Feature Request: Function to ‘report a post’

2025-01-21 13:48:08

adminsforum@hubzilla.org

Der Pepe (Hubzilla) ⁂ ⚝ wrote the following post Tue, 21 Jan 2025 13:48:41 +0100

Feature Request: Function to ‘report a post’

Not only because of the UK online safety act, but also because of comparable (albeit not as restrictive) laws in other countries and also because users generally demand the ability to easily respond to content that is supposedly or actually ‘forbidden’, I think it is ESSENTIAL that Hubzilla provides an easy way to report such content.

I would suggest adding this function to the post/comment menu (at the bottom right ⋮ of the post). A menu entry ‘Report post’ should be added there.

The function should be: A small text field opens in which the reporter can briefly describe their concern, and there should be a button to submit it. The admin of the hub receives an email (or a DM or a notification) in which the reporter's message and the local link (i.e. on their own hub) are listed. If he follows the link, he can check the post and has the option of deleting the post from his own hub or taking other measures after checking it, if necessary... or nothing at all if the post is in good order.

I consider the function to be very important (high priority), also to protect the Hubzilla project against accusations that there is no way to complain, and also to comply with the legal regulations in numerous countries.

@Mario Vavti

show all 24 comments

2025-01-21 14:20:33

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

Ideally, Hubzilla shouldn't re-invent the wheel in this case and become incompatible with everything else.

Is there something in the ActivityPub spec about reporting content? Do the various microblogging projects in the Fediverse have it implemented? If so, then so should Hubzilla (and, by the way, so should (streams) and Forte).

But even if reporting is proprietary, non-standard Mastodon technology which almost everything that's based on ActivityPub has taken over along with a whole lot of other Mastodon stuff, even then Hubzilla should take a look at it.

I mean, what are Mastodon users supposed to do if they want to report a Hubzilla user to Hubzilla admins? They don't even know they're dealing with someone on Hubzilla because nothing tells them. Thus, you can't expect them to first look up the admin of the hub in question on /siteinfo/json and then try to establish a bidirectional connection with the admin.

2025-01-21 16:02:31

Der Pepe (Hubzilla) ⁂ ⚝

pepecyb@hub.hubzilla.hu

@Jupiter Rowland I think you have misunderstood my FR a little. I'm not talking about Mastodon users reporting Hubzilla posts on their instance. I'm talking about being able to report content from a hub to the admin on that same hub. If, for example, a Hubzilla post appears in a Mastodon thread on Mastodon that a Mastodon user considers worthy of reporting, then they have to access the post, land on the hub and the report menu is available. The hub admin can then delete the post from their hub if necessary. Whether and how the admin of the Mastodon server does this on his instance is none of our business. That's their problem. They have to deal with it themselves.

2025-01-21 22:43:34

Scott M. Stolz

scott@loves.tech

At the very least, Hubzilla users need to be able to report content to their own Hubzilla Admin. In other words, users on example.com need to be able to report posts to example.com's administrator.

And in some jurisdictions, there needs to be a way for a third party to report public posts to the administrators, via email, an online form, or some other method.

None of this has to be communicated over Zot6 or ActivityPub.

2025-01-21 22:51:39

dynamic_hubzilla

dynamic@hub.netzgemeinde.eu

@Jupiter Rowland

Jupiter writes:

"Ideally, Hubzilla shouldn't re-invent the wheel in this case and become incompatible with everything else."

I think this is a really important point. It's not about building features for Mastodon users, it's about making reasonable efforts to facilitate compatibility with other platforms. Surely, all else held equal, being compatible is more desirable than... not being compatible.

2025-01-21 22:54:40

Scott M. Stolz

scott@loves.tech

I would suggest that the report function be available to people logged in locally and via OpenWebAuth. That should cover the requirement that users can report content to the administrators.

Contact information for the site can be placed on the Site Info page, if the server operator is required to post it in their jurisdiction.

