No, Google! Vivaldi users will not get FloC’ed.

Zalex108

@Catweazle said in No, Google! Vivaldi users will not get FloC’ed.:

@LorenAmelang , same

YouTube Free Expression Award

¬¬

Catweazle

@Zalex108 alt text

jamesbeardmore

@guigirl said in No, Google! Vivaldi users will not get FloC’ed.:

I loved this:

We believe in a future where the web can provide people with privacy, transparency and control while also supporting responsible business models to create a vibrant, open and diverse ecosystem. Like Google, we support solutions that give users clear consent, and do not bypass consumer choice

Hahaha, onya microsoft, you've still got your legendary comedic touch!

Haha I was about to say the same thing! Newspeak at its finest! I find it somewhat amusing that they dare even say this, given that Goggle and Microshaft are possibly the very worst offenders of all time regarding dark patterns, no meaningful consent, and bypassing consumer choice (although arguably Amazon and Facecrook are also pretty bad in this regard).

jamesbeardmore

@guigirl My work computer (Linux Mint Debian Edition) has the same files with exact same filesizes present. The date is slightly different but probably just relates to whenever I updated V. The _metadata/ dir contains a single json file that's 1.8k in length. All last accessed April 11th. I suspect the Floc processing files get created ready to be used, but if the code's disabled, they never get updated. Otherwise we'd surely not see exactly the same file lengths on two different systems with different usage patterns.

Haven't checked my personal (Gentoo and Devuan) boxes yet. I might see if I can find a Losedows computer to look at, to see if that has a Floc dir too.

mtrantalainen

How about supporting FLoC API but always using hardcoded FLoC group "69" or "31337"? That way servers wouldn't block Vivaldi for missing FLoC support but in worst case all Vivaldi users were considered a single identity.

FLoC API provides no way for the 3rd party servers to know if the FLoC group id is real or not.

Another nice option would be to use seconds since UNIX epoch as the FLoC id. All Vivaldi users would use one global id which is changed every second. Good luck trying to recover any data out of that from the 3rd party server log.

GearDoc47

@Catweazle said in No, Google! Vivaldi users will not get FloC’ed.:

"I think that in general the ads are overvalued, "

Don't let the advertisers hear that!!!
Where else will the ad agency's be able to make a living???

Catweazle

@GearDoc47 ,with a decent job, perhaps?

GearDoc47

@jamesbeardmore
Ah, the alternative truth we have heard recently in the USA........

code3

@mtrantalainen I think the main use case for faking FloC is to prevent fingerprinting. ( We need to blend in with the flock)

I don’t think websites will be able to block you for having FloC disabled, Trust Tokens, yes, but not FloC.

Changing every second is a bad idea, too fingerprint able.

One cohort for all Vivaldi users could work, but there are only 2 M V users.

The best answers are probably having a random cohort that changes each session, or the empty cohort. The empty cohort is for those whose cohorts would convey sensitive info.

code3

@jane-n Could you be more specific on what will be disabled? Wouldn’t disabling the FloC API make V users too fingerprintable? Wouldn’t disabling Trust Tokens be a very bad idea?

Chas4

It is a no FLoCin way from me also

This post is quoted in The Register https://www.theregister.com/2021/04/14/browser_makers_reject_google_floc/
Also "WordPress core contributor proposes treating Google FLoC as a security vulnerability" https://www.theregister.com/2021/04/19/wordpress_core_contributor_proposes_treating/

jamesbeardmore

@code3 Another option would be to take the approach of addons like "blender" (if it still exists), or Firefox's privacy.resistfingerprinting. From what I remember, using either the addon or the option results in you appearing to be running the latest version of the browser on a 64-bit x86 Windows 10 computer, using US English system locale, with only default fonts installed, etc.

Therefore, just set the Floc cohort to be the most common, widely-populated value. Update the value each week to what the biggest single cohort in your region (or simply the world) is that week. Blend in with the masses.

