abeorch reshared this.

threads - Link to source

“With Raid Immortality Spray, bugs will no longer be granted the sweet release of death, and will instead be burdened with forever walking the Earth watching friends, family, and loved ones slowly wither and expire around them,” said brand representative Sawyer Weidman

theonion.com/new-raid-immortal…

New Raid Immortality Spray Curses Insects To Watch Everyone They Love Die

RACINE, WI—Vowing that the new product line was the ideal way to make termites and beetles suffer for all eternity, household chemical manufacturer S.C.
The Onion Staff (The Onion)

abeorch reshared this.

abeorch reshared this.

mastodon - Link to source

The UK social media ban for under-16s does nothing to sort out the harms caused by platforms.

Harmful content is a product of a business model that uses data to drive engagement and target users with ads.

Digital ID checks feed more data into this system, while leaving the engine for harms unchecked.

Read our response ⬇️

openrightsgroup.org/press-rele…

#socialmediaban #socialmedia #stopkillinginternet #ukpolitics #ukpol #onlinesafety #ageverification #StopKillingInternet #StopKillingTheInternet

Starmer’s social media ban fails to address root causes of online harms

Open Rights Group has responded to Prime Minister Keir Starmer’s announcement that the Government will ban social media, including gaming sites, for under 16s in the UK.
Open Rights Group
#socialmedia #ukpolitics #ukpol #ageverification #onlinesafety #socialmediaban #stopkillinginternet #stopkillingtheinternet
This entry was edited (Tuesday, June 16, 2026, 1:55 PM)

reshared this

in reply to Neil Brown

mastodon - Link to source

slash

@pitrh I never set out to run a mail server. I just wondered how emails were delivered when my address was nowhere in the message. Time passes (that was dial-up...) and I passed through all the spam tools, until now I just run it myself.

AliExpress gets a unique email address. So I can kill that if it becomes a problem.

Right now, my biggest offender is my own credit union (do they have those in the EU?) They appear to have sold my email address, because I *only* use it for them.

@Peter N. M. Hansteen
abeorch reshared this.

mastodon - Link to source
The media in this post is not displayed to visitors. To view it, please go to the original post.

Ya lo dejó #Tamayo bien claro...
🔍 Te roban tus datos a diario sin que lo notes. Taller práctico con herramientas libres y fáciles para frenarlo.
🛡️ Antes, Install Party para dar el salto a GNU/Linux. 🐧

📅 Viernes 26 · 17:00 + 18:30 · Las Naves, València · ¡Entrada libre!

ℹ️ gnulinuxvalencia.org/2026-06-2…

#SoftwareLibre #Privacidad #ProtecciónDeDatos #GNULinux #LasNaves #València #Fediverso #FOSS #Linux #PrivacidadDigital #DatosPersonales #SeNosHaIdoDeLasManos #SoberaníaDigital #CódigoAbierto

Taller de Herramientas para Proteger tus Datos Personales + ¡Taller de instalación GNU/Linux! (26 de junio) - Asociación GNU/Linux Valencia

Te roban los datos cada día sin que te des cuenta. 🔍 Ven el 26 de junio a Las Naves y descubre cómo plantar cara: herramientas fáciles, libres y eficaces para blindar tu privacidad frente a data brokers y grandes tecnológicas — y de paso, ¡quizás te …
GNU/Linux València (Asociación GNU/Linux Valencia)
#foss #linux #valencia #fediverso #softwarelibre #codigoabierto #gnulinux #privacidad #datospersonales #protecciondedatos #privacidaddigital #soberaniadigital #lasnaves #senoshaidodelasmanos #tamayo

abeorch reshared this.

abeorch reshared this.

lemmy - Link to source

PRISM - a self-hosted OSINT platform with a real-time dashboard
The media in this post is not displayed to visitors. To view it, please go to the original post.

I've been building PRISM - a self-hosted OSINT toolkit you run yourself instead of pasting investigation targets into someone else's web service.