2025-01-21 23:15:08

Scott M. Stolz

scott@loves.tech

@dynamic_hubzilla

...it's about making reasonable efforts to facilitate compatibility with other platforms.

There are two aspects of this:

1. Minimum requirements for complying with various laws.
2. Making this work over federation.

Ideally we have both, but at the bare minimum, users need to be able to report content residing on the server to their own administrator, and the administrator needs to have the option of hiding the content or deleting the content from the server permanently in such a way where it can't be reimported.

2025-01-21 23:43:49

dynamic_hubzilla

dynamic@hub.netzgemeinde.eu

@Scott M. Stolz @Jupiter Rowland

"Ideally we have both, but at the bare minimum, users need to be able to report content residing on the server to their own administrator, and the administrator needs to have the option of hiding the content or deleting the content from the server permanently in such a way where it can't be reimported. "

Of course.

But we lack information on whether an ActivityPub standard exists nor whether it would be harder to implement Hubzilla reporting in a manner that adheres to such a standard if it exists. It might even make our lives easier. This wouldn't necessarily be for federation, just setting internal reporting up in a way that would be consistent with future federation.

It seems like a reasonable thing to look into, particularly as it might even save some work for the Hubzilla development team.

2025-01-21 23:48:47

Scott M. Stolz

scott@loves.tech

@dynamic_hubzilla From a programming standpoint, the reports get put into the database, and then are displayed somehow so the administrators can take action. There can be multiple ways for reports to get into the database. So it is entirely possible to implement local reporting, and then add federated reporting later.

2025-01-21 23:49:29

Harald Eilertsen

harald@hub.volse.no

A few thoughts pop into my mind about this subject:

If there is some sort of standard way to report content across sites, I don't see any reason for why we should not use it. Even if there isn't it's well worth having a look at other implementation, and making an effort at being compatible.
Reporting content cross-site and internally on one site don't necessarily need to be treated as the same thing.
It must be possible for an admin to delegate the moderation responsibilities to other trusted users on the hub. Preferably even to a group of users. One possible solution is to make use of the existing forum functionailty in Hubzilla. Just have a special forum that only moderators have access too. (With that solution it could even be possible to have a common moderation forum for multiple mutually trusted hubs if we want to.)

2025-01-21 23:56:33

Der Pepe (Hubzilla) ⁂ ⚝

pepecyb@hub.hubzilla.hu

My thoughts on this:

It is important that authenticated Hubzilla users get the option via the menu... not just those who have an account at the hub of the original posting.

Regarding the contact option for outsiders: If I view a (public) Hubzilla posting without being logged in anywhere, then the post menu (⋮) only contains the entry ‘Link to source’. Would it be possible to also offer guests an entry ‘Link to administrator info’ in addition to this, which then leads to a page on the home hub where you can find contact options (actually a link to /siteinfo would suffice)?

I think that with the combination of a ‘report function’ for authenticated users and a link to the admin (siteinfo/), everything that is necessary would be done.

I don't think we have to worry about anything more. Everyone can really ‘complain’ about that, and we shouldn't overdo it with the functionality either. And I hope that this also keeps the effort within limits.

2025-01-22 00:53:13

Der Pepe (Hubzilla) ⁂ ⚝

pepecyb@hub.hubzilla.hu

And again: please don't overdo it. The basic functionality that only authenticated users (including visitors) can report a post is perfectly sufficient. Maybe we should also decide to really adopt /siteinfo as the default into the app menu or into the main menu, so that everyone really has the opportunity to get information on how to contact the operator of the hub. Whether the site info is then filled with meaningful and usable information is the responsibility of the operator. If you don't provide any information, you may run into problems... but that's your own choice. Just like omitting certain information such as the community rules, terms of use, etc.

2025-01-22 09:30:16

Scott M. Stolz

scott@loves.tech

For authenticated users, it can have a "Report Post" link that sends that post to the administrator.

