The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was Keep. There seems to be a fairly substantial consensus to keep this article. - Smerdis of Tlön (talk) 16:25, 5 November 2008 (UTC)[reply]

Google platform[edit]

This article attempts to describe non-public information about the inner workings of a technology company. Although Google is obviously notable, internal information about the company's workings is not encyclopedic. Relevant information can be merged into the main Google article or a History of Google article, but as it stands this article does not meet WP:V or WP:NPOV.

The main problems can be summed up by the article's own introduction: This article describes the technological infrastructure behind Google's websites, as presented in the company's public announcements. As such, it has inherent WP:NPOV problems.
Because the article purports to describe internal knowledge about Google's operation, it cannot satisfy WP:V because that information cannot be verified by external sources. Many of the references are from Google's own public information, either first hand or second hand.
Those references which aren't directly or indirectly from Google are press reports describing Google building new facilities. While that can be verified, the lion's share of the information in these article cannot be. (Except, perhaps to say that large datacenters take a lot of power, which isn't unusual.)

While I love Google as much as the next guy, I don't believe that this article merits inclusion as it stands.JRP (talk) 07:15, 30 October 2008 (UTC)[reply]

Keep - this article describes the operation of possibly the most important website in history. We have articles describing the internal operation of cars, microprocessors, chemical engineering plants, and so on. I think the article should be renamed to something like Google service delivery and rewritten a bit. - Richard Cavell (talk) 08:50, 30 October 2008 (UTC)[reply]
- Cars, microprocessors, and engineering plants can (to a greater or lesser extent) be independently confirmed. A mechanic can look under the hood. Processors have been reimplemented by multiple, independent companies. Engineering plants similarly exist in many places from many companies and the process of designing one is well-known. However, this article is not on a cloned microprocessor or a common chemical engineering plant design but ONE specific platform for ONE specific company which has not been reimplemented, re-engineered, or independently evaluated. Without that, we can't possibly satisfy WP:V and we are at best left with trusting that their public communication is accurate or at worst dealing with press speculation. I'm not saying that there can't be an article, only that this article as it stands doesn't meet the bar for inclusion. JRP (talk) 15:45, 30 October 2008 (UTC)[reply]
  - So far as notability is concerned, I stand my ground. As to verifiability, I want the article to be sourced, as all articles should be. I like the example given below of North Korea's nuclear program. It is entirely acceptable for wikipedia to piece together information from what is publicly known, even though the subject is kept secret by the authority that owns that secret, and even though the information that is publicly known might be incomplete or just plain wrong. Google's shareholder reports should be accurate, since Google is bound morally and legally to report truthfully to its shareholders. - Richard Cavell (talk) 05:10, 31 October 2008 (UTC)[reply]
Keep - The inner workings of the most notable website is quite notable. The contents of this article definitely warrant inclusion in WP. The discussion of a possible rename should not be done here. Zain Ebrahim (talk) 09:42, 30 October 2008 (UTC)[reply]
- You are mis-using "notable" for a subjective evaluation of how important you consider something. Notability is not fame nor importance.
  And you haven't addressed at all the central challenge in the nomination, which is that Wikipedia editors piecing together the internal workings of a company by extrapolating what the company says in its public announcements, is a novel synthesis of sources, prohibited by our Wikipedia:No original research policy. Please counter that. (Hint: A brief search for books reveals that countering it will be relatively easy.) Uncle G (talk) 10:38, 30 October 2008 (UTC)[reply]
  - Even if we define notability by the existence of reliable secondary independent sources, Google could still be reasonably claimed to the most notable website.
    But you're right about the rationale - I haven't dealt with the central argument posed by the nominator. I'll just counter by saying that any OR should be removed (and the rest of the article shouldn't be punished). We don't delete articles about notable subjects simply because they're currently full of OR. Looking at the reflist and a quick search reveal several sources (I haven't checked for reliability, though) explaining Google's technology. If there isn't enough public information then that's another issue. Zain Ebrahim (talk) 11:34, 30 October 2008 (UTC)[reply]
    - My argument is that there is insufficient verifiable public information to make this article currently and what public information there is is tainted because it only comes from the source itself. That's not a solid foundation for a Google article. Imagine if we had an article about the internals of Microsoft Windows that only sourced Microsoft technotes and press reports? That wouldn't be good. In the MS case, we can crack open the hood (and many have) with tools and see how things actually do work. In the Google case, their datacenters are literally under lock and key. They don't allow the press inside almost any of them. There is no way to independently verify any of the information presented. JRP (talk) 15:49, 30 October 2008 (UTC)[reply]
      - I find it hard to believe that any information about google's infrastructure can only be found in their own publications. Also, for simple facts, there's nothing wrong with using primary sources (we can report on a company's balance sheet using its financial statements without violating WP:NOR). Zain Ebrahim (talk) 11:45, 31 October 2008 (UTC)[reply]

