[dao-]

I remember, I was around back then. :) Panorama was in some ways ahead of its time. The UI was nice visually but also somewhat heavy handed / not very beginner friendly to say the least, which contributed to it being used only by a relatively small share of our user base.

We're cautious about not repeating history. We're implementing tab groups from scratch and directly in the tab bar. Our Firefox View feature (Tools > Firefox View) may later get a more visual surface for managing groups.

[nokeya]

It was used by advanced users. And what advanced users do? Disable telemetry;) So, I suppose the share was a little bigger but not that much.

[eddyg]

Or, just don't disable (anonymous) metrics.

There's a BIG difference between tracking and metrics, but they are often treated the same, especially by "power users".

[titusjohnson]

There's no difference between tracking and metrics, they're the same thing. You get your metrics out of the data you track. Browser phone home? Tracking.

And there's no way for a user to validate that any tracking is indeed anonymous. The technical level needed to asses this is just... out of reach for everyone (the quantity of people who can properly verify this is small enough we can safely ignore it when speaking generally and use the coloquialism Everyone)

[eddyg]

There is definitely a difference between tracking and metrics. One saves and associates information like “when and where did this piece of data come from?” along with other “identifiable” information, the other simply increments a count for things like “advanced feature X enabled”. If you don’t see the obvious difference between this type of data, then that’s on you. The latter can provide extremely valuable signals, and when “power users” disable it (and tell regular users to do the same, spreading FUD) because they think it’s “tracking” (it’s not), that’s their problem when a tool/app/service starts moving in directions they don’t like.

[titusjohnson]

I was too succinct in my previous message, I guess.

To you and I there is a difference between tracking and metrics. To everyone else there is none. All my Mom knows is that the software dialed home. It is impossible to verify what it said to the mothership. The popup promised "Metrics only!" but then Little Snitch lit up like a Christmas tree 4 times! It's stealing my info!

How does the consumer know that the "metrics" didn't include home IP, OS version, and god knows whatever else? Again, metrics are a subset of data. Even the internet request to ship a fully anonymized usage info like "saved_files: 10, opened_files: 11" metric set contains Gobs and Gobs of identifiable information on and around the request itself. Does the company stash inbound request data for troubleshooting? That's fucking tracking bro. Your data dog instance is chock-full of tracking info.

It is not reasonable to expect end-users to be able to verify the claim that "only metrics are tracked". It is safer for everyone to assume this is a bald-faced lie, because at the end of the day it is impossible to verify to any level of certainty.

[krageon]

They're treated the same because they are the same, which people with domain knowledge (i.e. power users) are aware of.

A terrible way to get them to stop doing something you don't personally agree with is by starting your post with a bad idea, support it with a lie and close it with a personal attack.

[LtdJorge]

I explicitly enable it

[Y_Y]

Maybe instead of ubiquitous stupidity-tqxax telemetry we could have some neo-Nielsen families and get to pick a roughly representative sample out of voluntary, compensated users. A trusted third party contracts the victims and agregates the data. Don't ask me who regulates or pays though.

[jackstraw14]

> A trusted third party

And why wouldn't it be Nielsen :) I remember when they sent me cash in the mail as a kid, fun times.

[cbsks]

They still do that. Just a few weeks ago my daughter filled out their survey and got $5!

[jackstraw14]

That's so cool, mine was about 30 years ago. I had no idea they were still doing this!

[OutlawHusbando]

> We're cautious about not repeating history. We're implementing tab groups from scratch and directly in the tab bar. Our Firefox View feature (Tools > Firefox View) may later get a more visual surface for managing groups.

Hi! Thanks for letting us know about Firefox Team is being very careful this time, I want to share my idea too because I think the Firefox community have probably tinkered enough to figure out that Tree Styled Tab Group like Sidebery/Tree Style Tab is the best Tab Group implementation, because they're so easy to use, just drag and drop and work great in practice.

I hope you consider making Tree Styled Tab Group an option for vertical tab.

[justinclift]

> which contributed to it being used only by a relatively small share of our user base.

And that's why telemetry is such a brain dead idea. People then actually make decisions based upon "number of people using feature X" which is incredibly... lets just say "unwise".

[1oooqooq]

wonder if we should just accept it as "voting" and monitor their telemetry experiment and spam the option we would like

[ReadCarlBarks]

You can vote for features on Mozilla Connect: https://connect.mozilla.org/

[1oooqooq]

nice find. nobody uses that and sadly it became a desperate support forum which mostly goes ignored

[justinclift]

Seems like a reasonable idea. With this in place, why are Mozilla using telemetry instead?

[1oooqooq]

because Mozilla, the foundation, is hostile. it's a political money grab.

with telemetry they can justify whatever. they can go to the board and say "look people are clicking more on pocket, which we put prominently on the UI against everyone wishes, and they barely use bookmarks, which have critical bugs open for over 20years... who would have thought of this counterintuitive insight if it weren't for my genius persistence on trying and measuring new ideas uh? so are we good on the extra bonus to my close childhood friends?" ...see, it makes the pitch defensable, if you don't say the right parts out loud.

[justinclift]

Might be more accurate. ;)