Submitted by I didn’t believe it would be possible at first, but after spending the better part of a week on Chrome 46 I’m blown away. Memory consumption seems to have halved, groggy slow tabs are snappier than ever and my battery life isn’t shamefully bad anymore – also, my laptop’s fans aren’t constantly blowing.
It’s going to take a lot of convincing to get me to switch from Safari back to Chrome on my MacBook Pro.
as this topic seems to be on everyone’s mind at the moment, how much data does it send to Google?
If the level is close to zero, I might give it a go.
Personally, I think that the subject of ‘phoning home’ should be mandatory for reviewers to include in their reports.
I heard yesterday that a popular Windows Installer phones home to the instaler vendor not the vendor of the product you are installing, everytime the .msi is executed. IMHO this is getting silly.
If you use Chromium with duckduckgo as your default browser, you get all of the good things about Chrome without any of the google spying.
There was 1 information leak to google that was included in Chromium (voice search would send data), which was found by a Debian developer, and was removed from the default build.
Apply logic, please. The above doesn’t mean there are not other leaks. Don’t ever trust Google.
The good thing is, when Chromium wants to use some Google Service, it has to identify itself using developer keys (the process and the services it uses are on the Chromium wiki).
The bad news is, if you intentionally build Chromium without any key, it will bug you about it at every launch (see Dartium, that one was built this way).
Totally agree, i would not touch Chrome with a ten foot pole. I still use links2, netsurf-fb on linux and on mac and windows i use firefox.
That would apply for Linuxes like Debian or Ubuntu where you in fact use Chromium (not Chrome) when you install it from distro’s repo (of course you can install Chrome straight from Google, but then it’s full-blown Chrome). If you can live with nightly builds, then Chromium is also availbalbe for Windows and OS X.
In Chrome there are two trackings you cannot disable – first-launch tracking and RLZ tracking. And they both should be missing in Chromium (both in official nightly builds and the versions prepared by Linux distros).
However this voice search bug is a prime example how new features that undercut privacy can slip into stock Chromium code and stay there until someone finds them and disables. And of course there may be more of those.
The RLZ identifier is only used if Chrome was downloaded as part of promotional campaign (i.e pushed through OpenCandy when installing other software). The build downloaded directly from Google website is clean.
Apart from that pretty much everything can be disabled in privacy settings. In fact that’s all those unofficial builds like SRWare do – remove those settings completely. They rely on fearmongering to generate ad revenue themselves.
People are complaining that Chrome sends everything typed into address bar to Google. Well *OF COURSE* it does if you have search suggestions enabled. How else would it work? Every so-called botnet feature of Chrome serves a purpose. It’s up to you if you want to use these features or not.
Having said that, I don’t even like Chrome myself any more, for reasons other than privacy. It’s becoming bloated and introduces intrusive features that I don’t care for. User switching (with permanent icon in the titlebar) – that encourages people to share an OS account – a really bad decision, or that damned bell icon in the system tray/OSX’s menubar, custom notifications not using the OS-level notification centre, or most of the features from New Tab page moved to already overcrowded “hamburger” menu. Meh. Don’t even get me started on that extremely silly “you are now in full screen” popup on videos, that I can’t get rid of unless I add every website to some sort of whitelist…even if I won’t visit the site ever again. Yeah, running a video in full screen is apparently a big deal for Google devs.
There is no major browser that sends ‘close to zero’ data in normal operation. Every one of them is chatty as all hell…
https://www.google.com/chrome/browser/privacy/whitepaper.html
That spells it out for you. In short, it does mostly the same shit that all browsers do, plus some additional stuff if you log into google services. Don’t log into google services? Then it is no different from Safari or Firefox, which both also send data to someone in some form or another virtually anytime you do anything (type in the address bar or even just navigate to another page)…
Nothing wrong with being paranoid… Sometimes they are out to get you. The problem that you are not seeing is that everyone is out to get you – it isn’t just google
The main issue with chrome (where it differs from other browsers) is the integration with google services. If you don’t log into their services and choose a different search engine provider, then there really aren’t any significant privacy differences between it and all the other major browsers. If you do use google services, well than you pretty much signed up to be a data feed for them. Them’s the breaks…
Chrome has a couple of things that can’t actually be disabled as already mentioned above. Firefox’s chattiness can be easily managed but the point is moot on Windows 8 & 10. Edge is going to “phone home” no matter what you do as it’s an integral part of Windows 10 and there are several things it does that can’t be disabled nor blocked without air gaping the machine even if you’re using Windows 10 Pro.