Give it a domain, IP, email, phone, or username and it runs 22+ modules in parallel into one dashboard: WHOIS, DNS, crt.sh subdomains, GeoIP, threat intel (Shodan/VirusTotal/AbuseIPDB/Censys), breach data, username search across 3000+ sites (Blackbird + Maigret), dark-web mirror checks, and more. Results come with an entity graph, a GeoIP map, an OPSEC exposure score (0–100), and HTML/PDF/CSV/Markdown exports.

14 of the 22 modules work with zero API keys (missing keys degrade gracefully instead of erroring).

Stack: FastAPI + Next.js 14, runs with one docker compose up. MIT licensed.

Demo: getprism.su/
Github: github.com/NovaCode37/Prism-pl…

Built it solo - feedback welcome, especially on which modules you'd want added.

GitHub - NovaCode37/Prism-platform: All-in-one OSINT platform — 20+ modules, AI analysis, real-time dashboard, OPSEC scoring. Scan domains, IPs, emails, phones, usernames.

All-in-one OSINT platform — 20+ modules, AI analysis, real-time dashboard, OPSEC scoring. Scan domains, IPs, emails, phones, usernames. - NovaCode37/Prism-platform
GitHub
#selfhosted
This entry was edited (Sunday, June 21, 2026, 6:58 AM)

don't like this

reshared this

abeorch reshared this.

mastodon - Link to source

The Indian Restaurant is so secretive that I had to sign a legal agreement that I wouldn’t tell anyone their flatbread recipe.
It’s just a standard naan disclosure agreement.

#jokeoftheday #dadjoke #lawyerhumour

#dadjoke #JokeOftheDay #lawyerhumour

abeorch reshared this.

abeorch reshared this.

friendica (DFRN) - Link to source

Child accounts

!Friendica Developers

I am a business analyst that runs my own friendica instance focused around my friends and family..

Amoung them are some younger members who ae getting to the age where they will most likely start exploring social media.

For this reason I would really value the ability to create child accounts on friendica where their parents and/or I could retain a degree of control over what content they see.

Specifically I would like their parents as non admins to be able to:

See all the content they have access to.
Have the option of controlling which accounts they can follow and can follow them
Can limit their ability to post publically.
Limit their ability to create groups, pages etc.
Have notifications set for any follow requests they get
Control whether they can login or not (temporarily)
Have the ability to turn these controls on/off as they get older so they can retain the account as an adult.

I understand that these are a significant set of features.

I am wondering whether others have an interest in this kind of functionality amd would be interested in talkking more about it.

What would be the best way of supporting those with friendica coding skills who might like to implenent these kind of feature? How best could I contribute?

!Friendica Developers

reshared this

in reply to abeorch

friendica - Link to source

OldKid ⁂fedibook.de⁂

@abeorch @Michael 🇺🇦

I personally like the idea as such — and as the admin of a Friendica instance, too. But as Michael already hinted, there are a number of legal pitfalls involved. First off: the assessment below is based on the legal framework that applies in Germany and the EU (in particular the GDPR, the German Basic Law (Grundgesetz), and the German Civil Code (BGB)) — it may look different in other countries, and it isn't binding legal advice in any case.

How big those pitfalls are depends crucially on whether the instance is public or a purely family instance — that makes a huge legal difference, so I'll go through both points for each case.

The fundamental difference up front

On a public instance, I process the personal data of other people and am therefore fully a "controller" within the meaning of the GDPR. On a purely family instance, by contrast, the household exemption applies (GDPR Art. 2(2)(c) — "purely personal or household activities"), and a large part of the obligations simply falls away. However — and this is the most common misconception — that exemption only holds as long as the traffic actually stays within the closed circle. Friendica federates outward via ActivityPub by default; as soon as the child's content reaches other instances, it becomes accessible "to an indefinite number of people," and under the case law of the European Court of Justice (Lindqvist, Ryneš) the household exemption tips over at exactly this point. So "private," in legal terms, means: a closed circle and federation restricted accordingly.

1.) Access for minors (GDPR Art. 8)

Public instance: I may not simply grant a minor under 16 access without further ado — what's required is the verifiable consent of the parents (GDPR Art. 8 is fairly clear on this). As the admin, I have to be able to demonstrate


  • a) how old the child / adolescent is, and
  • b) that the person giving the consent is in fact the holder of parental responsibility.


