While this might be useful for a casual user, hidden URLs are a huge problem for web developers. Asking for a screenshot from client is not enough, now I'll have to provide additional instructions how to copy/paste full URL when reporting issues.
Not to mention all possible problems with misconfigured servers when www and www-less domains lead to same website, but some script refuses to work on one of those. While it was easy to spot with just a glance at URL bar, now one has to do additional clicking or opening devtools.
Probably we are at the crossroads where developers need separate, not dumbed-down version of browser.
I'm already annoyed by the hassle it has become to copy a substring of the URL into the clipboard, due to the schemes being hidden. Nothing has been won by hiding http(s):// as well as the www subdomain.
I had a frustrating experience with this only yesterday.
I was trying to search for the word "aquarium" but chrome kept filling in "https://aquarium.org", I would delete the ".org" and only the word "aquarium" was shown in the search bar, which was the word I was trying to search.
Of couse "https://" was hidden, so I was actually submitting "https://aquarium" which was not a valid domain and it took many frustrated clicks and enters to actually google the word that was shown. Absolutely infuriating as the true state of the search bar was hidden.
Even when the scheme is visible I have problems copying a URL substring. Chrome always wants to select the ENTIRE URL instead of just the subpath I'm double-clicking... endlessly frustrating >:(
Not for serious web development (dito for Safari): the developer tools are atrocious compared to Chrome, the battery life experience is way worse with FF, and while Chrome is a notorious CPU and RAM hog, FF is worse.
That combined with both FF and especially Safari being way behind in terms of standards adoption/development really sucks at the moment. Chrome desperately needs competition.
Google does have an amazing access to data on how 80 - 90% of users are using the Internet, for many Google is the entry point for their Internet experience. Maybe their data is telling them that the URL bar is basically unused?
That would be nice. They should share that data to help others understand the decision they are making. Or at least they could reference the data in their decision making.
there is a trick for at least chrome before 85, where if you install the google-made extension "suspicious site reporter" it will show the full url including protocol (which you can't even do with flags so it tampers with something internally which means they don't have to do this at all)
Developers at least the experienced ones never left firefox. For the new developers who got hooked on chrome over the last 10/15 years time to move to a developer friendly browser.
This is just simply inaccurate. I have 25 years of web dev experience and left Firefox because Chrome's dev tools were far, far superior to those of Firefox.
Blink is the most common browser engine. Firefox has some nice developer tools but if you don't test in blink throughout the day then you're just asking for problems.
Perhaps Mozilla should never have made breaking changes that pushed people away? The UI change that killed my extensions made me look elsewhere. Chrome's much faster js engine sealed the deal.
I've looked into switching back to Firefox, but what I've found is that they don't allow me to use my own extensions. I would have to use a beta version of Firefox or submit all of my extensions that only I use for approval to Mozilla or have to reinstall my extensions every time I close Firefox. None of these seem good options to me.
If you're relying on bug reports to find why a page is broken you're in for a bad time because 99% of the time the user isn't going to report anything. They're just going to think your page is broken and stop using it. Use a telemetry and error reporting service like Sentry or Rollbar. These services can strip sensitive data on the client before it gets logged.
Not every bug results in an error being thrown or any other signal that you could automatically detect. From my experience most of them are way more subtle.
While parent’s point is for devs, any kind of support situation falls into the same issues.
Grand-parents/friends/org users not being sure to be on the right site after a redesign, not seeing amazon in the right language, etc. There’s countless of questions that can be solved faster by looking at the URL.
How can I tell you don't work in IT? Almost all companies including mine have ONE site to choose from to do any one thing. If it doesn't work, they either ask their colleagues (which only works if it's not the first time someone is using this around them) or create a ticket for IT.
I actually wouldn't mind if the URL bar was replaced with a breadcrumb bar on some sites, like news and forums. Imagine something like
Example.com > Worldnews > 2020 > 06 > 14 > Big aquatic monster spotted outside of Tokyo
or
forum.example.com > Sport > Football > Spain > Real Madrid
It could then work like in Explorer in Windows 10, where you can press one of the breadcrumb separators and see a menu with siblings, or go straight to all news this month. It could use some manifest file in a standard format on the server for the directory information.
Of course, this should never replace the URL completely, you should always be able to get to it easily. But URLs aren't necessarily always the best solution for navigation. We tokenize code and apply different colors, mouse over pop-ups, and links, why should the URL bar be a long raw text string when it's really contains structured data?
