We don't care what happens to your website (unless it's Google)

I've been wondering for some time about the way deletion discussions formally use Google as the default search engine for the "find sources" line at the top of discussions. Why do we not offer similar links to other search engines? Are we down to one? Who made that call? Aside from sheer facility, why do we favor one over all others? Has this previously been a discussed topic? BusterD (talk) 22:10, 23 September 2011 (UTC)

I won't argue that Google is the best search engine available (that debate could go on forever without resolution) but it is by far the most prominent, the search engine has about a 2/3 market share. The term "google" is in the Merriam-Webster Dictionary (as a verb). That ubiquity makes it the obvious default search engine, and frankly for a quickie search it's fine to just have one search engine. -- Atama頭 22:37, 23 September 2011 (UTC)

Doesn't that put Wikipedia in the position of picking winners? Is that what we should be doing as a community? Has this EVER been discussed? I'll concede that other search engines haven't built the same massive collection of raw searchable data (I use googlebooks constantly), but I'm starting to chafe at the obvious COI between the most powerful internet company in the world and the most powerful user-built encyclopedia. Hence the title of this section. I myself could certainly use another method, and I'm sure other editors are sensitive to this issue, but IMHO to have formalized use of one company in procedures seems a poor course, long-term. BusterD (talk) 23:01, 23 September 2011 (UTC)

When it comes to AFD discussions, the ability to quick switch search venues from the web, to news, to journals, to books, and so on just by clicking on one button is beyond invaluable. This doesn't prevent any other search site from being used, but google's toolset is so complete specifically to help resolve AFD in the right manner. --MASEM (t) 23:06, 23 September 2011 (UTC)

So it's mere facility. And why isn't Google contributing for all this free linking and SEO? BusterD (talk) 23:17, 23 September 2011 (UTC)

Because they don't have to? Wikipedia is essentially taking advantage of free services Google is providing, and if Google indirectly benefits then it's a mutually advantageous arrangement. Also, I think you'll have to explain what COI there is with Wikipedia using Google for the basis of convenience links for search engines, it's not nearly as obvious as you seem to think. -- Atama頭 23:51, 23 September 2011 (UTC)

While I agree with Atama and Masem, I am curious as to your proposed solution. How would you propose changing the "Find sources" line - while still being able to keep the list of links compact, simple (ie: one-click), and reliable. --- Barek (talk • contribs) - 00:09, 24 September 2011 (UTC)

I don't have a solution; I have a valid concern. Wikipedia doesn't have to choose one particular search platform over all others, just because it's the most popular and high-performing as of this datestamp. It will be difficult to find an appropriate comparison, but I'll attempt one, albeit imperfect. Let's say that some important Wikipedia function, let's say the admin bit, only functioned with editors using Microsoft operating systems as a platform. For some particular technological reason, the only way blocking and protecting worked was with editors running Windows. Much like the situation I've described above, Wikipedia would be choosing a market winner. In this case of search engines there's only a superior advantage of functionality (at the current time), yet the community has sanctioned this favoritism of a company in a particular formal way. Google takes advantage of Wikipedians' voluntary contributions to make money. Wikipedia gets the same thing every Google user gets: free links to information. I ask for the third time, has this ever been discussed? Where? Who made this call? BusterD (talk) 00:19, 24 September 2011 (UTC)

I might not ever have been discussed (I don't know of anywhere it has been, anyway), but WP:SILENCE applies. No one's objected until now, so it's assumed not to be controversial. So... Unless you think there's an issue with using Google, I don't see what there is to do here. lifebaka++ 00:48, 24 September 2011 (UTC)

In the case of my example above, in such an obvious case of dependency, we'd have technological solutions constructed. No longer silent, someone has to raise an issue for the first time. While I don't doubt the good faith of the editors in this discussion, I have every reason to believe that a company like Google has many paid employees who monitor the quantity and type of links between the two internet institutions. Sure would like an answer to my question about discussion. I might link to this discussion on Jimbo's talk. BusterD (talk) 00:58, 24 September 2011 (UTC)

BuseterD, check out User:Kim_Bruning/Lost_functionalities for some added insight. Your fellow unpaid Google employee, Viriditas (talk) 01:23, 24 September 2011 (UTC)

