Google Wants to Kill URLs in Chrome

Posted on September 6, 2018 by Paul Thurrott in Cloud, Google with 33 Comments

With Chrome hitting the ten-year mark this past week, Google is moving aggressively to future-proof its web browser. And in the wake of sweeping user experience changes and a dramatic shift in how Chrome handles insecure websites, Google’s next major target is the URL.

“People have a really hard time understanding URLs,” Chrome engineering manager Adrienne Porter Felt told Wired. “They’re hard to read, it’s hard to know which part of them is supposed to be trusted, and in general I don’t think URLs are working as a good way to convey site identity. So we want to move toward a place where web identity is understandable by everyone: They know who they’re talking to when they’re using a website and they can reason about whether they can trust them. But this will mean big changes in how and when Chrome displays URLs. We want to challenge how URLs should be displayed and question it as we’re figuring out the right way to convey identity.”

This makes sense. Website URLs—Uniform Resource Locators—are a weird vestigial leftover from the overly-techie past, an often complex and lengthy string of characters that users should never really have to deal with directly. They’re particularly problematic on mobile, too, where on-screen real estate is at a premium. Worse, they can be insecure: It’s easy for hackers to change a few characters in a lengthy URL to misdirect users to an insecure site where they can be compromised.

As with all such changes, however, Google is sure to run into roadblocks. Indeed, as Wired notes in its report, this isn’t the first time Google tried to kill the URL: In 2014, it launched a campaign to replace lengthy URLs with an “origin chip” that displayed only a site’s domain name in the address bar until a user clicked on it. But push-back from users scuttled the plans.

Unfortunately, we won’t be seeing Google’s URL replacement anytime soon: The firm is currently investigating how its customers use URLs so that it can enhance security, promote the site’s identity, and enable easy sharing without compromising the experience.

“I don’t know what this will look like, because it’s an active discussion in the team right now,” Chrome director of engineering Parisa Tabriz said. “But I do know that whatever we propose is going to be controversial. That’s one of the challenges with a really old and open and sprawling platform. Change will be controversial whatever form it takes. But it’s important we do something, because everyone is unsatisfied by URLs. They kind of suck.”

Yes. Yes, they do.

 

Tagged with

Join the discussion!

BECOME A THURROTT MEMBER:

Don't have a login but want to join the conversation? Become a Thurrott Premium or Basic User to participate

Register
Comments (34)

34 responses to “Google Wants to Kill URLs in Chrome”

  1. Avatar

    yaddamaster

    Sorry, but this is a solution in search of a problem. The simplest solution is for website developers to simply stop making so many long and ridiculous urls and use friendly name mappings. There should be almost no need for long querystrings.


    For example, let's look at this page: you have your computer and domain. Can't get rid of that although it could be shortened to get rid of the www. Do you really need /cloud/172768/? No, you don't. And do you really need a "friendly" name in "google-wants-to-kill-urls-in-chrome"? No, you don't. "google-urls" would be enough.


    Companies have tried in the past to go back to AOL names - I forget the name of the most recent one but it was around 15 years ago and failed spectacularly. Google is going to fail as well - there's simply too much existing code out there that the ROI doesn't prove out.

    • Avatar

      LT1 Z51

      In reply to yaddamaster:


      You're right, URLs are not the problem, it's the people who use them. I don't get why a lot of websites have the whole word-word-word-word format now. Must be the web platform.

      • Avatar

        idontknow

        In reply to LT1 Z51:

        You both have valid questions. However URLs are supposed to be human readable and have an hierarchical meaning. So e.g. www.thurrott.com/cloud/ would give you all articles about cloud, which it does. If you're reading the article /2018/06/the-article/ then you can easily go to /2018/ to get all articles from 2018. Easy, yes and most actually understand this. Unfortunately not all website owners implement this in a good way.


        Yaddamaster's example of removing www is not always possible. E.g. if your website uses a CDN it's not possible since those only work with CNAMEs, i.e. can't point directly to a webserver's IP address. This does not stop the browser from hiding it though, which I believe some actually do.


