Slashdot Mirror
Chrome 64 Now Trims Messy Links When You Share Them (theverge.com)
Google's latest consumer version of Chrome, version number 64, just started cleaning up messy referral links for you. From a report: Now, when you go to share an item, you'll no longer see a long tracking string after a link, just the primary link itself. This feature now happens automatically when sharing links in Chrome, either by the Share menu or by copying the link and pasting it elsewhere. Even though it slices off the extra bit of the URL, this doesn't affect referral information. If you choose, you can copy and paste directly from the URL bar to grab the link in entirety.
I usually manually trim everything after the ? in the link, but now I won't have to
I have no problem with this feature under a "share" button, but plain-old copy and paste are not a tag-team synonym for sharing in any sane world.
The PC revolution was largely built on determinism at scale: the same operation repeated (on your machine, or the next machine) achieves the same results. This was pretty new in the world in the late seventies. It's why we became able to build more complex distributed systems than ever before; it's how we ultimately carved our way out of spaghetti-code mountain.
Now we take this boon for granted, and the pendulum continues to swing back toward infantilization.
Now copy, too, is apparently on its way to sloppy seconds (the way of all things shared too much, howsoever assiduously groomed).
Absolutely nothing wrong can happen when someone else can randomly change the link to where you are trying to get.
There is zero change that the link can be replaced to a fake location that looks like the original, but is nothing but a phishing website.
From the Android Police image, I see they've completely changed the URL even before the ?. Do they have a database of rules for this? I can't see how else it would work.
Obviously this is just a means for google to attack other companies -- the ones that actually need an identifier to be passed around with you to track a link rather than just having a huge database on you already (like google does).
As an experiment I tried copying a URL from Google Ads.
Guess what?
I got the full, mangled URL, not the clean version that they do for their competitors.
Thanks Google. You truly have your user's interests at heart.
From TFA: It also recently introduced automatic blocks for bad and unwanted ads...
In other words, it blocks ads that don't contribute to the Google revenue stream. That's what they mean by Bad Ads.
So google is removing their own trash from our copied URLs? About time.
Because really, they were the biggest culprit of this.
What's the deal with all the browser Slashvertisements over the last year?
"Chrome has new feature"
"Firefox has impoved its privacy-related features (yeah right)"
"IE made some performance improvements"
If Slashdot were still run by the original owners, you'd see stories about browsers that are way better - you know, with the basic fundamental feature of not being made to intetionally track you, unlike these three.
Palemoon
http://www.palemoon.org/
Waterfox
https://www.waterfoxproject.or...
I'm pretty sure the market will adapt to whatever Chrome does, ensuring that things don't break.
I never want to share an AMP link. I also never want to share an m.domain.com link like m.facebook.com.
Are we saying that there's no longer a way to get the full URL to the clipboard? That would be intensely annoying in many situations. Is this a feature on mobile only?
And how do they know what they can safely trim? It they only do this for sites they understand then this should mostly work well but you can bet on annoying issues elsewhere. I guess understanding some of the common platforms like WordPress, Drupal, etc. could help but there are so many versions of those with constant updates that it's bound to trip up sometimes.
So this feature will mostly work and provide some convenience for the masses but the price is going to be confusion and annoyance for those who know their way around a URL, plus random breakage.
Maybe this is a sign that websites are using URLs in the wrong way. Can't they just move all that stuff Google is trying to hide into cookies and/or form fields instead so the URLs are kept vaguely human readable and not crazily long?
How exactly does it determine what information to keep and which to dump. I've got a sneaking suspicion that the answer is there is a server it sends the URL to in order to be cleaned so that it will always be up to date for thousands of sites without needing to distribute changes. The downside of this is that (obviously?) they will now be tracking every single URL that you copy. #AllTechIsEvil #GoingAmish2018 ;)
Anons need not reply. Questions end with a question mark.
Google and any Google properties are masters at obscurification. The URLs you see in the location bar are not even close to what gets sent to Googole, nor what is - and more importantly - exfiltrated. They once again make oppression _easier_ since in the case of things like youtube, the URLs are typically hardened with some kind of MAC. Change one thing and it fails to validate (IP being reported for example).