The consequence is that I have to retain these proofs — i.e. process additional and in part highly personal data — and for that I again need a legal basis, data minimization, and documentation.

The verification of (b) in particular is hard to accomplish in any serious way on a public instance, and it is at the same time a child-protection problem: if any arbitrary person can register as the "parent" of a child account, then I am handing out full visibility into and control over the child — reading their messages, controlling whom they may follow and who may write to them, when they may log in — possibly to a stranger rather than to the genuine legal guardian. The very functions that are intended as protection for the child become a tool in the wrong hands: a person with bad intentions would thereby gain seamless surveillance, isolation from trusted contacts, and access to a child's private communication — that is, precisely the means by which grooming and abuse operate. So without reliable verification of the parent-child relationship, I'm not just building a protective function but potentially an instrument of abuse. That's why verification here is not a formality but the very heart of the matter.

Family instance: Here this critical point eases almost entirely. I know the families, the parent-child relationship is verified offline — a stranger registering as a "parent" is simply not possible within the closed circle, and the core child-protection problem falls away. The formal retention and documentation burden for the highly personal data also largely disappears, since there is no GDPR controllership.

2.) "See all the content they have access to."

Insofar as this refers to the child's / adolescent's public posts, it is unproblematic in legal terms in both cases. With private messages, however, the paths diverge again:

Public instance: I may not make the private messages accessible to the parents without first informing the affected third party who is corresponding with the child — and without having a legal basis for it. After all, the third party has not consented, and parental responsibility covers the child, not their communication partners. The viable legal basis for this is the third party's general right of personality (Art. 2(1) in conjunction with Art. 1(1) of the Basic Law — informational self-determination and the confidentiality of communication) together with the GDPR (I am the controller and need a legal basis + transparency under Art. 5, 6, 13/14). Thematically related, but in the individual case rather doubtful or not applicable, are the secrecy of telecommunications (Art. 10 of the Basic Law, Sec. 3 TDDDG) as well as Secs. 206, 201, 202a of the Criminal Code (StGB) — these mostly don't apply directly to a private, non-commercial instance.

In practical terms this means: either private messages are excluded from the parental view, or it would have to be marked — ideally visibly across the entire Fediverse — that this is a restricted account whose messages can be read along.

Family instance: A parent privately reading their own child's messages is acting within the scope of parental responsibility, not as a data-protection controller — so that is not a problem. With one caveat: as soon as the child is messaging someone on a foreign instance, that external third party still has their personality rights and knows nothing about the reading-along. So this part remains in play as soon as the communication leaves the family circle — one more reason to restrict federation for the child accounts.

What stays the same in both cases

Independently of the GDPR and the type of instance, the privacy of the growing child remains an issue. Sec. 1626(2) BGB obliges parents to take into account the child's growing need to act independently and on their own responsibility, and the child is themselves a holder of the right of personality. Seamless reading-along is care in the case of an 8-year-old, but a disproportionate intrusion in the case of a 16-year-old — and that applies in the living room just as it does on the family instance. This is precisely why the age-graduated, "growing-with-them" idea (winding the controls back as the child gets older) is spot-on from the very start.

Conclusion

Technically, the feature would be great. For a public instance, the parent verification is the critical sticking point above all — and that is exactly why, when it comes to minors, the gated approach (manually approved accounts with an offline-verified parent-child relationship) is not just cleaner but the only responsible option. For a purely closed family instance without open federation, by contrast, the legal pitfalls shrink down to a matter of family-law good judgment: staying proportionate and winding the controls back with age. But as soon as a family instance still federates publicly, you're partly back in the GDPR and affected-third-party territory when it comes to contacts outside the family. All of this refers to the legal situation in Germany / the EU — for a solid assessment of the specific individual case, professional legal advice would be advisable.

I wrote the text in German and had it translated using Deepl.

@abeorch @Michael 🇺🇦
in reply to OldKid ⁂fedibook.de⁂

friendica (DFRN) - Link to source

abeorch

Sorry for the slow reply. There is a bit to break down there. My response is somewhat unstructured. I apologise - might need to sort out my thinking a but more.

