Second Life — both the and the — compliant with European Laws and the European General Data Protection Regulation (GDPR). Please stop claiming the contrary, unless you wish to present proof from a court of law's sentence on the subject. To the best of my ability, I couldn't find any. In Europe, entities, public or corporate, also innocent until proven guilty :)

The GDPR is more concerned with users about what happens to their data, and less with what can legally be done or not.

Ultimately, what is (and that's the reason why fines are applied to non-compliant applications or companies developing such applications) is to . In other words: it's forbidden to send user's private data to a third-party (which may do with that data lots of things, many possibly not entirely legal... that is irrelevant) . What forbidden is to do it anyway without the user's explicit consent — what happened in the past with the many fines applied to non-compliant products, services, and companies.

Linden Lab states very clearly what is done with your data when you use the SL Official Viewer to connect to the SL Grid(s).

Now, if a sends data, or some of your data, to unknown third parties, , then the is not complying with the GDPR. For instance, it was usual in the past for event hosts holding discussion events in text chat to ask permission from all participants to post the chat log somewhere (on the community forums, for instance, or copying and placing it on a notecard to distribute in-world to others). Sometimes, one user or two would give that permission, and, as a consequence, some chat logs would have been redacted to edit out these users' comments.

With 'bots and other similarly automated tools, it might be harder to ask for permission to send a resident's data elsewhere. . The only thing that LL is supposed to do, in this circumstance, is to the ability for users to opt-in to whatever real-world data is being captured by the 'bot itself.

It is questionable if a 'bot may be able or not to access other residents' data. That's one of those cases where there had to be an actual litigation to get a court's view of the application of the law. My guess is that 'bots cannot capture any "real-world" data, so they're fine to do so — after all, the LL ToS states that, once you're in the public areas of SL, you automatically grant everybody a license to view whatever content is produced. Technically speaking, text is also "produced content", therefore, whatever you type in SL, is supposed to be publicly available to everybody else .

Now, the argument is... what if I such text and send it to someplace else where the LL ToS do apply, such as Discord (but it could also be on WhatsApp, or a non-LL wiki, or a personal blog...)?

Theoretically, again, the special license automatically granted to the content you create in SL . It might be argued that capturing text chat (or recording voice chat!) and is, indeed, a violation of the other person's (but not necessarily their ). That's why the current procedure is "ask first, post later, restrict to those who have explicitly given consent". The same obviously applies to textures, 3D models, etc., with an even more special exception set on Snapshots and Machinimas, which be posted "outside SL" without requiring authorisation, so long as the scene is entirely captured in publicly-accessible land.

That said... it's not up to SL to control what you do with content generated inside SL brought SL. What LL's ToS only says is: to join SL, you need to agree to a special set of mutual licensing between all participants in the same virtual environment, . Outside it, the usual legislation applies, according to jurisdiction; in other words, copyrights and authorship laws regulate what can and cannot be done with content from SL.

And that ought to apply to text, too, since text protected in most jurisdictions under the auspices of copyright & authorship laws.

Again, note that none of the above is related to laws. It's just laws.

And, I'm sorry to contradict you, but there are no privacy laws applicable to content, i.e. what you write in SL which is written on a channel... is, well, . Using that content SL does not relinquish your and rights over what you've written; the ToS only activates an automatically-granted license to your content to other SL users while they're connected to the virtual environment. What happens that is regulated by the ToS. It's just copyright laws. And, unless given special permission, you theoretically cannot copy that text (even if attributing the "authors" of such text!) and post it anywhere else.

That most certainly includes Discord.

A 'bot that just copies from one chat group to an external system — be it Discord, be it something else — is violating any laws. It's "only" violating copyright laws, if the 'bot was not explicitly given permission for that. It is also irrelevant if there are no obvious mechanisms to tell the 'bot to copy text from residents who given permission (there are, but let's assume there aren't for the sake of the argument!). If the 'bot cannot be trusted to only copy authorised content, it is not only in violation of RL laws, but it's in contempt of the ToS, and, as such, should be suspended/banned from SL altogether. The 'bot , beyond that, would be liable to be sued for content piracy , and, assuming that such a litigation case actually reaches the court, LL would respond to a request by the court to give out the name and address of the user owning the 'bot in question — so that person would be the target of a lawsuit (in some jurisdictions, piracy may even be a , so it might be more serious than just paying a fine; it would depend on the jurisdiction and the gravity of the offense).

