قالب وردپرس درنا توس
Home / Technology / Google & # 39; Incognito & # 39; The search results still vary from person to person, DDG study finds – TechCrunch

Google & # 39; Incognito & # 39; The search results still vary from person to person, DDG study finds – TechCrunch

An investigation of Google's search results of anti-tracking rival DuckDuckGo has suggested that escaping the so-called "filter bubble" of personal search on the web is a harmful difficult issue for what's put on Internet consumers who just want to stand out little

DDG expects not possible even for logged-in users of Google searches that also surf in incognito mode to prevent their online activity from being used by Google to program – and thus shape the results they see.

DDG says it found significant variation in Google search results, with most of the study participants who got results that were unique to them ̵

1; and some people simply do not see links.

The results in news and video info boxes also varied significantly, it found.

While it says there was very little difference for logout, incognito b

"It's simply not possible to use Google search and avoid the filter bubble," concludes.

Google has responded to the fact that DuckDuckGos research is "wrong".

Degrees of Personalization

DuckDuckGo says it has conducted the survey to test recent Google requirements to tweaked its algorithms to reduce personalization.

A CNBC report in September, referring to Google's access, and allowing the reporter to attend an internal meeting and talking to employees in the algorithm team suggested that Mountain View now only uses very little personalization to generate search results.

" A query a user comes with usually has so much context that the possibility of personalization is only very limited," Google fellow Pandu Nayak, who heads the search ranking team, told CNBC this fall.

On the surface, it would represent a radical reprogramming of Googl e-search mode operandi – given the company made "Personal Search" by default for self-logged users back in 2009.

Announces the extension of the feature, Google explained that it would "match" the search results for these logged-in users via an anonymous cookie & # 39 ;:

This app lets us customize search results for you based on 180 days of search activity associated with an anonymous cookie in your browser. It is completely different from your Google Account and Web History (which is only available to logged-in users). You know when we match results because a "Show customizations" link appears at the top right of the search results page. By clicking on the link, you can see how we've adapted your results and also let you down this type of customization.

A couple of years after Google deleted the Personal Search Switch, Eli Pariser published his now famous book describing the filter bubble problem. Since then, the bad press of online personalization has only grown.

In recent years, the concern has particularly spited over the horizon-reducing effect of big tech's subjective traits on democratic processes, with algorithms carefully designed to continue to serve users more of the same cases now widely accused of anchoring partisanian opinions, rather than to help expand people's horizons.

Especially so where political (and politically charged) topics are busy. And ultimately, algorithmic filter bubbles are accused of breaking democracy themselves – by creating highly efficient distribution channels for individually targeted propaganda.

Although there have also been counterclaims floating around academic circles in recent years involving the Ecclesiastes is even overblown. (Albeit sometimes from institutions that also take funding from tech giants like Google.)

As always, where the operational opacity of commercial algorithms is concerned, the truth can be a very difficult animal to dig out.

Of course, DDG has its own self-employed iron in the fire here – which suggests that "Google affects what you're clicking" – given it offers an anti-spam option for the same Google search.

But it does not deserve an immediate termination of a result of great variation in obviously seemingly "incognito" Google search results.

DDG has also made data from the study downloadable – and the code it used to analyze data open source – so that others could view and draw their own conclusions.

It conducted a similar study in 2012, following the former US presidential election – and claimed to have found that Google's search had put tens of millions of links to Obama tha n for Romney in the run-up to it.

It said it wanted to revise the status of Google search results now in the wake of the 2016 presidential election that installed Trump in the White House – to see if it could find evidence to back up Google's claims to have "de-personal" search.

For the last study, DDG asked 87 volunteers in the United States to search for politically charged "pistol control", "immigration" and "vaccinations" (in order) at 21:00 ET Sunday, 24 June 2018 – basically searching for private browsing mode and logged out of Google, and then again without using incognito mode.

You can read the full review of the study results here.

The results ended up being based on 76 users who searched for mobile phones were excluded to check for significant variation in the number of info boxes shown.

Here's the top line of what DDG found:

] Private Browser Mode (and Logged Out): [19659030] "Pistol Control": 62 variations with 52/76 participants (68%) see unique results.