Big surprise I'm sure but they like Mozilla also push hard for a "clean interface" which they define as removing things like the address bar and even status bars.. You know, the things that would tell a user what in the hell their browser is actually doing!
Get fucked Google.
Chrome is now much further away from being standards-compliant.
Google is orders of magnitude worse now than Microsoft ever was back when Microsoft was still relevant. Using their monopoly to extend their monopoly and break competition will eventually lead down the road to government intervention.
This is a "don't pay attention to the man behind the curtain" nightmare for those concerned about privacy.
You mean like a Google URL?
h++ps://www.google.com/search?source=hp&ei=63iLWqKmJMKmjwPMqouwBA&q=pineapples&oq=pineapples&gs_l=psy-ab.3...0.0.0.6378.0.0.0.0.0.0.0.0..0.0....0...1c..64.psy-ab..0.0.0....0.n3CzS30gTZ0
Beware of the Leopard.
Aren't a good percentage of the GET parameters floating around out there Google Analytics parameters (i.e. utm_campaign1, utm_a, etc.)? Aren't they just cleaning up their own company's mess?
They should probably make this an optional feature that can be disabled.
BTW - I think this proves that they're tracking everything people do in Chrome. How else could they roll out such a computer paradigm-breaking feature with such confidence (on by default)?
Does google (or any company) actually use, or even test, the crap they push out to us?
I keep thinking maybe the web would be better if Chrome did not command the browser market. But then I see Chrome doing things that make sense and really focus on using the web. But as with Internet Explorer domination, Chrome domination will not be healthy long term. Google will eventually be able to control what the web does, if that day isn't mostly here already. I mean not a single browser is even close in market share. Do we not already see most web sites solely focused on Chrome? Even Firefox is just a Chrome clone and so in Opera. Edge pretends to be Chrome to web sites.
It does not delete the tracking information, it just doesn't show it to you so you don't know its there or not.
I always admired how malware links trying to hijack my machine helpfully hid information that they thought I didn't need to know. Good to see that Google is learning from the best.
From the linked article:
"This feature now happens automatically when sharing links in Chrome, either by the Share menu or by copying the link and pasting it elsewhere. Even though it slices off the extra bit of the URL, this doesn’t affect referral information. If you choose, you can copy and paste directly from the URL bar to grab the link in entirety."
From the actual article, that Verge sponged from:
"The URL streamlining happens automatically when you use the Share menu in Chrome (but not Chrome Custom Tabs). You can copy to the clipboard or share directly to another app—no setup required. If you highlight the URL bar and select text manually, you can still get the full URL with all the junk at the end."
Whoever paraphrased that for Verge doesn't understand how to read.
FC Closer
Yes, because every website is meticulously maintained by highly competent administrators who know exactly what they're doing.
It doesn't have to be. When over half your users use Chrome, if it breaks in Chrome, it breaks for most of your users at once, and someone gets yelled at until it works again.
You already aren't allowed to just copy what is in the address bar with Chrome. Google lies about the URL by removing the required protocol then corrupts what you copy if you have just the hostname selected by hatefully prepending the protocol. They just can't stop lying and corrupting data.
Are we really going to have a front-page entry for every little change or feature in every web-browser out there?.. Seriously?
But, if we are, Google's own tracking of your usage of their search results is pretty invasive too. Will Chrome remove that tracking too, the way some Firefox-extensions do?
In Soviet Washington the swamp drains you.
Maybe I just haven't read far enough in the replies, but it seems like everyone so far is missing the point and reaching for other reasons/defects/etc. AFAICT, the motivation and everything else is very simple...
When one shares a link that includes tracking information, and someone else uses that link, it weakens the value of the tracked info/user, because it's no longer tracking one user. The further that link spreads, the more diluted and useless that tracking identifier becomes. If they can strip it before it gets spread around, they can maintain more accurate data.
