Datamol

Fri, 12 Apr 2019 01:52:54 -0700

!Hubzilla Support Forum
today I came content acroos that was shared via hubzilla that was nothing less then a right wing terroist blackmail
This brings back into my mind, that I want a moderation function similiar to how it is in mastodon.

There I could just click on report content to moderators, and they would take care of.

show all 35 comments

Haakon Meland Eriksen (Parlementum)

Fri, 12 Apr 2019 04:10:16 -0700

Hi, @paulfree14 It may be your wording, and I do not intend to be harsh, but your description says unwanted content 'was shared via hubzilla', but that must be tied to how you use it, as this is not possible any other way.

1) With each of your channels in Hubzilla you decide who its connections are, not the hubmin, so my first suggestion is to stop connecting your channel to people or forums with content you do not care for?

Perhaps you need to tighten how you connect or your permissions in your channel settings.

Note: it may be that you have additional protocol plugins on in your channel, i.e. not Zot, like Diaspora and ActivityPub, which may be more permissive than regular Zot, thus letting unwanted content into your stream.

2) Comments can be turned off.

3) Moderation is available in Zap, i.e. channel owner can decide to publish comments or not. I am not sure if this is in Hubzilla. If not, see 1).

Best of luck, I do not care for that kind of content myself.

Mike Macgirvin

Fri, 12 Apr 2019 04:50:04 -0700

It would be relatively easy to write a "nanny" plugin to do this, but we've never needed it here and have a different philosophy. This is a permissions based system. If you allow toxic content, that's your choice. You don't have to and the default settings are to only allow those you want to interact with. If the toxic content is coming through a contact/connection, you might think about your relationship with the intermediary. You can always superblock the author and this is incredibly effective. If the content is coming through the public stream, get rid of that nasty thing. Nobody can protect your from the public stream, but they can certainly waste their life trying.

Haakon Meland Eriksen (Parlementum)

Fri, 12 Apr 2019 05:28:20 -0700

Ah, yes, I forgot about the superblock! Thanks, Mike!

Anmol Sharma

Fri, 12 Apr 2019 06:19:59 -0700

But what is the content sender is admin Himself?

Anmol Sharma

Fri, 12 Apr 2019 06:26:58 -0700

I usually tag NSF* to bad talks content and Superblock toxic content.

Tanja

Fri, 12 Apr 2019 06:39:17 -0700

But what is the content sender is admin Himself?

In that case, I'd suggest moving to another server...

hEARt PhoniX

Fri, 12 Apr 2019 06:48:26 -0700

@Anmol Sharma In that case I'd review my choice of admin and make use of this awesome thing called Nomadic Identity. Ok, it's easy for myself, I chose consciously to be Zot only. Moving hubs may be more complicated for anyone using non-Zot connections. In that case you will need to take more actions yourself. I don't know, but I'd believe unfolowing your connections, using Zot to clone and move your channel, declare it primary, delete the old channel, reconnect to your connections.
If there is an easier way or I missed some steps or I'd do it all wrong, please add.

Giacomino

Fri, 12 Apr 2019 07:12:00 -0700

what bothers him is:
can not report a bad person to the administrator, can not play his role of "social justice," said in other words, and not the fact that these bad posts come in its flow (because for that in hubzilla we know that the problem it does not exist because hz is the best in this sense to decide what appears and what does not appear in our flow).

Haakon Meland Eriksen (Parlementum)

Fri, 12 Apr 2019 07:50:51 -0700

You can report to the administrator of the hub. It is required by law in a number of countries. In Hubzilla you find the hubmin by visiting /siteinfo . If you cannot find the hubmin's contact details there, then I suggest cloning your channel elsewhere and leave the hub as necessary.

Mike Macgirvin

Fri, 12 Apr 2019 13:58:44 -0700

I'm quite happy if somebody wants to build a content reporting addon. People coming from permissive networks do expect this kind of thing. And in the wake of the NZ shootings we're now expected to automatically take down content within minutes or face jail, even if we're fast asleep. I'm just saddened that it has come to this. It's not like these are new problems. We've had online spaces for decades now and those that deal with controversial topics have come to understand that you have to keep the door closed and only let your friends in.

It doesn't work the other way around. You can't let everybody in and be forced to confront the toxic players every time somebody gets out of line. What that does is increase the toxicity and it also increases the number of toxic players streaming through your front door to defend their friend and add to the harassment. It really creates a vicious circle.

Tanja

Fri, 12 Apr 2019 15:16:19 -0700

And in the wake of the NZ shootings we're now expected to automatically take down content within minutes or face jail, even if we're fast asleep.

The EU is in the process of passing similar legislation, and this so shortly after the copyright "reform".

I've been thinking for months that the best possible future for social media is a large network of thousands of small instances. If you're on your brother's server, or your best friend's, you're a lot more likely to behave, and if you're not, it's up to your brother or best friend to maybe rethink their relationship with you.

I've tried to write an accessible guide to running your own instance (originally based on Friendica, but if I publish it again, it will be for HZ / Zap), but documentation is actually not even the problem - it's convincing people that it's worth the effort.

