Minecraft to add telemetry

[belval]

Eh Minecraft is a video game, I just don't see the privacy risk with "typical" Windows-level telemetry happening there.

At some point we have to accept that this data actually helps them fix real issues too, they're not monetizing their players directly with it.

[slg]

>At some point we have to accept that this data actually helps them fix real issues too

It is strange that this community of all communities has such a resistance to this idea. Software developers should know better than anyone how difficult it is to identify and fix vague software problems without having specific details about the problem. Yes, there is a negotiation between the value of telemetry and privacy and often too much privacy is sacrificed. But I am always surprised to hear developers say all telemetry is bad.

[oneplane]

I don't think what type of application it is determines if a person has a right to their privacy if they want it. Same goes for data, we don't have to "accept" anything.

On the other hand, if it wasn't Microsoft's brand being connected to this and it wasn't called telemetry but "Automatic Bug Report Sharing" nobody would make a peep.

The same goes for the kind of data or application: if people know what type of data would be shared and what it would look like, they might indeed understand that this isn't some nefarious profiling but just normal software improvement. If you are a developer, bug reports are nearly useless unless it's super repeatable with normal conditions and a few clearly defined steps... or if there is a useful automatic reporting system in place.

But none of that reduces someone's expectation of privacy or acceptance of data sharing.

[falcolas]

> they're not monetizing their players directly with it

How do you know? Did they pinkie swear on it? Did they add any kind of T&C around their use of Telemetry specifically?

Telemetry that can be used to identify frequently used game functionality could equally be used to identify game functionality that can be monetized. Identifying how a user dies is a fantastic way to identify items which can be sold to prevent those common deaths.

[hulitu]

It did not helped them fix problems in Teams, Windows and Office 365. Why do you think Minecraft will be different.

[sneak]

Windows-level telemetry is not "typical"; it includes a dozen or more different spy subsystems. Most spyware apps only include 1-3.

If you are judging based on Windows as a benchmark, almost all software is going to come in as acceptable.

[tinus_hn]

Typical telemetry could well consist of ‘what processes is the user running’, ‘what kind of hardware do they have’, ‘where is the user’ and ‘what is their data that can be matched to their advertising identifiers’. Which is the typical data that can be monetized.

[II2II]

The problem is an irrational degree of distrust in the industry as a whole.

It's not that I blame those who distrust any form of telemetry. Many companies are eager to harvest as much personal information as possible. Depending upon one's definition of personal information, some of the data acquired by telemetry can be used for that purpose. In many cases the data collection process is deliberately opaque. In the remaining cases, very few people have the ability to verify that what is actually sent reflects what they are told is sent. That's before factoring in Microsoft's involvement here, since they have a negative reputation in some circles due to their past business practices.

[rnd0]

ha ha ha I came here to troll -or at least point out tongue-in-cheek the same thing.

But seriously though, it's not on the same level as, say -visual studio code sending back telemetry which includes potentially secret code or anything. Or Defenders willy-nilly sending "samples" back to the mothership for analysis.

I sincerely don't mind it if a game sends this info back, I think it's actually a good use of Telemetry.

[littlestymaar]

> At some point we have to accept that this data actually helps them fix real issues too, they're not monetizing their players directly with it.

What do we know about that?

Microsoft have long lost the benefit of doubt with their shady data collection practices.

[ginko]

> Eh Minecraft is a video game, I just don't see the privacy risk with "typical" Windows-level telemetry happening there.

But then Minecraft is a video game. Why do they need to spy on their customers?

[zh3]

Minecraft is a game played mostly by kids (minors). Tracking their behaviour, finding out what they try (even if just looking for ways to game the system) is pretty useful info.

[lmilcin]

But the explanation just does not satisfy me.

I mean, world building is an expensive but actually pretty simple, predictable operation. If they want to see how it performs on slower computer they don't need to get telemetry, just actually run it on slower hardware.

[kortex]

> In this release, we are re-introducing diagnostic tracking, which was part of Minecraft: Java Edition until 2018.

So it was there, they took it out, now it's back.

[foepys]

If it was "game crashed, here's the stacktrace" before and now (or later) it's "player put 20 blocks of gold on server IP 127.0.0.1 with players foo, bar, baz present", it's a difference.

[xahrepap]

According to Minecraft YouTuber/commentator Xisumavoid[1], it was removed for GDPR or a related law. And now they're adding it back in a compliant manner.

1: https://www.youtube.com/watch?v=4WEX_ICnZHE or the snapshot video just before it?

[vkoskiv]

The phrase "...to better understand our [users] and to improve their experience..."

always rings alarm bells in my head. It's corporate lingo for "We want to increase profits by selling your private data for as long as we can get away with it."

[pdpi]

It's very generic phrasing, and some products do hide nefarious behaviour behind it, but it's also just an accurate description of the most innocent, straightforward use of telemetry out there. I, for one, can't think of a way to monetise Firebase stack traces.

This is a game, so let's use other games as examples. Here's two pretty damn good talks from GDC about what the value telemetry brings to the table (the StS one goes into a whole bunch of other stuff that's sort of irrelevant but also serves to motivate the telemetry topic):

* https://www.youtube.com/watch?v=Urx7WQE6NY0 * https://www.youtube.com/watch?v=7rqfbvnO_H0

[pyr0hu]

What private data would you get by tracking minecraft players?

[bastardoperator]

This guy knows how to mine diamonds!