For unauthenticated visitors, the link can still say "Report Post" but should take you to /siteinfo page instead. But if the Administrator information is not filled out in /admin/site then the link to /siteinfo should not appear (because no contact information has been provided).

Perhaps there could be a check box which toggles the report post feature for unauthenticated visitors on or off, since single user instances don't need it and probably did not fill out the administrator information anyway.

2025-01-22 11:22:21

Der Pepe (Hubzilla) ⁂ ⚝

pepecyb@hub.hubzilla.hu

Perfect, @Scott M. Stolz !

That's exactly what we need, no more, no less.

2025-01-26 11:58:37

citc

citc@zotum.net

Without identify verification, public content removal becomes a tedious game of "whack-a-mole". Amateur administrators could satisfy themselves that they have done something superficial, but law enforcement notification could not occur.
It is also not understood how to manage public content where hubzilla clones exist.
Would a hubzilla administrator for clone A be notified of undesirable content on the source hubzilla account server?

2025-01-26 13:08:09

Der Pepe (Hubzilla) ⁂ ⚝

pepecyb@hub.hubzilla.hu

@citc

It is not the job of a hub admin to verify or establish someone's identity... and in most countries it is neither possible nor legally permissible for a hub admin to do so. An admin is not a law enforcement institution.

And an administrator is not a reporting (denunciation) institution. If there is suspicion of a really serious crime, depending on the local jurisdiction, there may be an obligation to report it to the law enforcement authorities and also to secure the relevant information about the post, which would then enable the relevant authority to identify the suspect.

Once the post has been removed from view, however, the admin's job is done.

Regarding the question about clones, I will test something similar.

2025-01-26 13:50:16

Der Pepe (Hubzilla) ⁂ ⚝

pepecyb@hub.hubzilla.hu

I just tried the following with a test channel:

The test channel is primarily on klacker.org
The test channel is connected to my main channel Der Pepe (Hubzilla) ⁂ ⚝
I cloned the test channel on hubzilla.de
I used the primary test channel on klacker.org to make a test posting
I posted a comment on the posting with the primary test channel on klacker.org, which is said to be ‘objectionable’
I posted another comment on the posting with the primary test channel on klacker.org, which is not too ’objectionable’
As admin of the primary instance (The Smoking GNU ⁂ ⚝), I viewed the thread in the public post stream
I then used the ‘admin delete’ function to delete the ‘objectionable’ comment as the admin of the channel (The Smoking GNU ⁂ ⚝)
The ‘bad’ comment can now be seen neither in the public stream nor in the streams of the channels connected to the test channel. It seems to be gone... even in the clone, where it can no longer be seen either.

2025-01-26 14:33:10

citc

citc@zotum.net

According to comments elsewhere, removal of content from public view may not be sufficient. If content remains in a database, the administrator has an option to delete the content from the database. Are administrators supposed to keep content for a suitable period of time,("statute of limitation" perhaps?) in case law enforcement seek evidence at a later date?

Impressive that clone content removal is successful; if content is shared/boosted/whatever elsewhere in fediverse, does that content also get removed? Is there an audit trail somewhere about those sharing/boosting/whatevering (such people may be liable for distribution of undesirable content)?

2025-01-26 15:19:41

Der Pepe (Hubzilla) ⁂ ⚝

pepecyb@hub.hubzilla.hu

The legal obligations for the storage of ‘evidence’ vary greatly from country to country. The same applies to the retention periods for ‘communication data’. Each instance operator must check this for their own jurisdiction and observe it accordingly.

Another experiment showed that content deleted by the instance admin is not automatically removed on other platforms (test was carried out with Mastodon), for example. In this case, the respective person in charge of the other instance is obliged to do so. However, since the admin of the initial instance does not know where the post has been federated, he cannot give any information to other instance operators. But that is not his legal obligation. Morally, it would probably be appropriate. But since it is not possible, nothing can be done about it.