Waitman Gobble

Fri, 12 Apr 2019 16:21:45 -0700

Who is the moderator for the EU? Should the button report content to them directly?

Marshall Sutherland

Fri, 12 Apr 2019 16:32:57 -0700

I definitely agree that superblock is a wonderful thing. I think I've only every used it 2 or 3 times, but it was exactly what I needed to remove extreme negativity from my online life.

lotfi

Sat, 13 Apr 2019 00:21:21 -0700

I've tried to write an accessible guide to running your own instance (originally based on Friendica, but if I publish it again, it will be for HZ / Zap), but documentation is actually not even the problem - it's convincing people that it's worth the effort

.
If one day you decide to do it for ZAP, know that there is at least one person already interested.

Anmol Sharma

Sat, 13 Apr 2019 00:48:50 -0700

@lotfi
There is #Hubzilla and #Zap package for #Yunohost already. You can do one click install,upgrade,remove,restore and backup.
The Zap package is new, so the package check is not yet done. I will soon update the Zap pacakge to 2.10. I have not tested fail2ban on Zap, but its working on Hubzilla.
Links:
Hubzilla for Yunohost: #^https://github.com/YunoHost-Apps/hubzilla_ynh
Zap for Yunohost: #^https://github.com/YunoHost-Apps/zap_ynh

There is Osada, Pleroma,writefreely,plume and Friendica packages too.
#^https://github.com/YunoHost-Apps/friendica_ynh
#^https://github.com/YunoHost-Apps/pleroma_ynh
#^https://github.com/YunoHost-Apps/plume_ynh
#^https://github.com/YunoHost-Apps/writefreely_ynh

https://github.com/YunoHost-Apps/osada_ynh

Mastodon package is on its way. PR is ready and its getting reviewed.
#^https://github.com/YunoHost-Apps/mastodon_ynh/pull/130

lotfi

Sat, 13 Apr 2019 00:56:39 -0700

Great! Thanks.

❦ CRIME LICH ❦

Sat, 13 Apr 2019 01:33:26 -0700

on my own server, i encourage my users to send me messages about all such users so that i can block server-wide. there isn't an automated reporting system, but i am responsive and so for each of them, the effort of manually reporting has been positively reinforced. if i didn't want to take on that responsibility, i'd have no business opening my hub to others.

paulfree14

Sat, 13 Apr 2019 07:23:10 -0700

thx all for engaging in this topic.

I just want to be clear of not being miss understood:

I am aware that I can use my permission system to choose whatever I want to see and what not. And that is great. I love this about hubzilla.
there is content that I consider as dangerous. If you're among the privilidged class of white cis male, it's likly that you will not be subject of this danger. Facists to just name one group, do use the internet to recruit and to spread their ideology. I want to protect innocent people. Limiting their reach does limit their ability to recruit and to spread their ideology. A report button and a moderation system that get's notified could be a part of countering this. If I see such content and choose to just block personal wide, that's similiar to looking away where inverention would be apropirated rather then to ignore.
I do not have the skills to build it myself

thx for listening

Giacomino

Sat, 13 Apr 2019 09:44:40 -0700

one of the big differences between h-zilla and diaspora or mastodon:
while in masto-spora things there are sites / nodes with, sometimes, thousand and even more people, where it really needs the existence of certain tools, in hubzilla (even in friendica I can say) we actually live, breathe decentralization, in the sense that our nodes receive at most hundreds of people and not thousand and more, and in this sense it is much easier to report to the administrator (and in a personal way) the arrival of a nazi-fascist dickhead.
Unfortunately, many people who arrive in hubzilla have not yet understood that there are great differences (hzilla vs masto-spora-roma......)

paulfree14

Sat, 13 Apr 2019 10:07:07 -0700

I understand @Giacomino, but it is still not solving the issue I'm bringing up.

Imagine that what you see is reminding you on an violent attack that happen to you.
You do not want to have to deal with discussing about any of that matter.
You want that others will take care of such.
Moderators could serve this role, who will, if not clear if content should be removed or not, discuss about it, without that you need to be involved.
Also can they take care to engage with other instance moderators/admins, if harmfull content was spread from there.

❦ CRIME LICH ❦

Sat, 13 Apr 2019 10:59:49 -0700

i, a hub admin, would like this, rather than to have the burden of explanation laid squarely on users' backs.

Giacomino

Sat, 13 Apr 2019 11:24:40 -0700 last edited: Sat, 13 Apr 2019 11:25:57 -0700

has happened in the past, we have eliminated and suspended ISIS accounts for example, but all done in a fairly simple way, direct communication with administrators, without the need for tools, exactly for the reason described above.

paulfree14

Sat, 13 Apr 2019 11:31:08 -0700

ok, that's good!
(thought it still would be usefull having it)

tonnerkiller

Sat, 13 Apr 2019 11:42:19 -0700

I just wonder: Did those ISIS idiots not run their own server?

❦ CRIME LICH ❦

Sat, 13 Apr 2019 11:56:04 -0700

there were fewer hubs and fewer users in the past, and how many of those specifically serving vulnerable marginalized communities?