This Google nonses of hiding everything except the domain is not a good solution IMO, it doesn't solve a problem and makes it harder to navigate, not easier.
I really dislike any attempt to modify strings like this. I find it invariably causes problems in edge cases. What if a site handles slashes differently to how Google expects? Where do GET arguments go? What if I want to modify the URL? Breadcrumbs are great when each part is navigable, but does example.com/worldnews/2020/06 actually lead anywhere, or is it an invalid address for the site? I have absolutely no interest in Google being allowed to dictate what should and should not be a valid address.
Probably worse than the change itself, though, is the tendency of anyone who makes such a change to start playing fast and loose with actually representing the underlying address. You mention Windows 10's address bar - it's one of the worst offenders. My Windows Explorer is currently sitting in my downloads folder, which is at "C:\Users\Wyatt\Downloads". The address bar reads "This PC > Downloads". When I click on the address bar to edit the address, it changes to just "Downloads". What part of all of this is in any way useful to me or the likely action I'm trying to take when I click on the address bar?
"This PC > Downloads" may point to the same directory as "C:\Users\Wyatt\Downloads", but Explorer may also handle or display differently or with different options. I've had various issues with this, such as not being able to copy the full actual path from the address bar, a sub-folder in one of these "This PC" folders or libraries showing no columns with no option to show them, and sometimes being indistinguishable from the Public folder. The full path matters in Explorer, Finder, and browsers, and should never be hidden without an easy visible way to show the full path or have it always show.
> I really dislike any attempt to modify strings like this. I find it invariably causes problems in edge cases. What if a site handles slashes differently to how Google expects?
I think it would have to be some standard format that websites use, not just string manipulation in the browsers. And certainly not some Google dictated feature! For the same reason, each part would have to be navigable on these sites, to work as I described. There's various possible solutions, like meta tags or some manifest like breadcrumbs.jsonld mentioned in another comment.
The fact that Windows Explorer doesn't show the full URL in special folders is a separate issue, I only mentioned it for the breadcrumbs example.
However, I think there is something to this idea - a breadcrumb style approach by default in Chrome would encourage developers to use paths in more standard ways that refer to resources, not heavy parameter coupling. As you noted, there are technical barriers to implementing this solution, which might encourage some other good things - servers providing resource discovery so that the browser can understand valid paths when visiting a site.
I find the way Explorer in Windows 10 handles this behavior to be annoying and inconvenient. It finds ways to change paths into new canonical locations, for example browse to C:\Users\Yourname and instead of giving you breadcrumbs like Local Disk > Users > Yourname, it simply shows "Yourname" as a special home folder. When you click back in the address bar, there are no breadcrumbs anymore, it's erased your trail.
Attempts to make things simpler by hiding the truth about where you really are in navigation seems like a way to make the web less discoverable except by Google. If you're on a web site you can usually learn more about its structure based on URL format. This makes that more difficult.
But there's nothing stopping websites from offering that without browser support. It can just show that at the top of the page. Everything it provides is under the authority of the website. The URL needs to be provided by the browser because it's not entirely under the authority of the website, but that's not the case for a breadcrumb bar.
> I actually wouldn't mind if the URL bar was replaced with a breadcrumb bar on some sites ...
Which sites?
Anyway, almost everyone else would mind. Especially if there was no option to revert to normal behaviour.
> It could then work like in Explorer in Windows 10, ...
That sounds like the worst of both worlds. If people want Explorer in Windows 10 behaviour - can't they just run Explorer in Windows 10?
If people want Chrome as it was yesterday, they've basically got no option now.
> But URLs aren't necessarily always the best solution for navigation.
The Chromium devs demonstrated their lack of interest in being able to navigate via URL / location bar a half decade ago when they changed the default on all operating systems to be single-click in location bar to 'select the whole address'.
For myself (and I'd wager most people), I want to clear the URL and go to a totally different URL much more often than I want to manually manipulate the URL I'm currently on, so I like the change in default. Many casual users probably didn't even know a quick way to select the whole URL when it wasn't the default.
There were several extensions that transformed location bar like you propose, but naturally just did simple domain, path and query segments transformation into clickable buttons producing breadcrumbs. It was a delight to use and I miss them in Firefox since Quantum leap prevented them to work any longer.
I’ve implemented this on my site. The pain in the ass is that Google will only sometimes show you the breadcrumbs, so it’s very difficult to tell if you encoded it correctly.