[moogly]

Not related to telemetry, but I checked out Minecraft about less than a year ago for the first time in about 10 years (I had to dig out my old account details from an ancient email and migrate my account through some strange form) and spent a week playing it until I had enough, and I was shocked, absolutely shocked at how little has been added to the game during those 10 years.

Sure, the fact that they have two divergent, separate games and are completely unable to move over to the C++ codebase and sunset the Java version due to the lack of modding (I remember the modding community screaming about needing a proper modding API back in the olden days, and Mojang hiring some of the Bukkit team back in 2012 to develop it, and it never came to fruition, still to this day), makes them having to do twice the work implementing stuff, but that's at least parallelizable work. I doubt the same people work on both codebases?

I feel I need to sit down with someone who's working at Mojang and ask wtf is going on over there because it could be fascinating.

[Freskis]

The amount of feature additions in the last 10 years has been enormous.

Is it possible you were playing an old "world" which doesn't have the latest features? Or were playing an older version (which is supported in the latest launcher)?

[zeta0134]

Maybe they'll notice that when I launch the game on Linux, I get about 10-20 FPS boost over Windows on identical hardware. I've always wondered about that; in theory shouldn't it be the reverse?

[half-kh-hacker]

On top of the JVM implementation potentially being faster, OpenGL drivers on Linux are sometimes better than their Windows counterparts; especially with the less modern OGL that Minecraft uses

[frostirosti]

I mean, it's the JVM in the end right? It depends on the JVM you're running on between windows and linux

[bool3max]

Probably the JVM implementation. I have observed the same effect on my system.

[altcognito]

Specifically they need to telemetry on the biggest change they made: how poorly does the new world generation and chunk loading work.

This was a huge change and it is honestly pretty slow even on beefy boxes.

[supperburg]

I really want to see a Minecraft that’s written in a fast language and also takes advantage of the huge amounts of storage and memory that we have now. The main problem with Minecraft is that you can see the end of the blocks even on high settings. Would also be cool to see slightly smaller blocks

[rndinternetguy]

They did that - it's called Minecraft Bedrock and is written in c#. The problem is that it's not a 1 to 1 copy of Java edition and much of the functionality differs, so most PC players prefer the Java edition as it reflects the original gameplay.

[literallyaduck]

Here you go https://www.minetest.net/

[sneak]

Are there any products whatsoever that come out of Microsoft that don't contain spyware?

[drusepth]

It's not productive to be hyperbolic or reductive. There's a huge difference between telemetry and spyware, especially in this context.

[genewitch]

Comfort curve 4000 keyboard without the "driver" software?

[josephcsible]

I'm not sure if there's an option to turn this off, but even if there isn't, since it's Java Edition, a mod could easily be created to turn it off anyway.

[juice_bus]

It is still quite frustrating that it has come to the point that you may need to mod a game you purchased to avoid telemetry.

[crorella]

Telemetry per se is not a bad thing when you make it with privacy at the center of it. If you offer users the chance to opt-in/opt-out of logging and then REALLY respect that decision by first) not generating data when the user does not want to and second) not using the data for purposes the user does not sign up for (in downstream, analytics mostly)

then telemetry can actually be beneficial to keep improving your product and features.

[oauea]

Interesting design choice to have the website not be scrollable while showing content outside the fold.

edit: Apparently this happens when running the "I don't care about cookies" browser extension, which is necessary to make the web somewhat usable after the EU's ridiculous ruling.

[yuuta]

The key problem is that Mojang does not allow users to disable it.

[zimmund]

In previous versions it was enough with snooperEnabled=false

[joshghent]

Weird that a Microsoft product would get hard to disable telemetry…

All jokes aside though, this doesn’t seem too bad and I can put myself in their shoes to understand why they want this data. It would be difficult to monotone this information (asides maybe a nice shiny new Surface machine).

[zapzupnz]

I'm surprised it hadn't already, to be honest.

[swalls]

Even if they're not doing anything nefarious with that data, I'm against telemetry other than crash reports because data driven design makes for boring art.

[skeaker]

This assumes that the only bugs or other glaring issues that can be found with telemetry are crashes. There can be non-design-related problems with a program that negatively affect the experience without crashing it that can also only really be found with telemetry.

[Shadonototra]

microsoft really turns everything it touches into a giant poop

[h54545nb]

Well I was going to re-install and check out new updates but I guess not anymore. I don't care how helpful it is, if there is no details on the exact specification of what data is shared and no way to disable it, I won't support the product.

[buzzy_hacker]

The details are in the link?

Contains following information: launcher identifier user identitifer (XUID) client session id (changes on restart) world session id (changes per world load, to be reused for later events) game version operating system name and version Java runtime version if client or server is modded (same information as on crash logs) server type (single player, Realms or other) game mode

[rozab]

If the reasons they give for this were honest, you'd think their time would be better spent building the kind of extremely basic integration tests that would detect if, say, their world generation broke making the game unwinnable.

https://bugs.mojang.com/browse/MC-236618

Inflammatory zingers aside, I don't mind telemetry in a product like this and I'm sure I'd want it if I were working on it. But I'd want automated tests first. Factorio's approach seems enlightened, but very rare in the industry.

[diegoperini]

No amount of integration tests can replicate millions of players, organically interacting with a video game. The state space is too large for that.

[alexander-litty]

Not a great example. This bug was in one of the weekly snapshot releases, which regularly and knowingly have some unplayable element to them.