Giacomino

Sat, 13 Apr 2019 13:45:11 -0700

everything can be useful, but I repeat, when you have just a hundred people or even double, at most, inside 1 hub, it becomes easier to control users.
...when you have 15000 people inside a node it is different.

❦ CRIME LICH ❦

Sat, 13 Apr 2019 14:42:36 -0700

my small hub still interacts with potentially hundreds of other users and i’d appreciate it if my users could hit a button to bring something to my attention rather than write up a full report.

Mike Macgirvin

Sat, 13 Apr 2019 15:53:29 -0700

If anybody wants to work on this, the primary thing that requires solving is that we know who owns the admin 'account', but we don't necessarily know which of their (potentially many) channels is the admin 'channel'. We also cannot just send them private mail or a private post because the permissions of their channel might not allow this.

So the choices are to send them an actual email (which is attached to the account) or provide a new notification class which is seen by any/all admin channels. For Zot6 and ActivityPub compatibility these should probably generate valid Flag or Block activities per the ActivityStreams vocabulary. Diaspora has no such mechanism.

Eventually it may be desirable to be able to receive these over federated networks from other sites (remote flagging - although this is a slippery slope). I don't know how other networks deal with this specifically but the addressee in this case should probably be the site itself (e.g. the baseurl). But check with the Mastodon or Pleroma devs.

Most of you are aware by now that I'm doing very little work on Hubzilla these days so you are in control of your own destiny. I will most certainly not be working on this because I am opposed to any form of centralised censorship. This is why I built a permissions based system where you are personally in control of your stream. Giving this authority to others is against everything I believe in - and eventually leads to tyrannical control structures that have total power over what information you are allowed to see. Always.

However, it's your software too and you are free to make different decisions to align with your own principles and requirements. As long as this doesn't interfere with mine, we're cool.

Marshall Sutherland

Sat, 13 Apr 2019 15:54:04 -0700

To run with the ISIS example, lets say they did run their own hub and ran it on their own hardware in ISIS-controlled territory, what is it you expect someone to be able to do? As a hub owner (and I'll pretend for the moment that I have more than 2 users), if a user complained to me, my first response would be to suggest superblock. It I wanted to keep keep the taint of their messages from touching my server, I could block their IP address from my network, but I don't know enough about the message delivery system to know if that would actually be effective. Perhaps if a lot of hub admins did that, it might isolate their hub to a large degree.

Such is the nature of a global, decentralized system.

Since I haven't seen anyone else mention it, I also have to say that, from a free speech perspective (and I recognize that free speech is not a globally recognized value), one of the prices of free speech is that dickheads are allowed to speak freely, too (up to the legal boundary).

❦ CRIME LICH ❦

Sat, 13 Apr 2019 16:19:03 -0700

the reports would be from users of my own hub about users from my or another hub whose content is on my hub. this would not be a report sent to offending user’s hub. upon receiving said report, i would then use the tools already available to me as hub admin to block all communications from that user on my hub alone. centralized censorship it is not, @Mike Macgirvin ⚡. i can’t and have no desire to control what goes on on others’ hubs, but on my own hub i am very proactive, even over my terrible radio connection that shits the bed when the weather turns, using the tools you’ve so graciously provided – thank you! now, to get things straight – is a hub aware of its own admins?

Mike Macgirvin

Sat, 13 Apr 2019 16:21:44 -0700

@Marshall Sutherland I think the major complaint here isn't the implementation - as we do have such mechanisms, but mostly the lack of a specific report/flag button to bring such content to the admin's attention. Your site and my site wouldn't have any need or desire to provide this, but some larger federated sites probably would. What they actually do about such a notification is undefined and is of course up to the whims of that admin.

Mike Macgirvin

Sat, 13 Apr 2019 16:33:29 -0700

centralized censorship it is not

It actually is. It's just on a much smaller scale than say for all of Facebook. But I understand where you're coming from and there's no need to argue the point. I understand what you're doing, and I have no qualms about you believing in it and I'm certain your site members appreciate it - it just isn't something I would personally do.

is a hub aware of its own admins

is_site_admin() determines if the viewer has admin rights. This is used to do things like for instance provide an admin page and is also used by the New Registrations notification for site with 'Approval only' registration. So there are existing models of how to implement such a thing.

Marshall Sutherland

Sat, 13 Apr 2019 17:03:25 -0700

As a bit of levity (which I'm sure will not be appreciated by all), I can imagine the hub admin of the hypothetical ISIS hub getting complaints about damn Westerners with their pornographic profile pictures with uncovered women's faces and their dangerous talk about blasphemous religions.

❦ CRIME LICH ❦

Sat, 13 Apr 2019 19:22:33 -0700

@Marshall Sutherland bigoted stereotyping aside, what anyone does with their hub is their business. my hub is my business.

❦ CRIME LICH ❦

Sat, 13 Apr 2019 19:24:07 -0700

@Mike Macgirvin ⚡ thank you again for the wonderful tools and guidance.