My thinking is primarily within what I think you would say were Private instances

1. Public vs. Private instances - One option would to be to restrict federation for child accounts - this would avoid many of these challenges as you say as it would effectively turn the account into something that is controlled by the 'parent' account

2. Supervised / supervisor account compared to Parent and Child Accounts - The functional capabilities are actually independent of the idea of whether the people concerned are younger or older than 18 - i.e the functionality could be implemented with no reference to age.

3. Current processes for Public instances - Are all public instances currently collect identity documents to confirm their age of users as above 18 or is it based on attestation - i.e. people signing up just say they are above the age of 18? - If this is the case could attestation that the person creating the 'child' account is the parent/guardian of the 'child' etc. t

4. Third party communication - My understanding is that following has / can have an approval acceptance process (follow requests) - I can't see why this could not resolve the issue of advising third parties that their content might be ready by the 'parent'

This actually promoted me to think - What are standard terms of service for Friendica instances? - My instance is private and well I've never really worried about that.

It still seems to me these challenges are not insurmountable.

Friendica Developers reshared this.

in reply to abeorch

friendica (DFRN) - Link to source

abeorch

I looked at ico.org.uk/for-organisations/u…

Quoting

It’s usually appropriate to let a person exercise a child’s rights on their behalf if:

you’re satisfied that the child is not competent; and
the person who has approached you holds parental responsibility for the child.

The exception is if you have evidence to suggest that this isn’t in the child’s best interests.

If you’re confident that the child can understand their rights, you should respond directly to the child. However, you could allow the person with parental responsibility to exercise the child’s rights on their behalf if the child authorises this, or if it’s clearly in the child’s best interests.

The way i read this you have a "best interests of the child" override but I am interested in other jusidictions - regardless the functionality can be used or not used depending on an individual sever admin's view.

Perhaps if you are in Germany and concerned it might breach their rights you could turn off the functionality.

Maybe I should add two user stories.

As a instance admin i can turn on , off or.require my approval for each child account creation

As an admin I can release a child account from any of the restrictions imposed by a parent

Also thinking about it also need a process for admins ans parents to add and remove other parents

Friendica Developers reshared this.

in reply to abeorch

friendica - Link to source

OldKid ⁂fedibook.de⁂

@abeorch @Michael 🇺🇦 There isn’t a single simple definition… in this case, it involves several laws, ranging from the GDPR to the Telemedia Act, the Constitution and the Criminal Code. And those are just the laws that spring to mind right now, without me being a lawyer.

Your suggestion to flag children’s accounts for other users would be a solution, but it must be ensured that all other users in the Fediverse can see and understand that if they message such an account, their data may be viewed by third parties – and that they must be aware of this before messaging such an account. In my opinion, this would mean that all apps in the Fediverse would have to initiate an extra query for this purpose and display a warning.

@abeorch @Michael 🇺🇦

friendica (DFRN) - Link to source

Self hosting Dotmakeup

!Selfhosted @Vincent Cloutier - I'm fairly new to running services on my vms but does anyone have any experience running DotMakeup ? I'd like to host an instance to share some Twitter and instagram accounts in a controlled way to my family on my #friendica instance.

I'm kinda of at that point where I can sort of use #DockerCompose but umm.. honestly not really sure where I would start from git.sr.ht/~cloutier/dotmakeup

~cloutier/dotmakeup - Twitter bridge to the fediverse - sourcehut git

git.sr.ht
#friendica #DockerCompose @Vincent Cloutier !Selfhosted

don't like this

Selfhosted reshared this.

in reply to abeorch

friendica (DFRN) - Link to source

abeorch

Soo.. I'm still trying to get #dotmakeup to work. I've got the docker container running with a few changes in the compost file to name my instance (I'll put twitter/insta keys in soon) but running it leaves me wondering.

Docker PS says :
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
2d5ad2cef33d cloutier/bird.makeup:latest "dotnet /app/Birdsit…" 23 minutes ago Up 23 minutes 8080/tcp, 0.0.0.0:5000->80/tcp, [::]:5000->80/tcp birdmakeup
6990de60413e postgres:15 "docker-entrypoint.s…" 23 minutes ago Up 23 minutes 0.0.0.0:5432->5432/tcp, [::]:5432->5432/tcp dotmakeup-d

So I think I should be looking at port 80 right? But just gets a connection failure..