That said, to comply with international copyright & authorship laws, group chats would only need to have a checkbox saying, "I authorise all text on this group to be sent elsewhere, outside the environment of Second Life". Such a checkbox would probably require being linked to a web page (or potentially a notecard or similar mechanism) where the actual usage of the text outside SL would be described. And, according to European laws, such a checkbox be opt-in by default — you assume that people "automatically" grant others the right to use their content, even if such content is "merely text". Doing so would, indeed, violate a lot of laws — and remember, it's not only Europe that has such laws, these days most of the Western world (and, in the US, especially California!) has some legislation regulating content-related rights in the digital world.

Now, where does the GDPR enter in this equation?

Imagine, for a moment, that the 'bot would, indeed, get permission (via the checkbox) to copy someone's chat to Discord. But what the 'bot owner would tell anybody was that their 'bot could actually hack into other people's computers, figure out their address/location, their computer fingerprint, and a plethora of other data which are publicly accessible, just harvested by the 'bot "because it can", and, worse than that, a constant stream of such data would be sent to a third-party that is confined to the European Union (such as Discord).

Now is a privacy violation!

In such a scenario, if discovered, the 'bot owner would be liable to the full extent of the GDPR laws, and I agree that they would have to pay hefty fines for such a violation, and — who knows? — possibly even face jail time (again, depending on the degree of offense and the jurisdiction). *without their explicit knowledgefull permission*

This is what happens, indeed, when Apple or Google capture non-public data (such as location information) without their users' explicit knowledge and without their explicit permission. The same also happens when web pages use cookies or any other similar tracking mechanisms which extract non-public data from individuals, not telling them what they're doing, nor telling them what they're going to do with that data, or where they store it.

In the 'bot scenario, it's clear that the 'bot owner is forced, by the GDPR, to what data is being gathered — beyond the text chat, that is — how it is being processed, where it is archived, who is given access to it, and so forth. *And all that be written in advance before the user clicks on the checkboxmust* know what they're committing to when they use a service, since it's very easy to capture non-public data outside the scope of such a service, and funnel it into data-crunching algorithms. So... no 'bots capturing chat without permission, period. No 'bots capturing any non-public data about users, period. That's a no-no, and liable not only to lawsuits, but even (possibly) criminal indictments.

But what about Linden Lab? How are they a "concerned party" in all of this, and thus liable to the threats you've made about terrible things that may happen to LL if they "fail to comply with the GDPR"?

It's simple, really. LL be accused of violating the GDPR the Second Life Viewer, or anything inside the SL Grid, be easily used to extract such non-public data and send it to third parties (much less their permission!).

But we're talking about a 'bot, the SL Viewer.

Also, we're not talking about the sent to Discord. That, as said, is covered under copyright & authorship laws, and, as said, the checkbox + a page/notecard explaining what will happen to the text would suffice.

Instead, we're talking about a 'bot that has the possibility of penetrating into the remote computers of all participants in the text chat and extract data from them, their knowledge. This is actually even than mere "privacy violation" — it's invasion of property, a cybercrime in most jurisdictions.

Needless to say, such 'bots don't exist :-) (but we're talking hypotheticals here).

But what if things are slightly less obvious, i.e., what if the 'bot owner claims that they are attacking directly any computers, but instead they're tapping into a stream of data captured via the SL Grid, which — for some weird reason — already does that on behalf of the 'bot owner. In other words: let's suppose that the communication between the viewer and the SL Grid somehow sends personal data from the user to LL's SL Grid, and, at the same time, from SL residents. The "normal" viewer could hide all of that inside its code, and regular users would never know; a 'bot, however, could tap directly into the communications and extract all of that for later storage.

It is conceivable that, in such a scenario, Linden Lab would be given a fair warning to close down such "information leaks" being accessible to third-parties so easily, and possibly given a reasonable amount of time to fix their code. After that time, LL continued to ignore the issue, they would be in contempt of European law, and could not only be fined, but, more likely, their services would be not made available in Europe, until they agreed to comply.

This is, indeed, what companies such as Apple, Microsoft, Google etc. have faced. They capture personal information from their users to do tracking & profiling; they did tell anyone what was going on; when the European authorities informed them that they were violating the law, they ignored such 'threats', until, well, they had to face a court of law, present their arguments (the hardest one to explain was not the data was captured, not even how it was being , but rather, knowthat* is what is highly forbidden, or even criminal, under the European laws), and face the consequences of the sentence.

