@matrix blocks were never part of the AP standard. don't advertise blocks originating from your instance and it won't be an issue

i DID NOT opt in to ANY bots sending me messages out of the blue
@matrix it doesnt "remove the purpose" if you fix your bot so that it doesnt spam remote users
@matrix i think *you* miss the entire purpose of why @nik even made the mrf for pleroma in the first place. it was to *discourage* block advertisements, not to add to the noise

@wowaname Yeah, I think I did tbh, because I thought the mrf contacted both sides

@matrix yeah no problem if you legitimately misunderstood. the mrf @nik wrote only checks blocks destined toward a "local" account: therefore it only notifies local users about other people blocking them. it was just to show mastodon users why federated blocks were a privacy risk. people like @sjw @crunklord420 and @p only adopted it because the people on their instances had fun seeing who blocked them

if people want to sign up on GLC and receive automated DMs from your instance, thats fine, but i personally dont care who blocks me and i want to avoid meta-drama as best as possible. i'm sure plenty of other people share my sentiment and treat it as spam
@matrix @crunklord420 @nik @p @sjw and if you absolutely want to advertise blocks originating from your instance (despite the spec) you can federate the Block and Undo activities, so that remote instances can handle these as they wish. overloading the DM functionality is a hack on top of a hack. people who want to be notified of blocks are already doing so by using the MRF on their servers, and people who dont can safely ignore the Block activities

@wowaname @p @crunklord420 @nik @sjw@neckbeard.xyz Tbh if you look at the source, it's a script kiddie hack.
It's supposed to chain the DMs into a thread and that work only like half the time.

@matrix @p @crunklord420 @nik @sjw bots on fedi are expected to be maintained, so if the hack breaks on you, youre still responsible for the pieces

basically ask yourself if you can justify the time cost of keeping it up

@wowaname
It didn't break. It's just poorly designed.
@p @crunklord420 @nik @sjw@neckbeard.xyz

@wowaname @crunklord420 @matrix @nik @p @sjw You *can* federate a Block but you don't need to and in fact, shouldn't.
Federating Blocks is only useful when for some reason you can't Undo the last Follow activity, which should be very rare.

Also I quite want to remove Followers-Only from Pleroma because:
- It requires that the remote servers are trustworthy (see pleroma-disrespectful, which is basically reverting one of our change; Also I think mastodon was leaking them to public in OStatus)
- Any interaction and specially replies to a followers-only post are basically leaks as they often can be seen without knowing the replied-to post
- Control of the follower collection is very loose (for regular collections it's Add/Remove IIRC, which works much better)

Only issue is that we use the Mastodon API and Gargron seems to know next to nothing about networking nor community.
@wowaname @matrix @crunklord420 @nik @sjw

> to show mastodon users why federated blocks were a privacy risk

I don't think this is a privacy issue for the person doing the blocking. What exactly is leaked? "This person doesn't wanna hear from you and does not want you to follow them any more, and the system has duly enforced this." I've never heard a real use-case. (If anyone's got one, I'm all ears, but the ones I have imagined are unconvincing.)

It *is* an issue for the person that has *been* blocked. Here is metadata about your account: two servers know it (and thus by extension, at least two admins plus anyone that gets unauthorized access to a server or a backup system that contains the DB dump), but you do not. Seriously, on what grounds should a user *not* be entitled to their own data?

> people on their instances had fun seeing who blocked them

Some people have fun with it, I personally think it's weird and passive-aggressive to not know who has blocked you, and I sure as hell don't want to waste time typing up a message at someone that won't see it and doesn't want to see it anyway. My use-case is that I don't want to waste time on that, since I'd rather talk to people that want to talk to me.

So we added it due to popular demand, but the more I think about it, the less sense it makes to hide this from the person that it targets.
@p @crunklord420 @matrix @nik @sjw
>two servers know it

gonna stop you there. anime.website is configured not to broadcast blocks to remote services. therefore only *this* server knows about blocks that i or my users instantiate. the rest of your point becomes moot afterward
@p @crunklord420 @matrix @nik @sjw
the way i understand it:
- muting is a purely aesthetic mechanism that affects only the person muting a user. it tells the instance to filter out posts from that user and deliver them to me in either a compacted view, or not at all
- blocking has the effect of forcing an unfollow and rejecting future follows from a user. blocking doesnt need to be advertised to let this happen, even though it creates visible effects (ones that dont always have to be caused by blocking). and sometimes this causes confusion because fedi is pretty broken still, sometimes follows glitch up, and we have to softblock people to fix that up. telling a user we blocked them is kinda misleading in cases like this. and in other cases where we actually want to block a user, it shouldnt really matter to the other user most of the time
@wowaname @crunklord420 @matrix @nik @sjw I have thoughts on this, but I also have breakfast. Normally I'd just reply when I got back, but given that I bitched and moaned about not getting an answer and just scrolling up resulted in an answer, I thought I should retract all the bitching I did: all the bitching about not getting this answer is retracted, with the author's apologies.
@wowaname @crunklord420 @matrix @nik @sjw