2025-01-26 15:30:08

citc

citc@zotum.net

If administrators do not know the locations of their users, compliance with the various laws would seem very difficult. There looks to be a risk: "administrator obeys laws is his/her country, but may break law of user X in user X country", Who is priority?

It would be interesting to know how certain laws (and lawyers!) interpret the concept of federation. If other fediverse servers must also remove undesirable content, those administrators must have some system to be notified about bad content.

2025-01-26 16:57:33

Der Pepe (Hubzilla) ⁂ ⚝

pepecyb@hub.hubzilla.hu

This is not only a problem of decentralised services, but a problem of any service that is globally available and usable... in principle a problem of the internet and the principle of the internet. A liability can actually only be claimed under the laws of the place of operation, if common sense is applied.

Should a service be in violation of the laws of another country, then the operator could prohibit the use of the service by users in that country (although in practice this would be impossible to monitor). This would also be limited to active use, but not to passive use by visitors. Otherwise, the only option left for those responsible in the affected country would be to somehow block the service or its use (active and passive) in their own country (we all know that this is easy to get around). It's simply an unsolvable problem. And it is also a very presumptuous view to demand that actors in a foreign country must comply with the laws of our own country when they act in that foreign country. They don't have to. What is permitted and prohibited for them is what they do to themselves.

2025-01-26 17:47:12

Tejan Ausland

tejan@hubzilla.monster

That's what the problem is. Too many countries wants to regulate social media and control your speech, and they are writing extra-territorial laws to do so.

The reality is that you must follow the laws in your jurisdiction, regardless of what foreign laws say. Unless you have a physical presence in that other country, they have no way of enforcing their law on you, especially if laws in your country conflict with their laws.

For example, you may encounter a situation where a country ruled by a dictatorship demands that you censor certain political groups or political thought (though crimes), but your country protects free speech. The dictatorship cannot use your country's courts to force you to comply with the dictatorship's laws since the dictator's laws violate your country's laws. They may have a kangaroo court that convicts you, but how would they execute their judgement if you never go to their country? All they can do is block your website, if they are technically capable of doing so. Although, it should be noted that there are some rogue countries that will actually kidnap you if you piss them off enough. Highly unlikely, but it has happened.

The only exception is if the two countries have a treaty. or their laws are similar. For example, if both countries require you to have a privacy policy, and you don't have one, they might not be able to enforce their laws on you, but they can contact your country and have your country enforce those laws on you. Or their laws require a report function, and your country requires a privacy policy. They can't enforce the report function on you, but they can enforce the privacy policy on you. You're still getting fined.

That being said, if you can comply with an ethical law, you should try to do so. Even if their legal basis of enforcing it is shaky, it still would be a huge hassle if you don't comply. Remember, we are dealing with powerful people backed by the government, who have guns and a monopoly on use of force. And they probably have "friends" in your country who can find something else you are in violation of.

So refusing to comply is at your own risk.

2025-01-26 19:38:35

Frank Casa

frank@frank.casa

@Tejan Ausland It is generally a good idea to not piss off people who can send men with guns after you. Or freeze your bank accounts.

2025-01-26 22:22:50

Rogue Runner

cisco@inlakech.site

It would also be a good idea to not fall into complete compliance with every centralized orwellian law that arrives by decree... lest you find yourself in a thread such as this one, and fall into complete compliance of laws that mostly everyone agrees is draconian overreach. The internet does not belong to any government, and should not be used for law enforcement... it's an information platform, and not a tool for enforcement. Let community enforce their moral etiquette... living in fear is how you got here.

2025-01-26 22:33:38

Scott M. Stolz

scott@loves.tech

Regardless of various laws in various jurisdictions, it generally is a good idea to have a report post function, especially if you have people on your server who you do not know well or run a hub that allows public signups.

Anything that violates our Terms of Service and Community Rules has to be dealt with, and the earlier we know about it, the better.

If a dictatorship in another country comes along and requests us to censor their political opponents, we don't have to comply. But if someone is harassing or threatening someone else, we want to know about it so we can protect our users.