I think that's a great example of the benefits though. HN could continue to use their fairly opaque URLs in the background, but instead show something like
news.ycombinator.com > 2020 > 06 > 14 > Google hides full addresses in URL bar on Chrome 85
This makes it easy to not only see where you are, but also quickly click on a part of the address to go to that hierarchy, or a sibling like yesterdays posts. It makes sense that a forum like this would have a way too see all posts from a day, month, or year.
Of course, most if not all users here are comfortable with URLs so they're probably not the ones that would benefit the most. But I think most common users, the ones who Google everything instead of typing in an address, would use the breadcrumb bar while today they probably see the URL as some weird text string they have little interest in or understanding of.
While we often create a mental map between some sort of logical hierarchy and the segments of the url this doesn't have to be the case. A specific domain should be authority on this, not he general-purpose tool used to access it.
At chrome://flags there is one called #omnibox-context-menu-show-full-urls, which I have turned on.
This enables you to right click on the address bar, and turn on the option "Always show full URLs". It will always shows the full URL including the protocol, but I suspect they will remove this flag at some point.
If you happen to close your browser and lose your tabs, use the reopen closed tab menu option, it'll bring all the closed tabs (even if multiple tabs).
I can't find this option on Chrome on Linux. I had to get an extension to show the full URL, but it only works for 'https://' URLs, not for 'http://'.
This drives me crazy when debugging. Whenever I copy-paste IP addresses from the browser address bar into my console, I have to manually delete the `http://` at the front. I work on a P2P project so this is an extremely common situation for me.
Are you running version 83 or later? I think they introduced the flag in that one.
Another solution for Windows and macOS users (no Linux sadly) is to use Edge Chromium, which does it by default, and you prefer to donate your data to Microsoft rather than Google like me :)
This is exactly what I'm afraid of. there used to be chrome://flags/#omnibox-ui-hide-steady-state-url-scheme-and-subdomains (when I google how to make chrome show the full url bar, this is the recommended answer) but it's been gone entirely for several Chrome versions. I even toggled a flag to undo flag deprecations in Chrome 78 to get this back but that didn't work very long - I think this flag has been totally dead since Chrome 80ish.
I personally don't care much what the default is for the normal user, but I want to be able to have my full urls.
OK, I just don’t get it anymore. I mean I’m a happy Firefox user, so it’s not like this personally impacts me, but how in the heck is seemingly nobody acknowledging that this has been the behavior in Safari for a long time now? This seems to be a recurring pattern.
I've got a couple of guesses. Apple is somehow regarded as being a pro-user company, not having plans of taking over the web. Also, Apple users are accustomed to UI changes that result in visual simplicity. Nevermind that their actions result in patronizing the user just the same as Google does, in their case those actions are more likely to be perceived as innocuous.
Cmd+f Safari, I'm equally surprised no one but you mentions it.
However: I believe Apple's motives are aligned with their users and they want their browser to be as safe and as easy to understand/use as possible. Their primary intention is to sell their shiny expensive hardware.
With Google it's more controversial, because who knows what's the plan. Combined with AMP there is a reason to be wary.
Ofc, one can make bad decisions based on good motives.
The other day, our 7 year old told me that [ is 5B and ] is 5D. I was quite impressed that he knew this, and I asked him how he knew it. He told me it was from reading the address bar in Roblox. Needlessly hiding technical details from kids is going to limit their learning.
This, exactly. People learn not only when they are forced to, but naturally from observing their environment too. The more opportunities to "spontaneously learn" you take away from them, the less they will learn.
Here's a comment I made from several years ago when Chrome tried to do before what it's trying again now (it's not the first time): https://news.ycombinator.com/item?id=7678729
Maybe this next point is starting to go into the realm of conspiracy theory, but I see far too much evidence of it every day: companies are doing this because they don't want users to learn. They want to keep users naive, docile, and compliant, and thus easier to "herd" for their purposes. They don't want people knowing the truth behind how things work; they would rather "developers" (and only those who explicitly chose to be one --- probably for monetary reasons) learn from their officially sanctioned documentation (which does not tell the whole truth), and not think or discover for themselves.
(I've memorised most of printable ASCII because I did a lot of Asm programming decades ago, so I instantly understood what you mean.)
Not sure how much actual conspiracy is in there, but I've definitly noticed that the gap between "consumer software" (highly optimized for ease of use but also highly limited and designed with a specific intention how users should interact with it) and "professional software" (powerful and flexible but only usable after extensive training, often command-line only) is widening instead of closing.
