And incorrectly. ChuckPA didn’t describe the data coming from the user. He quite clearly refers to the data coming from upstream, i.e. not the user. What he actually described was a user having x-rated content, user requesting metadata for matching, Plex then going off to outside sources for that data and subsequently caching the result. There was nothing about the cache being built from what the user has input on their server.
No, and they haven’t said anywhere that they are
where did the data originate from lol?
Wtf was the point of having global search settings for our libraries if Plex is going to override our very own set preferences.
If I had questionable content in a library, used ALL the tools currently provided to me to not have any of that content indexed elsewhere locally, then Plex should have had absolutely no part in feeding me other questionable content in my search bar results, especially when i already had indexing of my content turned off globally for the home and in search results on my server.
Ugh, what a cluster******
this is an oxymoron. plex searched for the content because a user initiated it. same same.
Hmmm, this seems pretty clear and does contradict what is said in policy.
Anyway, I’m done on the topic. It is an eff up and the press are picking up on it.
Whichever of the many publicly available metadata databases available on the internet that the Plex agents use.
It really isn’t. There’s a massive difference between sourcing metadata about a file from a user and using a provided filename to obtain public source metadata to pass back to user
ok i’ll present it in a different way. would plex have this information if it didn’t exist on a users server?
i appreciate your conversation. i’ll leave you with the last word. and not clutter the thread anymore.
If there was a film that has not been added to any Plex server ever then obviously not because there never would’ve been a need to find the metadata for a user. But the point you seem to be missing is that the metadata ChuckPA was referring to, and is cached by Plex, comes from upstream sources (i.e. the internet). It’s not populated by anything the user has recorded locally as metadata for the file.
Thanks @ChuckPa
This helps a lot, and judging from the number responses since you posted this, it has given us a lot to chew on. lol… Thanks especially for your forthrightness in spelling out what happened.
First question that popped into my mind was, if we already had the code, and all you had to do was toggle a feature flag, can’t you just toggle it back?
Ahhh, sounds like contractual obligations ($) led to a lot of this, rushing it through by a certain date, inability to roll back to a version prior to all of this, or even disable the “just enabled” feature flag, etc. OK, fine, it is what it is.
I still strongly feel these problems could have been avoided with a formal beta test cycle that included volunteer users outside the company.
I hope we’re talking days for the most common clients, and not weeks! 
Your server sends a request to the metadata back end and performs a search based on certain criteria.
It then returns the information containing the posters, etc…
That information is cached on the back end so if another user performs a similar search it can be served up without querying external sources.
In no way does Plex store that user X has performed a query for media Y. (I think)
On the other hand when searching for media in the search bar then your search terms are sent to a 3rd party. (justwatch) I’m not sure if this can be linked to specific accounts or if it is obfuscated. Thus what seems to be an issue is how this 3rd party querying is effecting our privacy. It would be fairly trivial to correlate this data with a specific person/account and that data could then monetised.
This is my concern anyways. Also, if this is used to provide 3rd party search results (watch this here) it’s an advertisement. If this is the case and Plex is profiting from it, it is morally wrong to subject paying users to this type of scheme. This needs to be CLEARLY stated in the TOS and we need an option to opt-out. Which thankfully it seems to be what will happen. But I and I’m sure many others would like a clear explanation on how this is working and to what degree.
Personally I’m going to leave it on because I really don’t care. But I want to be informed none the less and have the option to make an informed decision on how my data is being worked.
There’s just too much to sift through in these threads so please make a locked pinned post that explains in detail how this is all being worked.
First, enjoy your dinner. Ours will be on the table soon - I know how you feel 
Second, I’m really glad that it will have an off switch soon. I’m on the fence whether this Discover is useful and it actually might be with some polishing, but options are good - correction, vital.
Third, this is not personal so please don’t take it as such. But for heaven’s sake, PLEASE don’t ever do this to us again - that is, shove a half baked and downright unwanted feature down the throats of the entire userbase. Put it through an opt-in beta for the Plex pass users who can iron out the kinks if they so desire. Then make it an opt in feature for the masses. You know, like you used to do with (potentially) exciting new features. Honestly, I think you would have received nothing but glowing reviews if you had done it that way.