Therefore, it could be argued that LL, by refusing to shut down the data stream containing private data, was negligent, but could even be considered to be an accomplice of the 'bot owner — a facilitator, so to speak, for the 'bot owner's illegal activity. And, as such, LL would be liable to any lawsuits against them, . They weren't, but they made it and refused to change it — then they're also at least partially responsible.

This is, for instance, why Apple and Google are not very fond of allowing apps that show pornography to their users, to give a stupid example. Pornography for adults is legal across many jurisdictions. However, there is always the possibility that a child borrows their parents' mobile phone and gets caught watching pornography — truly nothing that hasn't been done in the past! Apple and Google are the "criminals" here, for selling/showing pornography to minors. However, they made it too easy for minors to grab their parents' mobile phones and simply use them that way. It would be like having guns in the house but giving the children the keys to them. These companies could, in theory, be accused of being negligent, and be required — by a court of law — to making sure that no children can ever watch pornography . Some jurisdictions allow parents to explicitly give permission to their children to watch pornography, but not all do. And, to make matters even worse, both Apple and Google receive fees from pornographic content sold through their stores, so, in a sense, Apple and Google are from pornography apps in their stores — which makes them also responsible for the "crime". As such, to avoid any problems or issues, pornography is banned on the Apple and Google shops — even if it could be legitimately been shown or sold under certain circumstances. It's just that it's too dangerous for Apple & Google to expose themselves to the risk that "something might go wrong".

There is a reason for the "Disneyfication" of a lot of products and services available via the Internet — it's the safest way to guarantee that no laws are ever broken, even if the content providers or infrastructure providers are not really part of the scheme.

Anyway, I digress. Let's go back to Second Life, and look at the of what's going on there.

*Linden Lab doesn't capture any private data from their users and sends it to other viewers.*

It's not merely a question of repeating what LL already claims on their ToS and related articles. If such data were being extracted and sent in either way, . 'Bots are unable to capture data . So, that can be easily demonstrated by just looking at the code; or, conversely, LL's alleged negligence and/or malicious intent is easily disproven by noting that LL's code does not allow any private data to be extracted and transferred by their viewers. That should be enough, in a court of law, to show that LL is not liable to any "leak" of the private user's data that they store; and since we have the source code of versions of SL (well, almost all, I'd say), LL could even prove that no data had been captured and/or leaked in the past, using their viewer.

But what about malicious TPVs, or, for that matter, malicious 'bots?

TPVs are easier to deal with: since the source code for SL is under the GPL, so must TPVs be, which means that source code is auditable as well. Indeed, to be listed as an TPV, there are several requirements that have to be met. Those who do comply with such strict requirements — by tampering with the code! — can be excluded from the SL Grid, at the Lab's discretion, without even a warning, because whereever the ToS rules, it is absolute :)

We all know about 'fake' TPVs 'pretending' to be legitimate and which allow all sort of malicious things — and which, inevitably, be banned from the SL Grid. While one may argue that malicious TPVs connect to the grid for a period of time before LL bans them, it's a fact that LL to prevent that from happening, and, as quickly as they are able to, they close the doors to the culprits. That is a "reasonable" response which would please any court of law. It's like saying, "look, I always lock the doors of the bank, and have a security system installed with cameras and connections to security providers, and all secret documents from my clients are hidden in a safe, of which only I know the combination, and there are always guards at the premises — nevertheless, the thieves managed to go through all that, bribing a guard, and blowing up the safe with dynamite during a heavy thunderstorm, so nobody noticed the noise." Even if those clients would sue the bank for negligence in protecting the most secret documents of their clients, they would be acquitted, by showing all the precautions they have taken to avoid being robbed. Alas!... Thieves, sometimes, are cleverer (or have access to tools, bribes, etc. which cannot be foreseen). All the bank can do is follow all established business practices in safeguarding access to the vaults, but... there is no security solution that works 100% of the time.

Linden Lab is in the same position. Sure, we know that there continue to be malicious TPVs available for download from "certain" websites (which will be up one day and disappear the next), and that there are a few malicious users downloading them, connecting to the SL Grid, and violating the ToS in every possible way. But they won't be able to do it for long. As soon as a new malicious TPV is released and tested, LL's security team starts their internal procedures to validate that specific TPV, try to understand how to correctly identify it, and block it as soon as possible. That's all we can demand from LL; there will be malicious actors which will able to get an edge over LL's own teams, and still get their viewers to connect. They're just not many, and they can't log in forever; as time passes, it becomes more and more difficult to 'crack' into LL's grid and bypassing all security, and many malicious developers have just given up, defeated, because the time and cost of constantly developing ways to bypass LL's security grows and grows, and the results are probably not worthy all that effort.