There are also definitly conscious design descisions how "cryptic" a particular feature should appear to users. I remember several bugzilla threads with discussion whether a config option should be exposed as an "ordinary" field in the settings or only as an option in about:config, so that normal users won't find it.
I think the value is not in memorizing such trivia; for a 7 years old it might be discovering the pattern of data encoding and its why and how. It opens all sort of paths of discovery in future for understanding software.
You'd be surprised how often veteran developers fail to grasp intermediate Unicode concepts (surrogate pairs, for instance) probably since they skipped over (or was not curious enough about) implementation details of such abstractions.
Sure, but it's the curiosity about how things work under the hood that matters.
If he sees [ being replaced with %5B,he'll ask why it does that. And that leads to learning.
When he told me this, I thought he was talking about ASCII, but I had to actually double-check with "man ascii", because 5B/5D sounded familiar but wasn't 100% sure he was right.
But the point is not that he memorises the ASCII table. The value is that he learns that computers internally represent letters/punctuation as numbers. The underlying concept is what's important, and the learning of specific values is mainly useful as a way of learning and reinforcing that underlying concept.
Firefox became quite fast again after Quantum. For those of us who never "bought into" the whole Chrome ecosystem, there's always been adequate alternatives. Will check out: https://www.palemoon.org/
Safari has been this way since 2014. I've never seen any pushback on Apple doing it over the past six years.
It's genuinely a benefit for the vast, vast majority of users, where the only important piece of information really is the domain name, to check which site you're actually on. And for more info, you can just click. Copying the URL becomes no more difficult.
The URL path beyond the domain is as useful to most people as an IP address, in other words not at all -- it's just noise. And displaying noise is bad UX. Pretty much only website developers and administrators and SEO people care about the full URL. Granted, there are a lot of those people here on HN, so I understand the pushback, but we're not most users.
But at the end of the day, I don't understand why people seem totally fine with Safari doing this, but not Google?
As long as you see the full url when you hover/click on the bar, I am all for it as well.
If find some of reactions on this ridiculously hyperbolic "biggest attack on the web in years" ? seriously ?
I get it, Google is a gigantic monster that does not necessarily act in its users best interests, but that does not mean we need to bring the pitchfork each time they launch an app update.
On that note, I noticed recently that Google search result links (on Firefox?) get rewritten. That is, you see the actual page URL when you hover over the link, but it's changed to their own redirect URL as soon as you click it.
I'm sure they've always been tracking these search result clicks, but I think this is a somewhat new behavior, and I find it highly deceiving.
I’m a developer. I use safari/WebKit for 95% of my browsing. You can enable the full address among a bunch of other excellent developer settings and move on with life with a browser that works great.
I strongly disagree. Ordinary users I interact with either understand the basic concept of the URL or understand it after an initial explanation. It becomes empowering to them in ways I often do not anticipate.
But why can't this be a toggle or user setting like in Safari? Why is it a 1-true-google-way of doing things when clearly there are users who want to keep it (even if its just web developers)?
What makes you think that once it becomes default, the switch won't remain to be able to turn it off?
Chrome is built by developers. Presumably, they pay attention to what developers need from it. Which is why their debugging tools overall are so amazing.
Perhaps you should withold criticism of what you assume they'll do until they, you know, actually do it.
Readit News
Not to mention all possible problems with misconfigured servers when www and www-less domains lead to same website, but some script refuses to work on one of those. While it was easy to spot with just a glance at URL bar, now one has to do additional clicking or opening devtools.
Probably we are at the crossroads where developers need separate, not dumbed-down version of browser.
I was trying to search for the word "aquarium" but chrome kept filling in "https://aquarium.org", I would delete the ".org" and only the word "aquarium" was shown in the search bar, which was the word I was trying to search.
Of couse "https://" was hidden, so I was actually submitting "https://aquarium" which was not a valid domain and it took many frustrated clicks and enters to actually google the word that was shown. Absolutely infuriating as the true state of the search bar was hidden.
google could make an amp-only web experience without dissent.
hide the URL bar
javascript -> webasm
hiding all this benefits data collection and advertising. Seems obvious to me.
Yes, we are.
Funny enough, Firefox has a 'Developer edition', but that's just the Beta build, with some features turned-on by default.
https://www.mozilla.org/en-US/firefox/developer/
Due to slow loading, I'm considering whatever they call the Microsoft browser today.