Thanks very much for the link. I'll read. I'm not trying to stir crap, I just thought while my title was somewhat provocative, I've raised a legitimate issue. And since Google knows my real name and address, I suspect I've signed my own death warrant. (that's a joke, actual Google employees, ok? Really, a joke.) BusterD (talk) 01:28, 24 September 2011 (UTC)

I think basically it's because WP:We don't care what happens to your website. It just isn't important. Also, the links at AFD are on pages that pretty much only a handful of editors (rather than our millions of readers) see. The typical AFD page gets 30 views in the course of a full week's discussion. Only a small number of those page views will result in someone clicking the link. Google is unlikely to even notice. WhatamIdoing (talk) 02:06, 24 September 2011 (UTC)

You mean, WP:We don't care what happens to your website (unless it's Google). We certainly don't care about Yahoo, Ask, or Bing. Google, we seem to be defending (as an institution) in several formal and informal ways. BusterD (talk) 02:22, 24 September 2011 (UTC)

BusterD, the best thing to do is write an essay about your concerns. Make sure you provide alternate proposals for consideration. Then, present it here (or elsewhere) as a link for discussion. Viriditas (talk) 02:59, 24 September 2011 (UTC)

Sound advice. I'm certainly not trying to put anyone on the defensive. I hope nobody takes it unkindly if I use the redlinked title above (but in my userspace, not project space). Still am nervous about those black helicopters circling in the rainstorm above. BusterD (talk) 03:05, 24 September 2011 (UTC)

I actually think BusterD's proposal is a great one. I won't argue that Google is the best service, but I will argue that all of the search engines have different algorithms and you have a good chance of finding different results on each.--v/r - T P 13:11, 24 September 2011 (UTC)

Proposed Links

What about abbreviate the searches? G for google, B for Bing, Y for Yahoo. We don't need the full names. Example:

(Find sources: "Subject" (G:B:Y) – news · books · scholar · free images)

What do you think?--v/r - T P 13:09, 24 September 2011 (UTC)

A number of alternative templates exist with links chosen as the most useful rather than just Google. Compare at Category:Search templates. It would be easy to have a random choice too, not sure how acceptable this would be for most users though. It might be an idea to think of designing an extension to the current template (i.e. the only change would be a "more..." button) that then optionally opens up to show a selection of alternates. We could then think of a poll for swapping links once users have a while to play with other sites. Fæ (talk) 13:21, 24 September 2011 (UTC)

I think random would be a bad idea, editors should know what to expect when they click the search links. I don't see a problem with adding more options though. Monty ₈₄₅ 15:39, 24 September 2011 (UTC)

I totally agree that random doesn't suit the nature of searches. Searches are about choices. It's somewhat revealing that the only templates categorized by company are the Google ones. This belies the assertion "we don't care what happens to your website." I think the approach User:TParis makes is a useful one, like tabs across the top of a WikiProject space. One click gives a different set of options. More choices is good, so long as (pointed out by editors above) the whole thing doesn't become cumbersome and slow anyone down. The effort to find and apply sources should be as effortless as possible, IMHO. The simpler and more elegant the technological solution, the better for now and for later. (After all, we're merely ten years in; imagine what the search engine options will be as more institutions digitize their existing resources.) BusterD (talk) 15:43, 24 September 2011 (UTC)

After thinking about it, I realized this issue is pretty comparable to the ISBN solution used to present links to multiple book sources. However, for practical purposes, I still think it's better to keep the links simple (ie: one-click, not needing to navigate through a sub-page as is done for ISBN links).

Ideally, I believe the best solution would be to have the "My Preferences" menu include a drop-down to allow each user to select a preferred search provider ... then a template able to utilize that preference to present custom links to each user who views a link formed with that template. But, that would be time-consuming to develop once proposed to the Village Pump. Barring that, as an interim solution, I have no major objection to providing multiple links as long as the presentation of multiple links can be kept compact (perhaps via a collapse/expand toggle for the search template).

It would still remain to decide which ones to link. Maybe Google, Bing, Ask, and AOL? Last I heard, Yahoo uses Bing as its search engine - so those results would be redundant to a Bing link - although I could see an argument if some users prefer the layout of those results in Yahoo over Bing.