There is, however, a catch. The above assumption is that malicious TPVs (or 'bots) can, effectively, . Now this is — you can only retrieve what the SL Grid allows you to retrieve, and such private data even comes close to the grid (it's stored on servers unconnected to the grid). There are no bugs to exploit, because the data is not there in the first place.

It would be like claiming that you could devise a malicious TPV viewer that, once you log in to the grid, enables you to steal from the bank accounts of any resident who happens to be in your field of sight. But that's impossible, because the information about payment methods is on the grid. Griefers and black hat hackers can that their viewer has such a mechanism built into it, but such claims are simply to fool the gullible — there is simply to capture data that doesn't exist in the first place!

Granted, it is that malicious TPVs be able to induce its users to enter payment data into it, which would then be forwarded to the criminals. Again, this is something that be boasted about, but very likely untrue. And the reason is simple: . What you , instead, is to enter your payment data , which the viewer can then tap into to emit a L$ buy order. But the data leaves that special LL webserver — which is connected to the grid anyway. What is being claimed, therefore, is that a malicious TPV includes a fake link to a fake page which looks exactly like LL's own page, and then use simple phishing techniques to lure innocent victims into typing their payment data there.

But that will work for the person actually using the malicious TPV. data might be compromised. But not .

Thus, in the scenario of the 'bot copying text to Discord, or some place else, where neither the ToS applies, nor (potentially) the GDPR, is an impossible one. The 'bot cannot copy data that isn't available. There is not even a requirement to give special permission to such a 'bot to get access to your private data, because . All it access is what a legitimate viewer can, and that means what can be captured from the stream of communication data between the viewer and the server. 'Bots can only "see" what a normal, regular user can "see" in their own viewers; not "more". There is no "more" to see!

That's why I'm often so confused about any claims that "LL is violating the GDPR by allowing 'bots to extract private data from their users". They are — such data is stored on the grid, so even the most devious hacker could not retrieve what is not there.

It's not as if every avatar carries with them a magic notecard where all private data is stored, and, somehow, a 'bot is able to extract that notecard from inventory, and tell nobody about it. While the whole notion of "opening someone else's inventory" is also impossible, even if it weren't, , nor any kind of asset, visible or otherwise, where real-world, private data is being stored, in any possible way.

It is also argued often that 'bots can, thanks to their magic properties, know the IP addresses of other users, and, through those addresses, easily figure out their RL locations. Again, that's a fallacy, because the SL Grid communication protocol does include IP addresses from . Sure, you get your own IP address, and that from whatever grid servers (simulators and asset servers) have been contacted by 'bot or viewer. But nothing is being transmitted from users.

It is conceivable that LL does, indeed, store such IP addresses for their own statistical purposes (after all, the ToS claims so, and we've agreed to it). But one thing is having malicious crackers attacking the statistics servers and access LL's databases — i.e., LL would be victim of a serious cybercrime. The other thing is claiming that, somehow, such information can be retrieved magically through a 'bot. It cannot. That data doesn't exist on the grid, and, as such, cannot be retrieved by a 'bot.

We all know how it's possible to capture people's IP addresses using media-on-a-prim, or by using a media stream, or even by asking innocently for a gullible user to click on a website somewhere, where the malicious actors can retrieve both the IP address and the UUID (and name) of the avatar. That's all true and correct. . All they do is explain that, by opening any such links, you're not on the SL Grid any longer, nor on any of LL's own servers, and, as such, ToS doesn't apply. And they made it so that all these options are opt-in — you cannot be to click on a link you don't want. That's as far as LL's responsibility goes — duly informing users that whatever they connect to outside SL is protected under the ToS.

So, what exactly that secret private data that somehow is being captured by other residents and their malicious 'bots, which they retrieve, and which would possibly fall under the auspices of the GDPR (or even other laws)?

Well, all I can think of — based on my own circle of acquaintances, reading on the forums, and so forth — is that residents are not aware that your current grid location can be tracked (perhaps not with extreme precision, but it ), but, to the astonishment of many, a 'bot can "know" what attachments you're wearing on your avatar. Oh, and don't mention the mysterious UUID, which anyone can capture in SL!

