Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Google

Google Testing a Radical Change By Turning People's Search Results Black (telegraph.co.uk) 107

Posted by msmash from the black-is-the-new-black dept.
Google may have plans to do a visual tweak to its search results. The company appears to be testing black search result links since the weekend, according to multiple reports. While some users are pleased with this tweak, many users have already posted their grievances on Google help forums. Some users note that it has become hard to tell which links they have already clicked. The Telegraph reports: Google puts a lot of thought into the exact colours it uses in its services -- and for a good reason. A few years ago its A/B test of different shades of blue -- nicknamed "50 shades of blue" -- earned the company an extra $200 million (£138 million). Designers at Google couldn't decide between two different blues, so they decided to test 41 shades between each blue to see which users preferred. In the test, Google showed each shade to one per cent of its users, and found that users were more likely to click on a slightly more purple shade.
This discussion has been archived. No new comments can be posted.

Google Testing a Radical Change By Turning People's Search Results Black More

Google Testing a Radical Change By Turning People's Search Results Black

Comments Filter:

  • In case anyone was wondering (Score:5, Insightful)

    by Anonymous Coward on Monday May 09, 2016 @03:42PM (#52078239)

    "I really feel like I'm frenemies with #Google. Black links instead of blue in the search results? No. Just No. Bad Google. Bad Google."

    We really need a new mental health initiative in this country.

  • Stupid (Score:5, Informative)

    by swalve ( 1980968 ) on Monday May 09, 2016 @03:50PM (#52078311)
    How do we know it's a link if it's the same color as the text? The whole point of hypertext is that links are called out visually.

    • Re:Stupid (Score:5, Funny)

      by 110010001000 ( 697113 ) on Monday May 09, 2016 @03:51PM (#52078329) Homepage Journal
      What are you saying? They it all looks alike? That is kind of racist.

    • If things stand out it spoils the flat look!

      [adjusts trilby, mounts fixie and rides off down the sidewalk while texting, oblivious to the little old lady he ran over]

      A/B testing indeed. Shitcockery of the first order.

      • Is your beard ironic too?

        • I had a beard when most Uxshuists' fathers hadn't started shaving, And I've flipped between mullahesque bush & Hungarian anarchist goatee since.

          If you do it for more than 2 days - the rise and fall cycle of your typical hiptard - then no, it's totally bastarding not ifuckingronic.

    • Re: (Score:1)

      by Anonymous Coward

      I like the way DuckDuckGo does it. Links are clearly distinct, and if you've been somewhere before a big check mark appears in front of the result.

      • Re: (Score:1)

        by Anonymous Coward

        I wonder how they do that since they're "the search engine that doesn't track you"...

        • Re:Stupid (Score:5, Informative)

          by mrchaotica ( 681592 ) * on Monday May 09, 2016 @06:20PM (#52079451)

          The :visited selector [w3schools.com], I'd assume.

          Distinguishing visited links has been built into browsers since the beginning of the Web, although styling it via CSS is newer.

          • Unfortunately, this also makes it quite easy to leak the browsing history to malicious sites. Even if you can't query the generated style for a link (some browsers now prevent this), you can use the canvas extension to render links to an image buffer and then just read back the colour values.

    • Re: (Score:3)

      by AmiMoJo ( 196126 )

      I had this happen to me. When you mouse-over an underline link appears. It's kinda crap because you have to mouse over everything to discover what is clickable.

      Like a mid 90s management sim where you end up clicking on the cheese plant in frustration.

    • How do we know it's a link if it's the same color as the text? The whole point of hypertext is that links are called out visually.

      No, no, this is good. This is the penultimate step in the interfaceless-UI that Google and their ilk are shooting for. Give it another couple years and it'll be black links on a black background, and you can simply shut your computer off.

      Or it's stupid.

      • It's even better when you're on a touchscreen device where there isn't a mouseover event. Fortunately for Google, no one uses mobile touchscreen devices to browse the web.

    • How do we know it's a link if it's the same color as the text? The whole point of hypertext is that links are called out visually.

      CSS can be used to change the followed link color. http://www.w3schools.com/css/c... [w3schools.com]

      The problem is if someone's browser overrides that setting, for example.

      Some people find darker backgrounds easier on the eyes--there is less light emitted so it is not as big a change from ambient indoor light.

      However, studies have shown that black text on a white background results in easier focus, so there are some people where black-on-white is better than white-on-black. https://ux.stackexchange.com/q... [stackexchange.com]

      Conclusion: if

    • The visited links will likely be darker, "blacker", but not black! It's been an issue in Google search for a long time, visited and non-visited links having the same color ("#1a0dab"). Why does Google keep the same color for both links is a mistery, though.