  • "Immigration": 57 variations with 43/76 participants (57%) see unique results.
  • "vaccinations"
  • & # 39; Normal & # 39; Mode:

    • "Pistol Control": 58 variants with 45/76 participants (59%) see unique results.
    • "Immigration": 59 variations with 48/76 participants (63%) see unique results.
    • "Vaccinations": 73 variations with 70/76 participants (92%) see unique results.

    DDG's claim is it really & # 39; impartial search results should yield essentially the same results.

    Nevertheless, the search results that volunteers received were – in the majority – unique. (From 57% at the low end to a full 92% at the upper end.)

    "With no filter bubble, one can expect to see very little variation of search results – almost everyone would look the same simple set of results, "writes it. "Instead, most people saw the results unique to them. We also found the same variation in private browsing mode and logged out of Google vs. in normal mode."

    "We often hear confusion that private browsing mode enables online anonymity, but this finding shows that Google tailored search results regardless of browsing mode. People should not lulled into a false sense of security as the so-called "incognito" mode makes them anonymous, "adds DDG.

    Google refused to provide a statement that corresponds to the study first, and instead tells us that more factors can contribute to variations in search results – flagging time and placement differences between them.

    The alleged results may vary depending on the data center a user request was associated with – possibly introduces any robotsearch based on microscope.

    Google also claimed that it does not personalize the results of logged-out users who browse Incognito mode based on their logged-in search logos.

    However, the company admitted that it uses contextual signals to rank results even for logged-out users (as the 2009 blog post described) – for example, when you try to clarify a questionable query.

    In which case, it said a recent search can be used for disambiguation purposes. (Although it also described this type of contextualization in search as extremely limited, it said that would not be dramatically different results.)

    But with so much variation that is evident in DDG volunteer data, there seems little question about Google's approach very often results in individual – and sometimes highly individualized search results.

    Some Google users were even served with more or fewer unique domains than others.

    Many questions naturally run from this.

    Like: Does Google make a small "contextual contextualization" audio as an adequate "de-personal" approach – if the name of the game pops on the filter bubble?

    Are the results served even marginally less clickable, biased and / or influential?

    Or actually any less "rank" from a privacy perspective …?

    You tell me.

    Even the same gang of links served in a slightly different configuration has the potential to be significant since the top search link always gets a disproportionate amount of clicks. (DDG says that the no-link is about 40%.)

    And if the topics like Google search are politically charged, small variations in search results – at least theoretically – can contribute to some major democratic effects. 19659002] There is a lot to chew on.

    DDG states that it is checked for time and place-based variation in served search results by having all participants in the study perform the search from the United States and do it at the same time.

    While it says that it is checked for local link inclusion (i.e., interrupting location-based variation) by combining such results with a localdomain.com placeholder (and & # 39; Local source for info boxes).

    Still, even if you take steps to control time-based variations, it has still been found that the majority of Google search results are unique to the individual.

    "These editorial results are informed by personal information Google has on you (like your search, browsing and purchase history) and puts you in a bubble based on what Google's algorithms are most likely to click on , "argues it.

    Google will oppose what is contextualizing, not editorial.

    And that some "small variation" in the results is a natural property of the dynamic nature of its Internet crawling search response business.

    Although, as mentioned above, DDG found that some volunteers were not served specific links (when others did), which sounds more important than "small difference".

    In the statement, Google sent us later describing DDG's attempts to control time and place displacements as ineffective – and the study as a whole as "wrong" claims:

    This study's methodology and conclusions are incorrect as they are based on the assumption that The difference in search results is based on personalization. It's simply not true. In fact, there are a number of factors that can lead to small differences, including time and place, as this study does not seem to have been verified for effective.

    One thing is crystal clear: Google is – and has always been – making decisions that affect what people see.

    This capacity is undoubtedly influential given most market shares taken by Google search. (And the big role that Google still plays in shaping what internet users are exposed to.)

    Clearly, even without knowing all the details about how personal and / or custom these individual Google search results were.

    Google's programming formula remains unlocked in a proprietary algorithm box – so we can not unpick it (and independently).

    And this unfortunate techno-opacity & # 39; the habit provides practical coverage for all kinds of demands and counterclaims – which really can not

    Until and until we can know exactly how the algorithms work to track and quantify consequences.

    Also true: Algorithmic accountability is a theme of increasing public and political concern.

    ] Finally, "trust us" is not Google's great branding once it was.

    So the devil can still (manually) become unknown from all these fuzzy details.

    Source link