Also, the proposal above only presents alternates to the main search ... but the additional links in that template would have alternates possible (ie: news · books · scholar · free images). Once alternates are provided to all of these, any template could quickly grow to a rather bulky header block. I would want to see proposals for how to manage all of the links in an easy to navigate and compact format. --- Barek (talk • contribs) - 20:24, 24 September 2011 (UTC)

Comment - A drop-down menu with multiple search options (stated by user:Barek above) would, in my opinion, be a significant improvement. Northamerica1000 (talk) 14:41, 26 September 2011 (UTC)

Comment - Perhaps the addition of a link such as this to the "find sources" line at the top of discussions would provide more comprehensive search options: Additional News Sources. Northamerica1000 (talk) 10:58, 26 September 2011 (UTC)

Comment - Another idea is to add this type of a link to a Wikipedia page which lists additional reliable news sources that have a reputation for editorial integrity and unbiased coverage. Northamerica1000 (talk) 12:10, 26 September 2011 (UTC)

I would very much support removing the link to web searches entirely. We should be encouraging users to default to Books, News archive and Scholar before bothering with a web search, since the former three concentrate reliable sources, and we have lots of problems with people using web only, or at least going there first, and the current setup fosters this by placing the web link first (it's not as if people don't know how to do a web search afterwards, if they want to). If we are to include any web links then they should appear after the others. I would have no problem in placing others there as a matter of "sheer facility" but only because it might be a benefit to our users. This may seem slightly tangential but it's not. Google is the only game in town for what we really need. This has nothing to do with allegiance or favoritism to Google and everything to do with usefulness for ourselves and our users. Whether it's fair to other search engines doesn't enter the equation because the reason we link to Google is completely unbiased. If Wikipedia was a free hostel and AfD was where scrambled eggs were served, Google would be a fork manufacturer and other search engines would provide only steak knives. Yes, it's "sheer facility", and what's wrong with that?--Fuhghettaboutit (talk) 12:32, 26 September 2011 (UTC)

IMHO, in an internet where nothing stays the same for long, putting all your eggs in one online basket would be a poor choice indeed. I wouldn't do it with a grocery store, I wouldn't do it with a pizza parlor, I wouldn't do it with an operating system. I certainly wouldn't do it with a web browser. Getting in bed with one vendor exclusively not only gives the vendor an initial advantage, but it gives the vendor leverage which can be used against you when they make changes in policy. And policies will always change. The recent New York Times paywall is one such policy change which could have a long-term negative effect; since many history cluster editors have been regularly linking to Times articles as historical news sources. When the vendor decides to put more books or more scholars behind a paywall or some other profit-based change (which seems likely if not inevitable), then all the links we're putting in to cite pagespace will have a larger effect to advertise one vendor exclusively. That's what's wrong. Today is fine; it's ten years down the road I'm concerned about. Maybe I'm concerned about nothing; maybe I have a point. Jimbo seems to take a bit of both sides, and that's perfectly fine with me. BusterD (talk) 01:43, 27 September 2011 (UTC)

But the problem I'm getting at is that the only real "eggs" for these search engines are books, news archive ans Scholar, which happen to be Google services. Web is the last search anyone should try, and I don't see any reason we shouldn't include links to others for web searches, but web searches should be put in the back room, put at the end as an afterthought. Meanwhile, there is no free search service that has any useful comparative tool to news archive, books and scholar. Worldcat is near useless for what we need. NYT.com is just one newspaper that is already searched by Google News. Again, there is no allegiance to Google, it's all about utility, and nothing else provides the utility. So I don't care what the name brand is, and if another brand pops up that provides a better utility not only would we switch without qualm, but doing so would be a few minutes work. Accordingly, I think the concern over "ten years from now" is a red herring.--Fuhghettaboutit (talk) 22:58, 27 September 2011 (UTC)

To use another example, why aren't we linking to WorldCat in "find sources"? IMHO, it's a way more reliable search engine for books and authors than Google; the difference is that Google links to actual content, where WorldCat merely gives extremely good information on where to find the works in a library. BusterD (talk) 01:54, 27 September 2011 (UTC)