Keep - Different reporters have visited Google and talked to them. Usually these reporters' findings are published in reliable sources. In some cases, Google would not tell them everything they wished to know, but if that happens, we can note it in our summary. I'd change the second sentence of the article to This article describes what is publicly known about the technological infrastructure behind Google's websites. Large companies don't tend to be 100% transparent, so there is no reason to single out Google for special skepticism. EdJohnston (talk) 15:34, 30 October 2008 (UTC)[reply]
- Findings published in reliable sources are only as good as their primary sources. (As folks following, for example, war reporting know.) In this case, as you acknowledge, all primary sources are the company itself. That's putting a lot of trust in them and makes those sources not be verifiable. There are verifiable sources in this article-- I cite the several press reports about their datacenters which can be independently confirmed-- but the largest portion of this is speculation and based on Google press information. Not a good basis for an article. JRP (talk) 15:45, 30 October 2008 (UTC)[reply]
  - I question whether a statement like Findings published in reliable sources are only as good as their primary sources can be found among our policies. Do you mean that Wikipedia should not have articles about secretive companies or organizations? We *do* have an article on the North Korean nuclear program. I have added a mention of Randall Stross's 2008 book, Planet Google, to the article. If you find that the article still contains undue speculation, you should say more about that on the article's Talk page. EdJohnston (talk) 20:59, 30 October 2008 (UTC)[reply]

NOTE Here's another quote from the article: In a 2008 book, the reporter Randall Stross wrote: "..Google's executives have gone to extraordinary lengths to keep the company's hardware hidden from view. The facilities are not open to tours, not even to members of the press." He wrote this based on his own experience of visiting the company and interviewing staff members. If this is true, then this article cannot pass WP:V. JRP (talk) 21:24, 30 October 2008 (UTC)[reply]

That's the quote I just added. I'm still waiting for you to nominate our article on North Korea's nuclear program for deletion :-). EdJohnston (talk) 21:39, 30 October 2008 (UTC)[reply]

Ha! No, North Korea and weapons of mass destruction is a well-researched article which deals in large part with the speculation itself, rather than being the result of the speculation. (And the speculation is sourced, so at least that the source said blah is verifiable.) That and the spies in N. Korea leak to the press. :) JRP (talk) 22:47, 30 October 2008 (UTC)[reply]

I think you are misunderstanding WP:NPOV. You are confusing two things - one is that there is only one point of view in the article (Google's), and the other if each point of view is sourced. Having only one significant point of view doesn't contradict WP:NPOV in any way, and this point of view is sourced, so it's verifiable (for example, it's verifiable that Google says it uses BigTable, not that they really do). Samohyl Jan (talk) 06:24, 5 November 2008 (UTC)[reply]

Speedy Keep The nomination is proposing a merger and so AFD is inappropriate. In any case, the content seems good and has adequate sources. Colonel Warden (talk) 10:41, 31 October 2008 (UTC)[reply]
Note: This debate has been included in the list of Websites-related deletion discussions. -- Raven1977 (talk) 05:33, 2 November 2008 (UTC)[reply]
Keep I just came to the page looking for the list of software in Google's technology stack, so it's definitely useful, although the list I was looking for is missing from the article for now. It may be about internal workings of a company, but they are doing things so differently than others (they rely on their own technologies) that it definitely warrants an article. Samohyl Jan (talk) 22:06, 2 November 2008 (UTC)[reply]

Relisted to generate a more thorough discussion so that consensus may be reached.
Please add new comments below this notice. Thanks, treelo _radda 20:50, 4 November 2008 (UTC)[reply]

Delete Synthesized essay on the inner working of Google? Unencylcopedic, inappropriate and not a concise "topic" - Jimmi Hugh (talk) 21:48, 4 November 2008 (UTC)[reply]
- Can you be more specific about what information would you delete from the article? And if you wouldn't delete anything from it, into what article would you add it? Samohyl Jan (talk) 06:24, 5 November 2008 (UTC)[reply]
- You failed to justify your claims that Google's infrastructure is unencyclopedic, inappropriate and "not concise as a topic" (whatever that might mean). This is not a vote, this is a debate. Zain Ebrahim (talk) 07:23, 5 November 2008 (UTC)[reply]
- My thoughts exactly. Going over my copy Computer Architecture: A Quantitative Approach and the IEEE Micro article, much of the "synthesized essay" can be verified. Rilak (talk) 07:48, 5 November 2008 (UTC)[reply]
Strong Keep - If this is "non-public information" then why is this subject extensively and comprehensively discussed in John L. Hennessy and David A. Patterson's Computer Architecture: A Quantitative Approach, Third Edition? Am I one of the elite few who have access to this book? I find this AfD to be ridiculous. We should be going after Pokemon, not notable academic subjects such as this. Rilak (talk) 06:35, 5 November 2008 (UTC)[reply]
Keep - notability well established, nom is otherwise outright wrong. Wily D 14:00, 5 November 2008 (UTC)[reply]
Obviously my hint above wasn't strong enough. I've cited some of the books for you. The References and Further reading sections of the article speak for themselves. Keep. Uncle G (talk) 14:14, 5 November 2008 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.