> telling a user we blocked them is kinda misleading in cases like this.

That's why people say this to each other. In implementations which have added unblock support, the immediate unblock makes it clear, too.

> it shouldnt really matter to the other user most of the time

"The other person doesn't need to see it" is a weak justification (in that it doesn't give a reason, but merely points out that the opposite behavior doesn't have a reason, which is incorrect) and doesn't really answer the question. Why should they be prevented from seeing it? It wastes time, an actual problem, and it's passive-aggressive. This is why the 'plonk' evolved: you tell people that you won't see what they have to say, and they are free to act accordingly. I have actually spent time on replies, someone says something that provokes some thoughts or they want citations for something or whatever, so I do some thinking and a little research and then reply, only to find a notification from blockbot. Thank you, blockbot, thanks Nik, now I know not to waste any more time trying to address that person.

Giving the user the option to advertise the block or not rather than making it instance-wide would be cool. (And time-limited blocking/muting, while I'm wishing for ponies.)

I don't think it should be private. It causes actual problems and for no good reason, as mutes solve the problem.
@p @crunklord420 @matrix @nik @sjw
>Giving the user the option

the problem is we have too many options on fedi already
@wowaname @crunklord420 @matrix @nik @sjw

> we have too many options on fedi already

:alexjonesshiggy2: The problem is that there exist things admins can do that most people cannot.
@wowaname @crunklord420 @matrix @nik @sjw

> gonna stop you there. anime.website is configured not to broadcast blocks to remote services.

Gonna stop there: you are the admin of your instance, and you have decided to not federate blocks. In the case that blocks federate--that is, the case under discussion, as the blockbots have zero relevance outside that context--everything I said holds, including the part where everyone fails to explain what's so private about it or why a server should store metadata about a user without alerting that user.

The case where blocks do not federate is not relevant: the blockbots do not trigger, and if the information does not leave your server (that is, it has not been given to anyone else, and has not triggered an action on any other servers) then that's just data you recorded for yourself.

And I still don't know why anyone has a problem with blocks federating.
@p @wowaname @crunklord420 @matrix @nik @sjw

this isn't a hill I want to die on, but I don't understand why people are so touchy about their public block activities showing up in a ui somewhere. your server sent the message, if you don't want people to know that information don't go shouting it at the top of your lungs maybe?
@xj9 @p @crunklord420 @matrix @nik @sjw the thread is about matrix's server shoving block notices down my throat, not about me opting in to see blocks that get sent to my server. both you and pete seem to be missing something key to the thread
@wowaname @xj9 @crunklord420 @matrix @nik @sjw It's a related topic, but I did read the thread, and the subtopic in which you tagged me was about the motivations for adding a blockbot.
@p @crunklord420 @matrix @nik @sjw @xj9 yes but my policy is to bring up issues with other admins before taking administrative actions

look where that led
@wowaname @crunklord420 @matrix @nik @sjw @xj9 I agree with this policy, although the admins of instances where I end up taking administrative action (just shitty bot floods from weedis.life and gab.com) both blocked me.

> look where that led

Not to wellactually any harder than my :facesofautism: Compulsions :autismapproved: require of me, but I think tagging more people in a paragraph with a tangent is the action that led us here.

@p
Telling someone you've blocked them invites bad feelings at best and retribution at worst.

I've previously blocked people during events they live-tweet for example. I doubt they ever knew. Purely an anti-"spam" measure.

The real problem isn't blocks propogating, it's user expectations, that *become* unreasonable during the transition to a federated model.

People used to Twitter, Facebook, et al have two general expectations when blocking:

1) The (logged in) entity cannot see my posts, or my comments on other people's posts, and vice versa
2) because of point 1, the entity cannot directly include my content in their post's metadata (it can't be a reply to my posts, or be a quote-tweet of one of my posts)

Moving to fedi, if the person being blocked is on a different server, now we have to figure out of these expectations are to apply to the remote server. If so, that server must know about it.