So long story short, using Microsoft or Google products expect to be tracked and cataloged. Apple? Guess it depends on what you’re using. I have little experience with Apple’s ecosystem. Linux/Firefox & other open source platforms privacy is easily managed by choices you make and not choices others make for you.
Seems like a prime opportunity for either Apple or Canonical to step up to the plate and advertise they will protect user privacy while it’s obvious both Microsoft and Google will not. I don’t really expect Apple to do so entirely because they sell ad services for iOS, but for desktop users that assurance would be an extra selling point.
Honestly, Chrome was never slow for me (Macbook Pro 15″2013). I also use Firefox, but whenever I did some week-long tests with Safari, it felt slower.
Totally agree. I started using Safari when it was still in beta. I switched to Chrome on my Mac after using it at work for awhile. It’s much faster than safari. When they do release a safari update, it often “catches up”. Now that Google has forked webkit with blink, apple doesn’t get free enhancements anymore from them so Safari falls further and further behind.
At this point, apple needs to either put a real investment in safari or kill it and make a deal with another browser vendor for iOS. Better still, just allow others to ship full browsers on iOS.
The last straw for me was an update on safari deleted all my bookmarks. I was able to restore them from time machine, but that should never happen.
Sorry to hear about your bookmarks…
But I’d like to point out that in, WebKit was needlessly forked by Google into Blink, because the Apple guys had already left the webkit bandwagon a couple of years ago… In April 2010 to be precise, back in the Safari 5 days.
WebKit2 is a API-breaking fork of Webkit1, so any changes made later by Google could’ve been backported but being a dev myself, I know for a fact that it is quite difficult when API drifts too much from one another… So possible, but as time go on, unlikely.
Why did the Apple Engineers forked the Webkit Project back in April 2010 (i.e. Around Safari 5.x)? Because they wanted a “separate process per tab”, the same feature found in Chrome, which coincidently not implemented by Webkit1 or Blink for that matter, but by Chrome/Chromium itself.
So for the last 5 years, any speed improvements made to Webkit2 is an Apple-only effort, as Google was not contributing to Webkit2… And I have family working at Apple on Safari/Webkit, and they have quite a team working on it, rivalling the size of the team working on Blink at Google (From all public account of the size of the team, which is sketchy at best, from both companies)
As for Webkit1, it is now a bit of an abandonware, with Blink being the bulk of their efforts.
All that being said, I use Chrome when plugged in the wall, because I know web sites that just do not work with Safari. But on battery, I use Safari, because it is softer on my battery.
ANY improvement to battery life with Chrome making it similar to Safari in that particular area, I will take (and drop Safari)
Edited 2015-08-22 22:39 UTC
Looking at https://en.wikipedia.org/wiki/Blink_(layout_engine) says Blink is fork of WebCore component of WebKit, and they are actually quiet different components – in development and licence. So I don’t understand it, and I am not looking that much into details in here but it doesn’t seem that chrome team where using old version of webkit and they just switched. I think they had enough of their own way in the background in chromium that they needed that other component also get aligned into their other parts like V8 and the rest.
Anyway I don’t think anyone should trust any of these browser binary blobs and start using chromium or firefox (which is really really slow and buggy for me), but chromium on arch or gentoo is actually quiet solid.
You just should not trust any Apple update to not delete data. I’ve had the same issue occur. I pretty much treat my macs like chromebooks at this point. Use non Apple services and as little of Apple’s software as possible and its a nice system.
Every website you visit gets to know your previous link by the HTML standards. So what’s it matter if your HTML is tainted by default, too. The whole purpose is to erase anonymity.
Only if you are going to a webpage by clicking on a link (HTTP-Referrer). If you just type the URL or use a bookmark it doesn’t. This has “always” been the case and was never a problem except for querystring trouble like “refer was http://sitex/login?uname=Me&pwd=YouGuessedIt“
For Firefox, Seamonkey and other Mozilla browsers you can disable the sending on referrer. In old Opera as well (don’t know about the new one). For Chrome & Chromium you need extension for this. For IE you can not do it. I don’t know about Safari.