User Agent Changes

  • The responses suggest that there's going to be a whole new can of worms from this. Wonderful.

    Also, let's not overlook that the others may jump on board, too. Most notable would be Firefox, especially considering how hard they've been hitting the privacy/security angle in recent years:

    "Apple (Safari), Microsoft (Edge), and Mozilla (Firefox) have also expressed support for Google's proposal to freeze and phase out the user-agent string, but have not announced detailed plans at the time of writing."

    BTW, I wonder, did Vivaldi have a heads-up about Google's decision, or was dropping Vivaldi from the UA so recently just a coincidence? If they did know, it certainly would have been yet another good reason to drop it, since UAs period wouldn't be long for this world (at least as we know them today).

  • As I ponder the proposed changes, it looks increasingly like having the present, detailed user 'agent string' in the initial user query packets will merely be replaced by a more extensive user data 'table' (ie Client Hints) made available soon thereafter in the connection. What a site does with the data will still be entirely up to the site and how it writes its site code... hence, no real improvement will necessarily result for the user-agent issues currently being faced by browsers and users.

  • @Blackbird said in User Agent Changes:

    If they use it to restrict site access (as some now do with user agent strings), then we will be right back to where we presently are.

    Yes. Browsers are still going to have to lie and say they're Chrome.

  • On the HTML5 Test Page Vivaldi identifies as Chrome 79 and scores 535 points. If I use developer tools to identify as Vivaldi the score is on 523. Am I using the right ID, and which result is accurate? Note that Vivaldi is not listed, and one has to type in a custom ID.

    The lower score for Vivaldi is due to less support in the User Interaction section.

  • @Pesala Yes, at the bottom of the HTML5 Test Page, they note: "The HTML5 test does not try to test all of the new features offered by HTML5, nor does it try to test the functionality of each feature it does detect." IOW, they're apparently profiling parts of Vivaldi as they believe it would behave from some database they have and reporting that result in their 'test' tables. A nice example of the User Agent String being used to shape a website response.

  • To repeat what @escape75 said, What is the proper way now for extension developers to detect that they are running on Vivaldi in order to fix issues ?

    For example, an extension from the Chrome store might override the new tab page, but since Chrome uses "chrome_url_overrides" it will use that, however Vivaldi doesn't support that so it needs to revert back to "chrome.tabs.onCreated" but in order to do that we need to know we're running on the Vivaldi browser ...

    What is the correct procedure to detect the browser ?

    Websites may not need to know which browser is being used, but extension developers do.

  • this breaks websites that display additional information alongside the push subscription popup as Vivaldi aligns that popup differently than other browsers. 😕

  • Moderator

    @Holly That's bad design for information display overall. Those sites should just display the info about why they need a permission before asking it.

    By the way, latest Vivaldi snapshots have a flag


    that makes the user agent even more private 😉

  • Moderator

    @madiso Thanks for hint.

    That gives websites a UA of a older Chromium:
    Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/75.0.3764.0 Safari/537.36

    that makes the user agent even more private

    Why do you think this older Chrome UA is more private? I do not understand why.

  • Moderator


    • Version number won't change
    • OS type and version won't be included (frozen to Windows)
    • On mobile the device model is not included, Android is frozen to 9
    • On Vivaldi-whitelisted sites this overrides the default behaviour (meaning those sites won't know about Vivaldi either)

    In few versions, this will be the default in all Chromium browsers.

  • Glad you added it to and glad you wrote this article! I'm a developer for and I was pulling my hair out why "Vivaldi" was in navigator.userAgent on but not on other websites. Now I know it's because you made an exception for us. Many thanks!

  • @druss The domain is for sale if you want to buy it 😉

Log in to reply

Looks like your connection to Vivaldi Forum was lost, please wait while we try to reconnect.