        The number part in the URL to this article is just laziness, it's the ID in the database for this article probably. Not needed if all articles have a unique name (or URL) instead, which probably most of them already have.


        Also, when you have a link in an email or in a comment it's nice to be able to guess what the link actually is about without the need to navigate to it. Case in point; yada.com/123123, what do think is on this page!? Compare that to yada.com/cats/black. You see? That's why I hate bit.ly and other URL shortening services, besides the redirection that will take place.


        Microsoft are worst, none of their services have nice URLs. Look at the mess in SharePoint and TFS or Office365 in general. Outlook is still on live.com domain.

      • Avatar

        RonH

        In reply to LT1 Z51:

        Word-Word in the url is long, but easy to read

  2. Avatar

    locust infested orchard inc

    Quote by Paul Thurrott: "Google Wants to Kill URLs in Chrome"


    Locust Infested Orchard Inc Wants to Kill Google's insatiable and compulsive desire for their abusive data thieving practices.

    • Avatar

      Winner

      In reply to locust infested orchard inc:

      Well Locust, you must be fine with your ISP knowing everything you do, your credit card company knows everywhere you shop and everything you buy on charge, your government tracks your license plate all over the place. Oh and your mobile carrier knows everywhere you've been and everything you've one on the internet from your phone. And they do sell that data. And of course Equifax lost your SSN and personal info in a massive data breach.


      Yet Google aggregates your info to give you better targeted ads, with no sales of your personal information. That's how they managed to stay in business and give you free services. And they've not breached your data. And for that minor item you get search, youtube, maps, assistant - all highly valuable services.


      I guess your hatred of Google works for you. Personally I think you're a bit misguided.

  3. Avatar

    curtisspendlove

    If I’m understanding what they want to do, Safari has been going down this path for quite some time.


    Honestly I’m fine with it. For instance, when I look up in the address bar on this page I just see “Thurrott.com” with a lock to the left of the name. If I care, I can click/tap in the address bar and it expands to the full uri and selects it (in case I want to copy/paste, etc).


    And (I think by default now in macOS) if you hover over a link you don’t see the link uri in the bottom toolbar. (You can change this in Safari preferences.) At first it bugged me, but I’m used to it now.


    If if you want to load it, you click it as usual; if you want to copy (or such) you right-click it (or option-click, I think...if you are averse to right-clicks).

  4. Avatar

    ben55124

    How about AOL keywords? /s


    Imagine a world without urls - where accessing websites required a web search. Go google.

  5. Avatar

    christian.hvid

    Didn’t some outfit called RealNames, supported by Microsoft, try to reinvent the URL in the late nineties? It failed completely, and this in a time when browsers were a lot more finicky about URL syntax than they are today, and address bar search wasn’t really a thing yet.


    URLs do not suck, anymore than phone numbers, postal codes or email addresses suck. They’re occasionally difficult to remember, yes, but Google itself is excellent at helping us with just that. This is just Google wishing they could replace the DNS system with something of their own creation.

  6. Avatar

    wright_is

    It looks like they have already dropped WWW. and M. from URLs that are displayed in the current version.

    According to El Reg, "www.domain.com" is displayed as "domain.com" and "whatever.www.domain.com" is displayed as "whatever.domain.com". Google said, that they are "unimportant" parts of the URL.

    I can't confirm this, as I don't use Chrome on any of my machines, currently. But I can see a few negatives with this already.

    If I am setting up a new website and want to see if the www.domain.com to domain.com is working, I won't see this, because Google has decided, I don't need to see it - there are other ways of checking, but simply calling up a page and seeing if the www disappears was the easiest.

    And, how long until some hackers work out a way to exploit this? If Chrome is deliberately not displaying parts of the URL, I can see that there might be some way to trick it into dropping other parts of URLs.

    How long untli they hide the URL bar completely and you have to actually click a down arrow or something to open it up?

  7. Avatar

    Rycott

    Customers already don't use URLs. They already just type things into Google search.


    It's super frustrating because they then ask what result it is and you have to do the search yourself to find out.