Folks. Plex is behaving like an old style horse trader. Once on NCIS Gibbs described a horse trade as: “You pick the best nag in the barn and work the deal until it falls through then you can get the second best nag for a song.”
Plex is doing the same thing as they have done before with intrusive features. They are intentionally putting out features that are way too intrusive for most users. Then they will back off those feature just a little until they end up with the feature set they wanted in the first place. It is a form of “bait and switch” but sneakier.
I do not buy that it was “accidental” as Plex has never show any tendency to having accidents.
There are ways to lie without lying and Plex is using the “lie” to get more and more user info on the Plex servers.
It is nothing but a grab for more power and money and, when Plex backs off just a little, Plex will get exactly what they have wanted all along; unfettered access to user search data so they can sell it to others. In fact, from some Plex posts, it seems that the data is already sold and Plex is trying to justify the enslavement of Plex servers world wide.
Plex has become quite corrupt in that they use half truths to lie to their users.
If Plex actually fixes this I will be more than surprised. Plex just wants to make money. Making money is a good thing for a business but Plex is making more and more money by selling more and more user data and then telling us it is for our own good. That is Plex is exhibiting many of the worst characteristics of the worst of the businesses in the world.
I will reinstall Plex (I am quite sure that Plex does not really care if I do or not) if I ever feel that my data is pretty safe but, until then, Plex will make me and my data their slave no longer.
All the excuses that people post like other companies that steal data does not matter at all. Just because some companies are thiefs does not mean that it is OK for others to lie cheat and steal. It just means that they are joining a larger set of crooks.
Quote - “Never ascribe to malice that which is adequately explained by incompetence”
Can we at least have the IP address to block Discover? If your investors are forcing this down our throat, it would be nice to block it from our end.
Is this a real, legal opt-out or just hiding? I want to make sure because yesterday there were some plex, inc employees who didn’t understand what opt-out means.
They’ve had open calls for feature testing for loads of other things that wasn’t as secretive. Examples like the DVR epg and guide data or even say the improvements to the Plex Transcoder, the new Movies and TV Plex agents as well as the most recent testing of what will be a new Music Agent.
All of those were good signs of Plex honoring that until they start working on more secretive ventures. At that point, I would have testers sign an NDA so developments are not leaked.
It’s not really idle though, and frankly I think this comment as a whole was in poor test. If I’m being candid, I think you’re venting a bit here yourself.
I’ve said it myself - with Discover in place without controls I can’t use Plex. My kids movies and shows are already on Jellyfin.
I think Plex does a better job with sharing to friends and family, and I’ve got a ton of home movies I’ve ripped and cleaned up over the years, dating back to VHS my from my dad in the 80’s. But while searching for something in there, my mom will get a ton of completely unrelated content.
So hey - how about take the feedback for what it is:
We want to keep using Plex, but Plex is making that hard. There’s enough bandaids to keep using it for now, but if that doesn’t last, if Plex refuses to allow for the use case it’s had for 14ish years, well, Plex doesn’t really make sense for those users anymore.
Tentatively happy about this.
Thanks for trying to get on top of this. I don’t envy you.
When you get a chance - and assuming you haven’t been instructed that there are to be no public commitments on the subject -, some further clarity on what the “off switch” will actually do would be appreciated. Forget the specifics of the UI for now - I think what a lot of us would like to know is, if we opt-out at our account level (or however it ends up being implemented), ie. we turn this and other online features off and opt to only access content on a personal server, what happens with our search terms? Will they only be used locally as has traditionally been the case when online features are disabled, or will they now permanently be sent externally regardless of how we configure the controls?
That is, will the off switch/opt-out actually turn it off/opt the user out of the function or will it simply hide the results?
You may or may not think it’s important but evidently it is important to many of the customers who’ve been posting about this here, and the few Plex employees who’ve put their heads above the parapet so far have been dancing around the question quite conspicuously, which hasn’t gone unnoticed.
When I know more, I will share.
Please be patient. I know it’s hard but a lot of good communicating has been done.
Let’s build on that and give the teams time to work their magic.
Thanks ChuckPa.