Even taking this approach, I'd still appreciate the option (maybe even default option) to simply not participate in the Floc scam at all. That way you're not just refusing to participate, you're standing-up and being counted, sending the message that you're refusing to participate.

code3

@jamesbeardmore said in No, Google! Vivaldi users will not get FloC’ed.:

addons like "blender"

Does it still exist? I saw an addon that had a blender icon and advertised the same thing, blending in, and right now I am using an addon by the same author that hides the font list from websites, similarly to Tor browser.

@jamesbeardmore said in No, Google! Vivaldi users will not get FloC’ed.:

Therefore, just set the Floc cohort to be the most common, widely-populated value. Update the value each week to what the biggest single cohort in your region (or simply the world) is that week. Blend in with the masses.

I think that would be the empty cohort, because it combines multiple cohorts like wheelchair cohort, hospital cohort, victim cohort, together to hide them.

@jamesbeardmore said in No, Google! Vivaldi users will not get FloC’ed.:

Even taking this approach, I'd still appreciate the option (maybe even default option) to simply not participate in the Floc scam at all. That way you're not just refusing to participate, you're standing-up and being counted, sending the message that you're refusing to participate.

Should certaintly an option, but not the default until V has a large enough userbase to use its own UA.

Edit: The best solution would probably be combining both solutions: at startup, select one of the top 100 cohorts to spoof.

wstonestreet

This post is deleted!

Panino

@juanvase: What that "crypto token" do anyway?

Pathduck

@Panino See this most excellent post from user @nomadic :
https://forum.vivaldi.net/topic/58228/google-extensions-crypto-token-what-it-does

Dr.Flay

Steve Gibson talked about Vivaldi's stance on FLoC
https://www.youtube.com/watch?v=7zlv-fRWRrM

paulojrmam

Will Vivaldi ever abandon the blink engine for one of their own doing?

iAN CooG

@paulojrmam said in No, Google! Vivaldi users will not get FloC’ed.:

Will Vivaldi ever abandon the blink engine for one of their own doing?

That requires a lot of coders, never say never but not in the near future. Make Vivaldi grow by spreading the word so they can earn more and then hire more devs. But you also know what happens when a company gets too big, and when it happens, only earning more money will matter. It's inevitable.

Ayespy

@paulojrmam Seems unlikely. The Web is hostile to new browser engines.

Back when Vivaldi founder Jon S Von Tetzchner was running Opera SA, he made the mistake of attracting and accepting outside investment funding. The investors eventually gained administrative control of the company.

At the time, Opera had the fastest, lightest and most versatile browser engine in existence, named Presto. Problem was, Opera only had something like a two or three percent share of web usership (numerically vastly more than Vivaldi enjoys this early in its development) and almost no web developers or web engineers took it into account in web development. They did not test in it to ensure their pages were working right. This led to a real compatibility problem between Presto and sites that were built for IE, FF and Chrome. Jon kept hiring additional developers to address the compatibility issues, patch Presto to properly interpret improperly-written sites, etc. But Opera could not keep up with the widening compatibility gap, and could not deal with the fact that thousands of sites deliberately sent Presto bad code and unworking pages.

At this time, Jon's preferred solution was to hire more developers (he had over 200 - like ten times what Vivaldi has now), but the investors wanted to save money, their solution was to dump Presto and select a new, "compatible" browser engine, and so Jon ultimately left the company. Presto was abandoned, and Chromium (ultimately Blink) was adopted as the Opera engine. Problem was, they were not looking to make the best browser now, but to sell the company and make a ton of cash. The rest is history.

Jon gathered a team and launched Vivaldi browser, with the aim to serve the public Opera had abandoned. It would take a decade or better to fully write and deploy a new browser engine, Presto was still owned by Opera and still incompatible with thousands of websites, and Jon had a tiny team. They studied which engine to use, and settled on Blink but with major architectural differences from the way Opera was approaching that adoption of a new engine.

So, Vivaldi. Like the old, classic, Presto Opera, it is uniquely designed, and uniquely flexible. But it is mostly, by default, compatible with the modern web - so a huge team of developers to address the compatibility problem is not needed. But there is also not the time, the economics, nor resources to start all over again building a new and unique browser engine.

I don't think we'll be seeing one.