    • How many people remember the way your search terms used to be highlighted in color on Google's cached copy of a page?

      I would always click to the cached copy rather than the original page. When your eye was immediately drawn to the highlighted words you were searching for, it was a huge timesaver (especially on large pages; no need to use your browser's Find command).

      Now multiply that time savings by the billions of Google searches that are conducted every year. The loss of that feature is a major hit on t

    • Given the idiotic changes to Google and YouTube that have broken just about everything including the "Back" button, it doesn't surprise me that the next step is to break hypertext links.

      Funny how it was just a few years ago the world was screaming bloody murder over standards compliance.

  • I'm not sure it's related, but I noticed over the years that Google's topic-sensitive ads grow less and less prominent compared to regular search results such that it's harder to tell the difference.

    The placement and color of ads has grown closer to the regular results, such as the fading of the background color of the ads to almost white. I have interpreted this as increasing corporate slimeballery on Google's part, but welcome alternative interpretations.

    And isn't this an accessibility issue, per ADA "Sec

    • Google has ads? It is 2016. Why are you still seeing ads?

      • Re: (Score:1)

        by Tablizer ( 95088 )

        I rest my case.

    • Re: (Score:2)

      by DarkOx ( 621550 )

      Yup one step above plastering fake "download" links all over.

    • Re: (Score:2)

      by Lehk228 ( 705449 )
      Section 508 covers government agencies, google isn't (yet) a government agency

      • Re: (Score:1)

        by Tablizer ( 95088 )

        Okay, you are probably right, but the Americans with Disabilities Act extends similar requirements to private businesses I believe.

  • they decided to test 41 shades between each blue to see which users preferred. In the test, Google showed each shade to one per cent of its users, and found that users were more likely to click on a slightly more purple shade.

    Way to give a meaning to noise.

    • Way to assume that they didn't have sufficient sample size.

    • stupid idea. why?

      how color accurate are pc, mac and tablet display colors?

      answer: TOTALLY NOT THE SAME. even on the same model.

      so many different variables. even if you look at an angle, unless its an SIPS style display, the color will change.

      way to make something out of pure noise, google.

      (laughs)

      google - nerds who think they are smart, but really just *think* they are. to the rest of us, you're just youngsters who were overly ego stroked.

      • Why would any of those deviations matter if one distinct color was in fact statistically significant in what people liked? So people preferred that one color across a range of devices and deviations. If it's statistically significant, why would Google care about the deviations?

      • Re: (Score:3)

        by plover ( 150551 )

        how color accurate are pc, mac and tablet display colors?

        answer: TOTALLY NOT THE SAME. even on the same model.

        Which is why you do field testing.

        It doesn't matter now recently you calibrated your reference monitor, the resolution you display them at, or how precise your systems are in the lab. It doesn't matter if Google put in 0x0000FF and Joe Sixpack's crappy old VGA CRT displayed red text. By putting the changes out in front of millions of users, they got the average of all the devices and all the users. The color each specific monitor displays isn't important - what the entire collection of users responded to m

        • Re: (Score:2)

          by pz ( 113803 )

          And that apparently well-designed test earned $200,000,000 in extra money, with a WAG cost of under $100,000. Holy cow.

    • Re:Cat got your tongue? (Score:5, Informative)

      by Sowelu ( 713889 ) on Monday May 09, 2016 @04:05PM (#52078457)

      Each one had a sample size of 1% of their user base, which is probably a larger sample size than all human-research studies published in a given year combined. I'm betting that when you plot colors on an axis against user behavior, it makes a nice curve too.

    • they decided to test 41 shades between each blue to see which users preferred ...

      Way to give a meaning to noise.

      Yeah, I'm pretty sure differences between one monitor or screen and the next, will totally overshadow the majority of the differences among those '41 shades'. Hell, on my desktop monitor I can select among four wildly different colour, contrast, and brightness profiles with the touch of a button.

  • I actually saw them in a private browsing session and was confused why every link was 'read' (in private browsing). Was even more of a head scratcher when searches in normal mode were blue...

  • This just proves... (Score:3)

    by Junta ( 36770 ) on Monday May 09, 2016 @04:03PM (#52078431)

    That black links matter.

  • Even though I had google as my default search engine in my browser, google kept on popping up a window on my searches telling me how to enable google as my default search engine.

    .
    It got to the point, I started to wonder about google's deteriorating quality. So I went looking for another search engine.

  • First of all I'm having a problem with the way this is being approached.