#dotmakeup
in reply to abeorch

friendica (DFRN) - Link to source

abeorch

I think this is saying that its listening on port 5000 - but I'm unable to connect

root@docker:/home/abe/dotmakeup# sudo lsof -i -P -n | grep LISTEN
sshd 622 root 6u IPv4 5916 0t0 TCP *:22 (LISTEN)
sshd 622 root 7u IPv6 5926 0t0 TCP *:22 (LISTEN)
docker-pr 1083 root 8u IPv4 7788 0t0 TCP *:5432 (LISTEN)
docker-pr 1087 root 8u IPv6 7789 0t0 TCP *:5432 (LISTEN)
docker-pr 1200 root 8u IPv4 8118 0t0 TCP *:5000 (LISTEN)
docker-pr 1205 root 8u IPv6 8119 0t0 TCP *:5000 (LISTEN)

friendica (DFRN) - Link to source

New to Authentik - Challenges with getting my first application working

!Selfhosted - I'm new to #Authentik - I've just spun up a test instance and tried to connect my first application (#Postiz) but I'm kind of stuck.

I've installed both Postiz and Authentik using Docker Compose - as provided by the relevant apps.

As far as #Authentik goes it looks like its approving authentication requests (its showing successes in the GUI) but the application just returns to the login screen without progressing.

One thing that I found was that within the configuration of the Application and Provider one URL provided by Authentik is the /application/o/postiz/.well-known/openid-configuration which lists several URLs used by the application - one is /application/o/authorize/ but unlike all the others this returns a 404 error when I try to load it via the browser.

I am struggling to work out if - this URL should 404 should occur and also how to diagnose what the problem is.

I've popped messages in their relevant Discord forums but any ideas or input would be greatly appreciated - I'm figuring getting Authentik is going to be key to getting other applications going.

#authentik !Selfhosted

Selfhosted reshared this.

in reply to abeorch

lemmy - Link to source

brewery

I have self hosted for years using Authentik and have 50 + services. Postiz has been one of the most challenging to spin up and maintain. The documentation is not up to date, the software is constantly changing and I had to tweak lots to get it to work. Even now, it seems to crash after a few days and haven't had the chance to investigate. All these posts I set up didn't go through and now if I turn it back on, it'll release them all in one go before I can get in to stop it, or have to revert to a fresh install...

The vast majority of services and not like this so don't get disheartened! Try some other ones first them come back to this when more experienced. I recommend Bento PDF, ConvertX, outline wiki, vikunja and immich (if you have the storage space).

At first, try to stick to software on Authentik's guides (integrations.goauthentik.io/). Once you do a few OIDC and forward proxy services you'll get the hang of it.

When I get a chance I'll share my docker compose SSO settings and authentik redirect uri that worked for me with the caveat that the software crashes for some reason.

Integrations overview | authentik

What is an integration?
integrations.goauthentik.io
in reply to brewery

friendica (DFRN) - Link to source

abeorch

Oh cheers. Yeah I think I need to test setting up another app to confirm whether its Authentik or Postiz. The other app I am looking at is an open source core banking system so I think I'll leave that one for a while and as you suggest try something more straight forward.

I'm keen to get Postiz working because I belong to a sailing club and its a time vacuum posting updates all over the place.

I'm interested do you you anything for social listening - I'd like to pull together a few social feeds from Twitter, Instal, Facebook and provide them to the few of us that are working on keeping the club's profile up.

Selfhosted reshared this.

in reply to abeorch

friendica (DFRN) - Link to source

abeorch

So an update. It turns out that all the issues were actually to do with Postiz rather than Authentic. - I tried installing a few other apps and linking them to Authentik but I ran into a few issues. Then I took Postiz and tried turning off the Atuthentik authentication and then just getting it to work with native user accounts. This initially didn't work - it turned out that there were a few issues.

1. The startup procedure for Poztiz actually left the backend API service not running properly. I had to run the docker compose, get all the other services running and then stop the postiz container specifically wait a bit more and then restart just that container. That got the API service running.