That combined with both FF and especially Safari being way behind in terms of standards adoption/development really sucks at the moment. Chrome desperately needs competition.
Can anyone reproduce this?
At this point ie becomes more useful.
I've looked into switching back to Firefox, but what I've found is that they don't allow me to use my own extensions. I would have to use a beta version of Firefox or submit all of my extensions that only I use for approval to Mozilla or have to reinstall my extensions every time I close Firefox. None of these seem good options to me.
Deleted Comment
Grand-parents/friends/org users not being sure to be on the right site after a redesign, not seeing amazon in the right language, etc. There’s countless of questions that can be solved faster by looking at the URL.
Example.com > Worldnews > 2020 > 06 > 14 > Big aquatic monster spotted outside of Tokyo
or
forum.example.com > Sport > Football > Spain > Real Madrid
It could then work like in Explorer in Windows 10, where you can press one of the breadcrumb separators and see a menu with siblings, or go straight to all news this month. It could use some manifest file in a standard format on the server for the directory information.
Of course, this should never replace the URL completely, you should always be able to get to it easily. But URLs aren't necessarily always the best solution for navigation. We tokenize code and apply different colors, mouse over pop-ups, and links, why should the URL bar be a long raw text string when it's really contains structured data?
This Google nonses of hiding everything except the domain is not a good solution IMO, it doesn't solve a problem and makes it harder to navigate, not easier.
Probably worse than the change itself, though, is the tendency of anyone who makes such a change to start playing fast and loose with actually representing the underlying address. You mention Windows 10's address bar - it's one of the worst offenders. My Windows Explorer is currently sitting in my downloads folder, which is at "C:\Users\Wyatt\Downloads". The address bar reads "This PC > Downloads". When I click on the address bar to edit the address, it changes to just "Downloads". What part of all of this is in any way useful to me or the likely action I'm trying to take when I click on the address bar?
I think it would have to be some standard format that websites use, not just string manipulation in the browsers. And certainly not some Google dictated feature! For the same reason, each part would have to be navigable on these sites, to work as I described. There's various possible solutions, like meta tags or some manifest like breadcrumbs.jsonld mentioned in another comment.
The fact that Windows Explorer doesn't show the full URL in special folders is a separate issue, I only mentioned it for the breadcrumbs example.
However, I think there is something to this idea - a breadcrumb style approach by default in Chrome would encourage developers to use paths in more standard ways that refer to resources, not heavy parameter coupling. As you noted, there are technical barriers to implementing this solution, which might encourage some other good things - servers providing resource discovery so that the browser can understand valid paths when visiting a site.
Attempts to make things simpler by hiding the truth about where you really are in navigation seems like a way to make the web less discoverable except by Google. If you're on a web site you can usually learn more about its structure based on URL format. This makes that more difficult.
The only difference between
and is a little bit of reformatting and upcasing and linkifying (or otherwise making selectable) the individual path segments of the URL.And probably some clever logic to deal with the randomforum.php?fid=12345&tpcid=984.3&page=5 goop that is still all-too-common... :/
You say that as though websites like that are random small sites. HN has that kind of a URL, so do YouTube and Google.
Which sites?
Anyway, almost everyone else would mind. Especially if there was no option to revert to normal behaviour.
> It could then work like in Explorer in Windows 10, ...
That sounds like the worst of both worlds. If people want Explorer in Windows 10 behaviour - can't they just run Explorer in Windows 10?
If people want Chrome as it was yesterday, they've basically got no option now.
> But URLs aren't necessarily always the best solution for navigation.
The Chromium devs demonstrated their lack of interest in being able to navigate via URL / location bar a half decade ago when they changed the default on all operating systems to be single-click in location bar to 'select the whole address'.
I'm beginning to think they are not our friends.
https://www.ghacks.net/2011/03/01/improve-firefoxs-urlbar-wi...
They by default worked very similarly the aforementioned Windows Explorer, what in focused state with keyboard input turns into "raw" text field.
https://developers.google.com/search/docs/data-types/breadcr...
Sadly not used everywhere, but maybe browser support would encourage its usage by site owners.
news.ycombinator.com > 2020 > 06 > 14 > Google hides full addresses in URL bar on Chrome 85
This makes it easy to not only see where you are, but also quickly click on a part of the address to go to that hierarchy, or a sibling like yesterdays posts. It makes sense that a forum like this would have a way too see all posts from a day, month, or year.