Literally all accusations made towards me about "hacked" Mastodon software are in regards to point 2. Mastodon doesn't really enforce that like Twitter does. If a *user* blocks me, but the instance doesn't suspend me, and my reply-to is their status url, their instance will still show my reply under their status to anonymous users and anybody who hasn't blocked me.

imo the solution is better communication with the end user, and less gaslighting. If it's decided that blocks should federate in *an attempt* to enforce the sillier expectations of a block, the user should be told "instance.tld will be notified about this block, and they may choose to alert @person." if this is undesirable, a "local block" may also be an option to add.
@wowaname @matrix @crunklord420 @nik @sjw

@r000t @wowaname @matrix @crunklord420 @nik @sjw

> Telling someone you've blocked them invites bad feelings at best and retribution at worst.

Decades of USENET and email and plonk announcements and everything was fine until people started trying to take the internet seriously. It's an automated plonk, this is useful to people that are not jerks. What you see is selection bias, anyway: you don't know who sees that they've been blocked and just moves on with their life because they don't make noise about it.

This is what people said about instance-block transparency, but it didn't turn out to be the case. Anyway, if you don't want to tell people, tell your server not to tell people. :shrug_akko:

> I've previously blocked people during events they live-tweet for example.

Why block instead of muting in this case? (Also: timed mutes and and a pony.)

(As a side note, the lower barrier to creating an account on fedi means that if someone is going to do live coverage of a scheduled, long-duration event, it might be a better idea to just make a new account and repost some of the highlights from their main account.)

> Literally all accusations made towards me about "hacked" Mastodon software are in regards to point 2.

Indeed, I'm familiar with these. Even admins don't get this a lot of the time. (I think you might remember the big "PLEROMA NAZIS ARE CONDUCING PASSIVE SURVEILLANCE" debacle when someone found out that posts federate.)

> Mastodon doesn't really enforce that like Twitter does.

Yeah, it was a mistake for Twitter to do it, starting with the dumb attempt to hide posts, and Masto's added too much Twitter-pandering fluoride to begin with.
@moonman, there was an error, some or no recipients may have received their tip.
@p @wowaname @crunklord420 @matrix @nik @sjw I think the people who think they should be able to keep it a secret who they blocked are cowardly and deserve contempt.
@leyonhjelm @p @crunklord420 @matrix @nik @sjw @wowaname
Im not sure why anyone would gaf one way or the other.
> get blocked, miss out on awesome posts! lol
@p @Secftblgirl @crunklord420 @leyonhjelm @matrix @nik @sjw @wowaname nah it makes me experience here much more pleasant. I dunno why people still defend this at all. The net effect of these block bots is more harassment (oh that’s why people defend it maybe?). All I see when they’re mentioned is some variation of “look how mad!”
@karolat @Secftblgirl @crunklord420 @leyonhjelm @matrix @nik @sjw @wowaname

> The net effect of these block bots is more harassment (oh that’s why people defend it maybe?)

I'm going to assume an implied "Present company excluded" here, since I'm only speaking for myself and for my part, I like the bot for exactly the reasons I named. If my reason were anything resembling a harassment vector, you would probably have seen that manifest by now.

> All I see when they’re mentioned is some variation of “look how mad!”

Sample bias. As covered elsewhere in the thread, you don't see the times when a reasonable-ass person sees the plonk, says "All right, guess I should get on with my life" and does that. This is my usual reaction, except when it's baffling, like Divine Asses.
@leyonhjelm @crunklord420 @matrix @nik @sjw @wowaname It's a passive-aggressive move to say "I want your instance's server to enforce this, but I don't want you to know about it." I keep hearing from people that are on instances without a blockbot, people in a thread reply to them, I say "Tell this guy I have blocked him and he can stop replying to me." Except in cases where the guy's just trying a low-level troll (and good luck trolling a former /b/tard anyway: fedi only has a couple dozen or so competent trolls among the sea of people yelling "KEEP SEETHING I OWN YOUR FREE REAL ESTATE" in reply to an Alex Jones emoji), they know not to bother. Saves me the trouble of plonking them myself.
Sign in to participate in the conversation
Game Liberty Mastodon

Mainly gaming/nerd instance for people who value free speech. Everyone is welcome.