This isn't a them just cleaning up their own mess. This isn't them helping to keep your shared URL's shorter or cleaner. This isn't to help protect anyone from leaking tracking ids. This doesn't cut down on the amount of tracking done to users. It just improves the tracking they're already doing, all while (effectively) hiding that from you a little bit so you're less likely to be bothered by it.
To reinforce that theory, just look at the links they create in hangouts and gmail. Here's what "https://slashdot.org/" looks like when you "copy link address" and paste it: ... they're not trying to shorten jack shit. There's no real benefit to the user.
https://www.google.com/url?q=h...
Back to the share link thing, IMO, there should at least be a config item to set which keys get trimmed per-site, and maybe allow that to be configured by the site via a META tag.
Agreed. I may be accepting of there being a "copy raw url" from the context menu, but would prefer a copy to behave as always with a "copy simplified url" from the context menu.
With respect to safely trimming, google has a done a lot of work on determining the canonical form of a URL to limit redundant search results. I'll bet that it uses patterns in the URL only partially, while it uses the results rendered for their crawlers for various forms of the URL to determine the insignificant parameters.
I expect this is just step one. Once Chrome users are used to this, then Google will introduce step 2. That may be Google rewriting the links to channel everything through AMP, or it may be something less blatant but more insidious.
It's just another reason not to use IE6... er, I mean, Google Chrome. They're just going to keep taking further and further advantage of their dominant market position to go into full-MS mode. Twenty years from now, people will be amazed that anyone saw Google as anything but the second coming of Y2K Microsoft.
#DeleteChrome
Co's will just start putting tracking data in other parts of the URL, such as:
www.foo.com/tracking-crap1234/page-x.com
Table-ized A.I.
But now they've broken it. Why? To make tracking more accurate.
When you follow the link, the web server gets the tracking data. When you send a copy to someone else and they follow it, the tracking info pointing to you is gone so the web server doesn't think you clicked again.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Copy and paste works as expected. This affects the Share buttons only.
...it's a link to a Getty Image?
Now there's an exploit waiting to happen. Google, what ever happened to your byline to do no evil.
Since it’s mostly technical users who know what they are doing with URLs they should keep the full url. But I bet the XULless browsers will be putting this feature in right now.
FTFY
People are not wearing enough hats.
it does have a couple downsides, albeit nitpicky ones. For example, it eliminates anchor tags that will bring a user to a specific part within a longer article, so visiting a link that has been shared in Chrome will land you at the top of the page.
Why anchor tags aren't supported? Answer: no reason (AKA incompetence). Extracting that information is trivial and also storing it; sample approaches: including it within the main id (e.g., id=12345 for whatever.com/subwhatever and id=12346 for whatever.com/subwhatever#firstarchor), creating a secondary id (e.g., id=12345&id2=1 and id=12345&id2=2 for the previous example) or even keeping the anchor tags in the new URLs because they aren't too messy.
If you have enough resources to store/index as many pages from as many domains as required, you shouldn't find any problem to also deal with whatever number of anchor tags. Or, in other words, the first rule of any let's-slightly-improve-something-that-is-fine-anyway action should be making sure that it can still do what the previous version was able to do.
Custom Solvers 2.0 = Alvaro Carballo Garcia = varocarbas.
Altering the way a basic copy+paste function works is really not something I want, even if good intentioned. Why not just create a plugin that users can opt-in to instead? I rarely use Chrome (Pale Moon user here) but what they're doing sows a bit more distrust in them for me personally.
It is pitch black. You are likely to be eaten by a grue.
As far as the google link spaghetti goes, this is honestly part of why I choose bing. I can search for something and bookmark the actual URL from the search heading to look at later if its something I don't feel like opening at that time. Personally, I think this "feature" is pretty much crap. How reliable is this going to be? Is google going to guarantee that whatever it trims from a URL is going to keep the link functional? I don't use Chrome with exception to testing web code. Decisions like this will certainly keep me away from a good number of their products.
Another way browsers prevent me from seeing the links i'm clicking on.