2. Because I was just running it internally I didn't have SSL certificates configured and was just using http - consequently I had to introduce NOT_SECURED: "true" into the environment section of the Postiz service in the compose file and then repeat the process above. This resolved a silent error in my browser where cookies used to manage the authenticated browser session were discarded because I wasn't using SSL.

So success at least getting Postiz running. I am now going to step back, reconfigure it for Authentic retry.

I am going to guess that these issues might be relevant if you are running Postiz behind a reverse proxy (I'm not sure if this is being done with the docker compose set-up). I possibly also need to look at how much memory and CPU I have on my test VM.

I've posted similar details in Postiz Discord. If anyone else is using #Postiz and would like to exchange notes - please do drop me a mention. Always good to have other people to talk to about these things and share notes.

#Postiz

Selfhosted reshared this.

friendica (DFRN) - Link to source

TOCs - Must haves

!Friendica Admins - I have realised that when I set up my private server I never really considered terms of service text -

What have you included in your Terms of Service page and why? Are you based in the EU and have you included specific Privacy statements and do you have references to GDPR? - I would really like to work on appropriate text to include for both public and private (closed registration for groups or friends & family ) servers.

While the Friendia.ca text provides a starting point. I'm not sure that I should be blindly copying and amending that text.

Friendica Social Network | Tos

friendica.ca
!Friendica Admins

reshared this

in reply to abeorch

friendica - Link to source

Andy HΞ3

Yes, there is a minimum age requirement, but it's not verified. The terms make clear that this server is not intended for younger users that may need closer supervision; our moderators are not able to constantly monitor all public posts.

The server is situated in Asia-Pacific, where most of our users are based. The choice of Irish law is for pragmatic reasons, as some of the operators have connections to that country. The terms aim to set out a framework we intend to follow as a minimum standard. This does of course offer limited protection to the service provider if their assets and interests are elsewhere. It's still probably better than opting for a jurisdiction with draconian laws. Indeed, there are places in the world where people have served prison sentences merely for clicking "Like" on Facebook posts that were deemed illegal in that place.

On a more cheerful note, the terms are written to be accessible, intended to set clear expectations, so everyone can get the most out of and enjoy the service.

in reply to Andy HΞ3

friendica (DFRN) - Link to source

abeorch

Yeah I mean its a helpful to have something. I wasn't sure about whether your choice of jurisdiction is something that you can elect but I can understand the thinking. You might want to check how enforcible that nomination is.

I can see that not too far from now I might be running into issues with New Zealand and Australian social media laws and it sounds like the UK is going that way as well.

Friendica Admins reshared this.

abeorch reshared this.

lemmy - Link to source

TIL “The Front Fell Off” is from Australian comedy news duo Clarke and Dawe
And, they have years of other comedy news skits, including Defence where they discuss aircraftless aircraft carriers. Hoo-rah.

- YouTube

Disfruta de los vídeos y la música que te gustan, sube material original y comparte el contenido con tus amigos, tu familia y el resto del mundo en YouTube.
ABC News (Australia) (YouTube)
#til

abeorch reshared this.

in reply to wuffah

lemmy - Link to source

ᛒᛚᚢᛖᛇᚦᛖᚱ (BlueÆther)

and I'd just like to say that Clarke is a New Zealander that had to move across the ditch because NZ is just too small to contain such talent

Look up Fred Dagg:
youtube.com/watch?v=AYvMeT2GC1…
youtube.com/watch?v=AeUzrsjwF4…

fred dag 01 Country Calendar 1974.avi

Disfruta de los vídeos y la música que te gustan, sube material original y comparte el contenido con tus amigos, tu familia y el resto del mundo en YouTube.
Gavin Storey (YouTube)
abeorch reshared this.

mastodon - Link to source

Casting a large net here but does anywhere in my circles do #accessibility consulting for the web?

Edit: thank you everyone for helping me out! I have some solid leads I’ll track down on Monday.

#accessibility
This entry was edited (Saturday, June 13, 2026, 7:25 PM)

abeorch reshared this.

⇧