Of course, most if not all users here are comfortable with URLs so they're probably not the ones that would benefit the most. But I think most common users, the ones who Google everything instead of typing in an address, would use the breadcrumb bar while today they probably see the URL as some weird text string they have little interest in or understanding of.
Dead Comment
This enables you to right click on the address bar, and turn on the option "Always show full URLs". It will always shows the full URL including the protocol, but I suspect they will remove this flag at some point.
Now how does this new flag interact? Has anyone enabled both to see?
1: https://bugs.chromium.org/p/chromium/issues/detail?id=106157...
2: https://bugs.chromium.org/p/chromium/issues/detail?id=883038...
This drives me crazy when debugging. Whenever I copy-paste IP addresses from the browser address bar into my console, I have to manually delete the `http://` at the front. I work on a P2P project so this is an extremely common situation for me.
Another solution for Windows and macOS users (no Linux sadly) is to use Edge Chromium, which does it by default, and you prefer to donate your data to Microsoft rather than Google like me :)
Seriously, I keep looking at the address bar to make sure the URL is still there and I'm not dreaming.
I personally don't care much what the default is for the normal user, but I want to be able to have my full urls.
Is there also a #upgrade-to-firefox-immediately flag ?
However: I believe Apple's motives are aligned with their users and they want their browser to be as safe and as easy to understand/use as possible. Their primary intention is to sell their shiny expensive hardware.
With Google it's more controversial, because who knows what's the plan. Combined with AMP there is a reason to be wary.
Ofc, one can make bad decisions based on good motives.
Here's a comment I made from several years ago when Chrome tried to do before what it's trying again now (it's not the first time): https://news.ycombinator.com/item?id=7678729
Maybe this next point is starting to go into the realm of conspiracy theory, but I see far too much evidence of it every day: companies are doing this because they don't want users to learn. They want to keep users naive, docile, and compliant, and thus easier to "herd" for their purposes. They don't want people knowing the truth behind how things work; they would rather "developers" (and only those who explicitly chose to be one --- probably for monetary reasons) learn from their officially sanctioned documentation (which does not tell the whole truth), and not think or discover for themselves.
(I've memorised most of printable ASCII because I did a lot of Asm programming decades ago, so I instantly understood what you mean.)
There are also definitly conscious design descisions how "cryptic" a particular feature should appear to users. I remember several bugzilla threads with discussion whether a config option should be exposed as an "ordinary" field in the settings or only as an option in about:config, so that normal users won't find it.
You'd be surprised how often veteran developers fail to grasp intermediate Unicode concepts (surrogate pairs, for instance) probably since they skipped over (or was not curious enough about) implementation details of such abstractions.
But the point is not that he memorises the ASCII table. The value is that he learns that computers internally represent letters/punctuation as numbers. The underlying concept is what's important, and the learning of specific values is mainly useful as a way of learning and reinforcing that underlying concept.
I'd stay clear of that project and use mainstream Firefox instead. And afaik they still don't support WebExtensions.
Suppose you always had to tell people to 'Google it'
Suppose 'I feel lucky' was always the default, and the result was sold to the highest bidder.
Deleted Comment
It's genuinely a benefit for the vast, vast majority of users, where the only important piece of information really is the domain name, to check which site you're actually on. And for more info, you can just click. Copying the URL becomes no more difficult.
The URL path beyond the domain is as useful to most people as an IP address, in other words not at all -- it's just noise. And displaying noise is bad UX. Pretty much only website developers and administrators and SEO people care about the full URL. Granted, there are a lot of those people here on HN, so I understand the pushback, but we're not most users.
But at the end of the day, I don't understand why people seem totally fine with Safari doing this, but not Google?
If find some of reactions on this ridiculously hyperbolic "biggest attack on the web in years" ? seriously ?
I get it, Google is a gigantic monster that does not necessarily act in its users best interests, but that does not mean we need to bring the pitchfork each time they launch an app update.
And also precisely because of AMP, this might be a very dangerous step towards blurring the lines between original and AMP pages.
I'm sure they've always been tracking these search result clicks, but I think this is a somewhat new behavior, and I find it highly deceiving.
Right now it is a setting to enable in Chrome.
What makes you think that once it becomes default, the switch won't remain to be able to turn it off?
Chrome is built by developers. Presumably, they pay attention to what developers need from it. Which is why their debugging tools overall are so amazing.
Perhaps you should withold criticism of what you assume they'll do until they, you know, actually do it.