    It seems intuitive that if you have discovered that you can make 200 million dollars just by changing shades of blue, you would have already reached a point where you knew that blue was going to be *the* color.

    This testing of black now seems to be saying that Google hasn't even determined which color is best, much less which shade of which color. This is assuming that the cites about the two hundred million dollars and the rest of the a

    • Re: (Score:2)

      by plover ( 150551 )

      Interesting that 2 of the first 3 search results for black returned hits on the Black Panthers.

      Well, the title does say 'radical change'.

    • I'm sure that they determinED blue WAS the best choice. Actually they just confirmed what Yahoo and others had already proven. When they did that, 94% of users were using Windows with a CRT and mouse.

      Ten or fifteen years later, most users aren't using Windows, a CRT, or a mouse. They're using Android, a 4" LED screen, and a finger. To me it makes sense to go back and double-check UI choices that were good before and see if they are still best.

  • I've gotten the black links on one of my machines.
    Functionality aside, they're just ugly.

  • To assert that Google "earned additional revenue" from this change, you also need to demonstrate that the addition revenue measured did not come from revenue they would have received regardless, had they waited longer.

    By the similar kind of "logic", it wouldn't surprise me that one could justify using cocaine to treat ADHD in children. (They really did pay more attention for the duration of the comparative study.)

    In any case, what seems clear enough is that this comparative study shook loose a significant

  • Sad--when I saw the headline I was hoping that Google would stop with the blindingly white background on everything and provide a dark interface. At least for now I can use Stylish to make Google dark, but there's way too many blindingly white web sites and apps out there and if Google were to change, it might encourage some others to do so as well. Twenty years from now I'm sure Slashdot will be covering studies showing that this generation's eyes have been destroyed by all the excessively bright white UIs
    • Maddox, [thebestpag...iverse.net] is that you? ["...I've chosen a black background for most of my text because it's easier on the eyes than staring at a white screen. Think about it: your monitor is not a piece of paper, no matter how hard you try to make it one. Staring at a white background while you read is like staring at a light bulb (don't believe me? Try turning off the lights next time you use a word processor). Would you stare at a light bulb for hours at a time? Not if you want to keep your vision."]

  • Got my hopes up, I thought maybe they were testing a black background instead of white. I have been using a combination of the Stylish Add-on and the Dark Fusion style from userstyles.org to achieve this for quite some time now. But, it would be nice to be able to set this right inside of Google's Options.
    https://userstyles.org/styles/... [userstyles.org]

  • I see a blue link and I want it painted black...

  • > test 41 shades between each blue to see which users preferred
    > found that users were more likely to click on a slightly more purple shade

    Note the word choice- we're told that users "prefer" this shade of blue. They didn't get this by asking them, or doing a study about what is most pleasing to look at. Their metric? Clicks. They measure everything by clicks.

    Pretend I have a magical color- we'll use Octarine- that ups the click rate to 100%. Wow! Users must really "prefer" that color, right?

    • When I got to "They measure everything in clicks" I suddenly realized if instead they strobed and cause seizures they would click every link.

  • ... it was a metaphor. You know, like "dark web" or some such thing... "black search results"... But no, the headline is saying literally what Google was doing: making the links black.

    I'm not sure I see any added value to this, and I think some strong arguments can be made that it is a bad idea.

  • " In the test, Google showed each shade to one per cent of its users, and found that users were more likely to click on a slightly more purple shade. "
    This seems like a classic example of how to misuse statistics.
    I'm 100% more likely to click on a link that's relevant to me, I don't give a flying fuck about what color the link is. And neither does anyone else.

    Make the links Black, who gives a shit.
    But when you do that, you better dam well make previously visited sites a different color so that we don't los

  • This is simple, and happens in almost all organizations. They have a new group manager who has to show some achievement. Oh! Look, I have revolutionized google search!!! Now all links are black." WOW. Talk about innovation.
  • I've been using DDG, and it has the superior black font.

  • I'm really hoping someone at The New Slashdot (nice having you aboard whipslash!) will un-fuck the link to comments so they're:

    a) below TFS

    b) not always black, so I can determine if I've visited the particular comment thread yet.

    Those changes were certainly backwards-facing.

  • I hated it so much I stopped using Google and moved to duckduckgo. I hope they record the fact I stopped using Google after using it all the time for years and change me back.

  • I remember when Gmail was blue and didn't strain my eyes... Can I have that back please? And don't touch the effing links. Thanks!

  • ....and I want it painted black.

Slashdot Top Deals

It seems that more and more mathematicians are using a new, high level language named "research student".

Close