Aye, indeed, you can track an avatar without their knowing about it. Probably deep inside your fantastic mesh head is a script which periodically 'pings' an external server, telling your location, so that the head's content creator knows what popular areas you're visiting, in order to figure out where to put more ads or something. While I have zero evidence of the existence of such scripts, it's they exist.

A similar technique are visitor counters. They can be placed inside prim, and made invisible; no avatar will know the visitor counter is there, unless they know where to search for them. All major shops — and possibly almost all venues out there — generally want to have an idea about how well-attended their shop or event is, and that means having visitor counters.

Visitor counters do not register merely that "someone is here". They can capture the avatar's precise location, their name, and if they're wearing scripted attachments, for instance. And by cleverly combining arrays of visitor counters, you can get a pretty good idea of how a specific avatar is moving across the grid. Visitor counters are ubiquitous and collect vast amounts of statistics, all the time.

Even better than visitor counters are 'bots. All viewers list the avatars that are in the region — on the mini-map, on the list of nearby people in chat, triggering alerts when some people log in and out near your avatar, and so forth. In order to be able to track all of that, the viewer store a list of avatars in a region, and that means that a 'bot can retrieve that list — usually very quickly. It's actually much less laggy to capture that information via a 'bot (which will cover the whole region in a matter of seconds — and doesn't even need to for that!) than by using arrays of visitor counters spread across a region. 'Bots are just specialised viewers which retrieve that information as easily as a viewer.

The lack of understanding about the kind of tools that are required for the viewer to render scenes also leads to some misinformation. When the statistic/metric 'bots were engaged in a huge scandal in 2023, one of the things that shocked residents most was that 'bots could know what you're wearing! Why — obviously! How else would be able to render avatar, if they didn't know what is being attached? Obviously, this is the kind of thing that never crossed people's minds — how does the viewer how to render scenes? The most obvious things are somehow rarely mentioned. While it's true that the viewer can only retrieve certain information (and not more!), it's also true that, , a viewer some information.

That also reminds me of how people were accused of "stealing full-rez textures" — presumably using a malicious viewer and/or a 'bot and/or some kind of weird device — when they were told that nobody to "steal" textures. . Obviously! How could the viewer how to apply textures to some object or mesh, if it didn't have the textures for that? And, naturally, if you want your object to be rendered with high-rez textures, — and archived for later re-use. You just need to go through your cache and pick the textures you want, in the fullest possible resolution. !

And if a texture is just another file in your disk... why, then you can copy it and re-upload it as your own. It's sad, but that's what is known as the "analogue hole": if you can see something in computer, it means it has been downloaded to it, in some way — and if you know where it is stored, you can make a copy.

I remember being argued that textures should be stored on disk, , from where they can be so easily "pirated". Instead, they should be encrypted with a password that nobody knows but... but whom? Because, ultimately, whatever is encrypted on disk, has to be decrypted to be sent to the texture memory; that means that whatever encryption mechanism (and key) are used to encrypt the textures, the reverse process has to be in place , since you cannot an texture. You need the decrypted version for that. And that decryption happens "somewhere" — in memory. So you can always write some application which retrieves the texture from memory as from disk — encrypted or not, since, at some moment in the rendering pipeline, all textures be in place, and that means that they must have been decrypted first...

Last but not least... UUIDs. It is often claimed that UUIDs are some sort of "magic private number" identifying and avatar, and that collecting those is "a privacy issue".

Well, it be, if such a key were, in fact, a "secret" of some sort. But regarding asset content & storage, an UUID is just an entry in a table — it references a certain bit of content. Every item in SL must therefore have an UUID, and anything that requires addressing such content will need its UUID.

What that means is that the UUID is some sort of secret that must be hidden at all costs. Rather, almost all LSL scripts operate upon UUIDs; you can get other avatars' UUIDs either through scripting but also via many TPVs which list them with the avatar's profile. The "panic of 2007" when residents learned that people were — gasp! — lists of UUIDs for nefarious purposes is long gone (hint: the "nefarious purpose" was actually a simple but easy way to spam avatars with unsolicited IMs; LL put an end to that mostly by limiting how many IMs can be sent per object per unit of time; nobody is saying that being spammed is a welcome thing, and I'm obviously glad we don't get any more such spam, but it's hardly a terrible conspiracy to "steal" our most closely guarded secrets).

What, exactly, are the "privacy issues" that you always mention — on this feedback issue as well as on others?