  8. Avatar

    longhorn

    URLs are not a problem. Casual users don't need to look at URLs because they are clearly identified as secure if a certificate authority has approved them. A user who is unable to understand the meaning of a green padlock should NOT be using the web. Maybe 0,001 % of the population are too dumb to use the Internet and we shouldn't redesign URLs for them.


    For example in Firefox; thurrott.com is marked secure and if you hover the green padlock you'll see that it has been verified by: COMODO CA Limited. If you click the green padlock and then click More Information you'll see that the certificate expires March 8, 2019. You can also set site permissions etc.


    Google is searching for a problem that doesn't exist. What's the real reason Google?


  9. Avatar

    Jeff Jones

    The "web" aspect of the world wide web was designed around linking to other sites. I wonder how one would share or link to another site if their web browser hides all the URLs?


    I'd hate for forum conversations and other URL sharing tasks to take 2-3 steps longer just to shield people who might get confused easily from extra information.


    • Avatar

      hrlngrv

      In reply to DataMeister:

      Indeed, consider how many people post sharable links to images (Imgur, Postimage), videos (YouTube, Facebook), even files (Google Drive, DropBox, OneDrive). It'd also make a lot of user-to-user support forums a lot more difficult.

  10. Avatar

    RM

    "The firm is currently investigating how its customers use URLs" . . . means they are collecting how you are interacting with the URLs in this field which means they are collecting the URL's also. Hopefully in an anonymous way. URL's do suck, but hopefully this is done in a standardized way across all browsers. Maybe the DNS servers need an Alias option where if you type in google it equates to google.com. Then it will act the same way for all browsers.

  11. Avatar

    skane2600

    "So we want to move toward a place where web identity is understandable by everyone: They know who they’re talking to when they’re using a website and they can reason about whether they can trust them."


    Sounds a bit like Google wants to create a walled garden of websites. Perhaps they could contact Steve Case for some insights.

  12. Avatar

    johnbaxter

    Old saw: Phone tech support person: "Put the computer back in the box, take it to the store, and tell them you're returning it because you're too dumb to own a computer."


    Seriously, I see Google's public-facing point. But...what they really want is more searching to give them even more data. (They found an empty drive in a data center.)

  13. Avatar

    Brockman

    Any reason to think Google won't just shove this down the web's collective throat the same way they have done with AMP? I'm all for faster pages, but Google bypassed the obvious but more challenging answer (get people to stop making crappy, heavy websites) in favor of their own "open" (but privately hosted solution) solution that sites now need to adopt at risk of being pushed into search oblivion. I fully expect they'll come up with something just as self-serving here, pushing us further from a truly open web.

  14. Avatar

    Stooks

    Firefox is my solution to Chrome. If you care about your privacy you would ditch Chrome.


    The next Firefox coming soon (63?) follows Safari and automatically blocks cross sight tracking and auto video play. Two things Chrome should be doing, but it would directly impact Google revenue stream aka sucking down your information to sell it.

  15. Avatar

    Saxwulf

    “People have a really hard time understanding URLs,”


    What "people"? Google engineers?


    Uninformed users will always click dodgy links no matter what the underlying string looks like.


    Do you really think that Google are the right people to be messing with web addresses? I wouldn't trust them to empty my rubbish bins. Sift, scan, sell.


    Yes. yes, they do.

  16. Avatar

    Martin Pelletier

    Don't we do something similar already? Typing something in the bar, the search engine show possible links and you click the one you need?



  17. Avatar

    harrymyhre

    I read once that Tim Berners-Lee didn't want urls exposed.

  18. Avatar

    dcdevito

    I agree with this wholeheartedly, but the "organic" web crowd will hate this, just like they hate AMP.

  19. Avatar

    Chris_Kez

    I hope that whatever they do, they do it as part of a standard approach in concert with the World Wide Web Consortium and other browser makers. I don't want to see another AMP-like splintering and segregation of the web.

  20. Avatar

    Daekar

    URLs do not suck. People are just stupid. This is a non issue.

Leave a Reply