Nogburt (talk) 21:49, 14 November 2008 (UTC)

Ok so there are political issues because the cloud spans many borders... reword it if you must but multiple jurisdictions are one of the single biggest hurdles for cloud computing. The latter part is a quote, and I would say a relevant one.

Listing companies is useful iff the companies listed are already associated with 'cloud computing'. Google and Salesforce are two obvious examples. Facebook might be another. Listing SAP, Microsoft, etc. regardless of whether they are getting involved now, is more likely 'advertising' for those companies. Granted there are categories which contain this information but listing a few (2-4) of the best examples of cloud computing actors is IMO a good idea.

The key characteristics section, as I have explained before, was intended to list exactly that... characteristics. One man's pro is another man's con, and many issues (eg security) would appear in both lists anyway.

Sustainability is a big issue today, and cloud computing, by vastly improving resource utilitisation, is (or at least can be) a sustainable solution. My partner is a sustainability engineer so this is stuff I'm exposed to every day; I assure you that very few computing solutions come close to the benefit that cloud computing is able to offer. Conversely, these installations do centralise consumption of copious quantities of power so there are associated concerns about sourcing cheap power (which essentially translates to dirty coal). -- samj _in^out 22:02, 18 November 2008 (UTC)

Sorry about being too brief above. My issue with the quote in the political issues section is simply that the section itself is begins with the full quote as opposed to original prose. I guess you fixed that?

If you think that sustainability is a big issue with cloud computing, I'll go with that as you seem nearer to the subject than me.

Thinking about it, I guess a lot of what is getting me about the article is its heavy use of very hierarchical lists as opposed to prose. I'm not a cloud computing professional by any means. I came upon the article trying to learn about cloud computing but I didn't seem to get whatever it is that a reader might want from it. I spent more time thinking about the quality and presentation of the content I was reading (primarily as a result of the many lists) than the content itself.

But regardless of my lack of technical knowledge on the subject, as a reader there is something that can be improved. Such an article isn't written for cloud computing professionals anyways (they can read it of course, but it ought to help lay readers understand its subject). Often times those closest to a subject are less inclined to make a lay-friendly article than those less near to it. I'm sure some investment banker might have jotted in some nice things (and lists of relevant companies) for mortgage backed securities back in 2005. Perhaps many of these things would turn out to be the opposite of useful. Those nearer to a subject can be more prone to the latest industry trends which may not be optimal content here.

I could put it this way: a good Wikipedia article should have content that is given in such a way that a fair subject-lay reader who spent some time just on that article, and perhaps a few of its linked sources, could take what he learned and replicate the content and manipulate it into some other format.

If I applied this rule of sorts to this article, I could rewrite a lot of it in prose; but it would be a poor work. But it would be poor on account of the content that I set out to replicate in prose more than my ability to write prose (or lack thereof). There's actually a lot of content here, but I just can't translate it into prose without some additional knowledge of the subject.

I'm something of a humanities man (although I know quite a bit about economics). Here's the key question: Would someone from some non-tech savvy background get what a reader of a Wikipedia article wants to get out of this article?

--Nogburt (talk) 09:38, 22 November 2008 (UTC)

The article's already long - turning the lists into prose as you suggest would likely make it less readable. 93.3.248.168 (talk) 02:28, 25 November 2008 (UTC)

Definitely several of the lists should stay. But in many other places the lists should either be made into prose or perhaps even broken off into separate articles. Actually that would also be a thing that we could discuss. Is this article too big for just one article? ----Nogburt (talk) 16:54, 25 November 2008 (UTC)

At 34k, not quite yet but it's something we'll likely have to look at before too long. -- samj _in^out 01:06, 22 January 2009 (UTC)

Cloud image

I just came across this article due to my Twitter search stream for "Wikipedia" picking up a tweet making fun of it [4]. I'm usually a big fan of any kind of graphics in articles, but I have to admit that File:TheCloud.svg doesn't seem to really add anything to the article, and may in fact may detract a bit. In case I'm missing something here, I wanted some input from the article's regular editors rather than removing it unilaterally. Cheers! —Elipongo (Talk contribs) 09:49, 18 January 2009 (UTC)

I replaced the previous image with this one, as I thought it was a bit too technical for the intro section and wasn't even a particularly good diagram to introduce the topic. I'll admit the current image isn't really much more than decorative though. Letdorf (talk) 19:22, 18 January 2009 (UTC).

This was a placeholder for the replacement that never came, and a bit of a joke in itself. It's already been discussed and removed before, but I'm still not sure of a decent replacement - I tried a few with users hanging off it but none of them looked great. Anyway it's not urgent. -- samj _in^out 20:22, 20 January 2009 (UTC)

Per se, humour does not make a text, an illustration or any other form of knowledge inapproriate, uneducational, or unencyclopedic; see the Name of the Rose. The problem with the image then was that it lacked a caption, thus not providing a context and leaving the uninformed reader unenlightened as why it might be humouristic. I've know written such a caption[5], and I'm not certain that it cannot be improved. ¨¨ victor falk 11:21, 18 February 2009 (UTC)

No, the problem is that it is so dreadfully amateurish and whimsical that it negatively impacts the article. If we're going to use humour at all, I'd consider waiting for something which doesn't require a multi-line caption in explanation. This should be removed again. Chris Cunningham (not at work) - talk 12:15, 18 February 2009 (UTC)

As there's been no further comment regarding this, I'm removing the image again. Images are not mandatory, and this one is whimsical at best and idiotic at worst. We don't need a placeholder. Chris Cunningham (not at work) - talk 11:01, 27 February 2009 (UTC)

Agreed, good call. -- samj _in^out 14:42, 28 February 2009 (UTC)

Resolved by creation of File:Cloud computing.svg -- samj _in^out 09:14, 3 March 2009 (UTC)

[1] Gartner Says Cloud Computing Will Be As Influential As E-business

[2] ttp://www.dell.com/content/topics/global.aspx/corp/pressoffice/en/2007/2007_03_27_rr_000?c=us&l=en&s=corp

[3] ttp://www.marketwatch.com/news/story/dell-has-every-right-cloud/story.aspx?guid=%7B6CBB1978%2DE5CB%2D4280%2DA1C5%2D5D1DFBD463D5%7D

[4] ttp://tarr.uspto.gov/servlet/tarr?regser=serial&entry=77139082

[5] ttp://www.dell.com/content/topics/global.aspx/corp/pressoffice/en/2007/2007_03_27_rr_000?c=us&l=en&s=corp

[6] ttp://www.google.com/trends?q=cloud+computing

[7] ttp://news.google.com/archivesearch?q=%22cloud+computing%22&as_user_ldate=2001/01&as_user_hdate=2008/12&scoring=t&sa=N&sugg=d&as_ldate=2004/01&as_hdate=2004/12&lnav=hist3

[8] ttp://tmportal.uspto.gov/external/PA_1_0_LT/OpenServletWindow?serialNumber=77139082&scanDate=2008080708479&DocDesc=Prosecution+History+for+Canceling+NOA&docType=HCN&currentPage=1&rowNum=2&rowCount=12&formattedDate=07-Aug-2008

[9] ttp://tarr.uspto.gov/servlet/tarr?regser=serial&entry=77139082

[10] The dangers of cloud computing

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]