Are you merely confusing copyright/authorship issues — which require explicit consent to be given! — with privacy issues, or are you talking about something completely different?

Since the target here seems to be Discord, what exactly are you afraid that Discord finds out about in-world group chat that would be a violation of your privacy or of anybody else's?

What can 'bots find that are privacy concerns and which, for some reason, the SL Viewers cannot? (Name one!)

While your concern for LL's legality status is admirable — nobody wants LL to be the victim of endless lawsuits in different courts — why do you presume that is compliant with the GDPR? Note that this is not the same as saying that individual users, through their activity in SL, might be or might not be compliant. In other words, where is LL essentially benefitting (financially or otherwise) by deliberately ignoring some circumstances where the GDPR would apply, or, through their inaction, appear to be condoning such circumstances?

One thing is "demanding" that a group chat having a 'bot that exports text to Discord or other platforms has, indeed, an opt-in checkbox, and a space for a link or for some text (not unlike the Estate Covenants) which clearly identifies, as you so well put it, "what is used, shared, for what and with a mandatory opt-in + related TOS". Of those things, only one is obvious, i.e. is shared, namely, (there is nothing else to be shared). To be fully compliant, the 'bot-enabled group clearly state what happens to the text after it leaves the SL Grid via the 'bot.

Granted, one might argue that a few things are possible...

In other words...

Instead, the challenge is for LL's legal team to make an over-arching statement that LL is aware of the many national and international laws regarding safety, privacy, copyright/authorship, minor protection and so forth, and will endeavour to comply with all of them (so long as they aren't contradictory!) and to give residents the required tools so that they, in turn, are also compliant with such laws. And then, of course, each case is a case, and up to LL to "decide" what their policy will be regarding issues raised now and the future. They already use that approach for a lot of things, the latest of which having to do with the ever-conflicting issue of age play vs. paedophilia... the first being "allowed" or at least "tolerated", the second being obviously absolutely forbidden (since it's a crime in most of the world).

... the reason why Google, Apple, Microsoft, Facebook & friends are under the EU's scrutiny is because they never felt the necessity of telling their users what real-world data is being collected by them and what is being made with it — namely, profiteering from selling such data

... Facebook's desire to enter the Metaverse bandwagon was because they were angry at Google and Apple, which, in an effort to remain GDPR-compliant, required all apps on their stores to disclose what data they collected and what they did with it, something that Facebook did want to do, and accused the two companies of acting as a "duopoly" by imposing their own rules over hardware only controlled by . Thus, Facebook wanted to sell their own device — the Oculus — where they would implement rules instead. Suffice to say that this was enough for the EU to start taking a look at what Facebook does on the Web...

... I still wonder how exactly Google, Microsoft, Yahoo and a few others "get away" without disclosing that they thoroughly comb through their users' mailboxes in order to extract profiling data, which gets resold, or at least used for providing better-targeted ads to their own ad agencies. Sure, recently, I have seen some popups from Google, explaining that we have to give them explicit permission to track us across all their platforms, but their explanations about what exactly they are not described in what I would consider to be "clear" language. Apple do the same as the others, but they usually escape being noticed, because they don't sell or give away their users' data to (and Apple, as a company, is liable to bribery; they have way too much cash in the bank, besides their market value in shares, obviously). That doesn't mean that Apple doesn't use that data — regardless of they do with the data, they clearly state data is being collected and for what purpose (which they only do... when a gun is pointed directly to their heart, I mean, wallet)

But I digress. The point is that it's not fair to compare LL with Big Tech, which play by their own rules, unless they're curbed by legislation. LL comply with pretty much every legislation out there :) and does so willingly, sometimes erring on the side of caution, just to be absolutely sure they cannot be seen as deliberately trespassing upon any laws, willingly or not.

That said — I think it would be fair to give a more detailed description of the immensity of data that is being sent by every viewer to LL's servers, which LL vaguely refers to as "metrics" or "telemetry data". It is highly likely that this is, indeed, just what LL does with the data. But I personally don't . I only know that the data being sent . And it's definitely quite a . To the best of my knowledge, while I have gladly accepted all ToS ever created by LL since mid-2004, I don't exactly recall anything specifically mentioning what data they collect from my computer and what they do with such data. Probably it's buried deep in their documentation and it's only my fault for being lazy and not having dug deep enough...

I would recommend a checkbox in the group tab where people can check, "Group chat linked to a third party service." And display that in the group info.