Live data from Hacker News

Chrome phasing out support for User-Agent

infoq.com

111–120 of 329 posts

Re: Chrome phasing out support for User-Agent

#111
I log this for coarse statistics about what our user base is running, but that is about it.

The good news: IE use is down over the last year to only about 40%.

The bad news: the growth elsewhere is all Chrome, with less than 1% Firefox or Safari. There’s a tiny sprinkling of Edge, as well, but I forget the numbers on that.

Our users are state and county offices and medical facilities, rather than private individuals, so the users are somewhat captive to whatever their organization mandates.

The only browser detection we do is in client side scripting to detect if the browser can directly display a PDF inline (or not, in the case of IE11)

Re: Chrome phasing out support for User-Agent

#112
New proposed syntax adds even more noise:

    User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) 
    AppleWebKit/537.36 (KHTML, like Gecko)
            Chrome/71.1.2222.33 Safari/537.36
    Sec-CH-UA: "Chrome"; v="74"
    Sec-CH-UA-Full-Version: "74.0.3424.124"
    Sec-CH-UA-Platform: "macOS"
    Sec-CH-UA-Arch: "ARM64"
Why not getting rid of the `User-Agent` completely?

It's already bad infrastructure design to have the server do different renderings depending on `User-Agent` value.

Re: Chrome phasing out support for User-Agent

#113
post #51

Earlier quoted context omitted.

I had been thinking recently as I've been using Firefox more that Google maps had got clunky. With a little fiddling prompted by your comment, it turns out Maps sniffs specifically to reduce fluid animations on Firefox (and probably some other browsers).

Thnak you, this had been bugging me for a while. Looks like I'll need to permanently install a UA-switcher extension. Yesterday I saw a HN comment saying you can add the (?|&)disable_polymer=1 parameter to the end of YouTube URLs to make the site much faster - iirc Polymer is extremely slow on Firefox only. This extension was also linked: https://addons.mozilla.org/en-US/firefox/addon/disable-polym... Unfortunately t…

Some advice about ReCaptcha, the audio test is way easier and usually only makes you do it once or twice (as opposed to the 5-10 times you usually have to do when you have disabled tracking). Sometimes it will say you aren't eligible or something, just refresh the page and it will let you try again.

Re: Chrome phasing out support for User-Agent

#114

does this mean there will no longer be a way of determining if the device is primarily touch (basically all of "android", "iphone" and "ipad") or guesstimating screen size ("mobile" is typical for phones in the UA) on the server? https://developer.chrome.com/multidevice/user-agent i wonder what Amazon will do. they serve completely different sites from the same domain after UA-sniffing for mobile. is the web just goi…

Presumably you'll grab the dimensions (could cache after first load) and then render dynamically based on that. If you're doing some sort of if statement on the server to deliver content based on screen size you're probably doing it wrong. Obviously I can't speak for every mobile user, but for myself, it's infuriating to have a completely different set of functionality on mobile.

> If you're doing some sort of if statement on the server to deliver content based on screen size you're probably doing it wrong. Obviously I can't speak for every mobile user, but for myself, it's infuriating to have a completely different set of functionality on mobile.

there's not a "right" and a "wrong" here; it's about trade-offs.

you're either stripping things down to the lowest common denominator (and leaving nothing but empty space on desktop) or you're wasting a ton of mobile bandwidth by serving both versions on initial load (the most critical first impression).

you frequently cannot simply squeeze all desktop functionality from a 1920px+ screen onto a 320px screen - unless you have very little functionality to begin with. Amazon (or any e-commerce/marketplace site) is a great example where client-side responsiveness alone is far from sufficient.

https://www.walmart.com/ does it okay, but you can see how much their desktop site strips down to use the same codebase for desktop and mobile.

Re: Chrome phasing out support for User-Agent

#116

New proposed syntax adds even more noise: User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.1.2222.33 Safari/537.36 Sec-CH-UA: "Chrome"; v="74" Sec-CH-UA-Full-Version: "74.0.3424.124" Sec-CH-UA-Platform: "macOS" Sec-CH-UA-Arch: "ARM64" Why not getting rid of the `User-Agent` completely? It's already bad infrastructure design to have the server do different renderin…

It's great design if you're trying to push Google products.

Re: Chrome phasing out support for User-Agent

#117

does this mean there will no longer be a way of determining if the device is primarily touch (basically all of "android", "iphone" and "ipad") or guesstimating screen size ("mobile" is typical for phones in the UA) on the server? https://developer.chrome.com/multidevice/user-agent i wonder what Amazon will do. they serve completely different sites from the same domain after UA-sniffing for mobile. is the web just goi…

browser feature detection is the way grown up developers have been doing this for several years now. user agent sniffing is dumb because it bundles a ton of assumptions with a high upkeep requirement, all wrapped up in an unreadable regex. It's been bad practice for ages; I'd be surprised if that's how Amazon is doing it still.

Re: Chrome phasing out support for User-Agent

#118
post #59

Earlier quoted context omitted.

Google is probably so big that we might as well consider Chrome and rest of the Google as separate entities.

You can see Chrome devrels on Twitter expressing disappointment with Chrome-only web sites, saying that they raise the issue internally. Of course we have no visibility into what happens after that, but it's an indicator that you're right.

Considering that there's been an internal and external bug filed about the US states not being in alphabetical order on contacts.google.com for years, making it impossible to type 'new y' to get New York, I don't think raising it as an issue will help much.

Re: Chrome phasing out support for User-Agent

#119
post #36

The weird thing about this is that the only company I've seen doing problematic user-agent handling in recent years is Google themselves. They have released several products as Chrome-only, which then turned out to work fine in every other browser if they just pretended to be Chrome through the user agent. Same with their search pages, which on mobile were very bad in every non-Chrome browser purely based on user age…

"Oopsie" said Google to Firefox.

Re: Chrome phasing out support for User-Agent

#120
post #36

The weird thing about this is that the only company I've seen doing problematic user-agent handling in recent years is Google themselves. They have released several products as Chrome-only, which then turned out to work fine in every other browser if they just pretended to be Chrome through the user agent. Same with their search pages, which on mobile were very bad in every non-Chrome browser purely based on user age…

Every single Google product is slower on Firefox and it’s hard to not call this malice and artificial. Many people check out Gmail and GMaps on Firefox and go back to Chrome because of their clunkiness on Firefox.

I try out google products on occasion and find them rubbish compared with alternatives, so go back to non-google products.
Post reply on HN