Another example of internet policy creep and effects thereof: Like many, I enjoy watching "The Daily Show" and "The Colbert Report" online. I preferred online viewing because of the fewer commercials than cable TV. At first, I watched these shows on their respective Comedy Central websites linked from Viacom, and they inserted one commercial between the clips. Then they moved to two commercials. These days, there's a Comedy Central commercial at the top of the show, then a couple of commercials, then several (often as many of four) between the segments. Of course, I could watch on Hulu, and if I pay for HuluPlus, I can even watch on my mobile device. These days, the Viacom sites work less reliably, so I'm stuck with Hulu. I'm not saying Viacom is forcing me to use HuluPlus, but the trend is obvious, and not accidental. Let's apply this anecdotal analysis to my concern here: In my view, when Google has Wikipedia lined up exactly the way they like it, they could conceivably drop a paywall down, and our only option would be to remove the links we've applied on millions of articles, or put our users in the position of paying for access. Now one strategy for dealing with this is to "relax and enjoy it", and another might be to protest the Google campus, but the wrong approach would be to hold our hands over our eyes and ears and sing la-la-la. IMHO, ignoring the inevitable can't be considered a reasonable strategy. I'll concede my examples aren't precisely applicable, but close enough to demonstrate my concern. BusterD (talk) 02:42, 27 September 2011 (UTC)

Well, if it works today, I'm not certain why we'd need to change it today. When things change, we can change things, this isn't controversial. Now, we still might want to change the way we're doing things, but the possibility of circumstances outside of our control changing in the future isn't a good thing to base changes here off of.

As for GBooks vs. WorldCat, generally the former is going to be more immediately useful in an AfD, where users might not be able to locate a copy of the book (suppose, for instance, that it's checked out) and a relevant passage in it within an AfD's seven day time limit. I'm also unsure of how useful WorldCat would be if I didn't already have a good idea of what I was looking for (i.e. book title, author, etc.). We should certainly make WorldCat more prominent and encourage its use to find sources (I wasn't aware of it until just now, for instance), but I'm not at all convinced of its utility for AfD specifically. We also ought to look into how both platforms choose their results, as two searches for the text "wikipedia censorship" (without quotes) returns 10 hits on WorldCat and about 1600 on GBooks, with none of the first page GBooks results being the 10 WorldCat results. Cheers. lifebaka++ 02:50, 27 September 2011 (UTC)

Wow. I disagree violently that we should wait to respond to change even if we can see it coming down the road. To paraphrase R. Buckminster Fuller (I think from the beginning of Operating Manual for Spaceship Earth): "Imagine you're on a boat at sea, and the boat sinks. Swimming around, you come across a piano lid, and you climb on. When the rescue comes, you're fine. You may feel strongly about your survival, you may even grow to have an attachment to the design. But just because the piano lid was successful in your situation doesn't mean it's the optimal design for a life preserver. That's what we do every day; we latch onto piano lids as we're swimming by. What we should do is think hard about such situations BEFORE they arise." (Sorry, Buck, if I've misinterpreted you. You know you're my main guy.) I don't disagree that Google provides a better product, but WorldCat is a valuable professional research tool available for free as well (at present), and shouldn't be disregarded as an essential search tool. Putting all our search eggs in one corporate basket is just foolish, given the history of the internet. (I'll confess to watching James Burke's Connections series on YouTube the last few weeks, so I've been thinking hard about the way the future will change the internet.) Hey, I may be wrong; but I could be channeling Willie Stark, too. BusterD (talk) 03:07, 27 September 2011 (UTC)

But... We don't see anything coming, right? That's rather the point. Of course, if we know something is likely to happen, we should prepare for it, but I don't see the point in making changes based on vague possibilities. I'm not aware of anything coming up that we'd need to adapt to, and I haven't seen you point out anything specific either. The OMfSE example is disingenuous, because on a seafaring ship you're perfectly aware of the possibility of falling into the water (for one reason or another), whereas you are asserting we should prepare for some vague and undefined possible future. I'm sorry if I didn't make my point clearly before. lifebaka++ 04:52, 27 September 2011 (UTC)

I'd say disingenuous is a poor choice of words, because (with no evidence but my own 22-year experience with the internet) I do imagine at least one Google profit-raising policy change coming in the next ten years, and probably more than one. That's what my NYT/Daily Show experience examples represent, a patient corporate ramp-up of profit-intended practice. Fuller's point was that we never consider these things at all until it's too late, then we grab onto the first available solution, or hang on more tightly to the ones we've used in the past, even if impractical. If consensus doesn't agree with my position, I can handle that. For my part, I'd prefer to be candid about what I see as red flags down the road instead of saying "I told you so" afterwards (my Willie Stark metaphor). Even now I'd say the mere appearance of gfavoritism improves their stock value. Are you saying we shouldn't consider investing in a poncho or an umbrella just because the weather looks fine today? BusterD (talk) 11:07, 27 September 2011 (UTC)