Lemmy’s issues are pretty gigantic as well - to the point where many people outright refuse to fund its further development, thus impacting the future of the Threadiverse (one example is the inordinate amount of time spent performing moderation activities rather than actually working on adding features to the codebase).
Anyway, it’s good to have multiple alternative options - having more software implementations of the ActivityPub Protocol is unquestionably a good thing imho.
The social engineering though is NOT a good thing imho - fortunately it’s an option that can be disabled, though unfortunately it is opt-out rather than opt-in, and not transparently handled at all.
PieFed.social is outright censoring votes now, deciding what people are allowed to see or not - and this affects not only people choosing to use PieFed, but also Lemmings (& users of Mbin, Mastodon, Friendica, nodeBB, etc.) as well. The Algorithm has returned to social media, taking away your personal control and instead putting it into the hands of Big Daddy who knows best what’s good for you.
Then again, Lemmy.ml is famous for doing this consistently for literally YEARS. Every single community on there is an echo chamber where certain “undesirable” types of people are not allowed to interact. You could prove me wrong btw by going into any active community and speaking plainly in a negative manner about Russia, China, or North Korea. I’ll wait…? 🤪
Remember that at one time Lemmy devs also implemented a slur word filter, DIRECTLY into the codebase, in a HARD-CODED manner no less, and when the community cried out against that, Nutomic said in response:
If you dont like it, fork it. Stop bothering us about it, we will never fully remove the slur filter.
We hoped for better from PieFed. Instead we merely got “different”. Though the pull towards authoritarianism is hard to resist - so often it is the quickest and by far least painful path towards a desirable solution (see e.g. Anakin Skywalker in Star Wars), and I can’t fully judge anyone for considering it. Worse, it might be the height of naïveté to even so much as think that anything else could exist outside of pure theory (aka fantasy)? After all, who is offering a “better” platform - YOU? (me? anyone else stepping up to offer?)
So then what is “the solution”, in your mind? Based on your instance, I am guessing you will say Lemmy? In that case, what about PieFed.zip, running its own version of PieFed with the voting quota anti-feature stripped out? Maybe at the end of the day, the software becomes just a starting point, and it’s the instance admins that are the ones who decide which parts of the software will run or not.
(I still cannot see myself donating to the further development of Lemmy software though - if only for the reason that the enormous amount of time spent moderating the Lemmy.ml instance seems to leave little time leftover to actually work on new code changes, though ngl the genocidal attitude towards actively wishing the deaths of everyone living in a Western civilization does put me off a tad bit as well!)
I feel like we’re having two separate conversations.
What the Lemmy devs do with their instance is separate from what they do with their software, imo.
I’ve said this many times before but as I see it we have two bad choices:
Lemmy, whose devs are arguably questionable with worrysome personal politics and philosophy but who do not bake said philosophy into their software.
Or Piefes, whose devs are arguably much more acceptable and well-meaning, but who bakes their social philosophy into their software.
To me, these are both bad options, but I’ll pick the side that develops agnostically without nannying tendencies. That they chose to enact stricter political bounds within their home territory is irrelevant to me because I don’t spend time there and that feels well within their rights without encroaching on the autonomy of others, to me.
Or, in short: I don’t care even a little bit about Lemmy.ml, I only care about Lemmy.
Lemmy, whose devs are arguably questionable with worrysome personal politics and philosophy but who do not bake said philosophy into their software.
But as OpenStars pointed out, the Lemmy devs did just such a thing in a much more egregious (hard-coded) way previously. The slur filter is now optional after considerable time and outcry.
Piefed’s vote restriction is an option that admins of any instance can fiddle with, or, effectively, disable entirely. Like the slur filter currently in Lemmy, and thus less forced than the slur filter as it was initially rolled out on Lemmy. The issue, for me, is that I made a home on Piefed.social, specifically, and now it’s… not home for me.
If you don’t care about what admins do on their own instance, this shouldn’t bother you. I care - both in the abstract and insofar as it affects me, and so am… winding down my participation.
If it sounded like what I was trying to say was “PugJesus should have just picked and stuck with Lemmy!”, then I apologize, because that is enormously not what I’m trying to say.
If anything I’m saying I understand your frustration and was just lamenting a lack of a good solution, and attempting to point to the insufficiency of all available options.
I only targeted towards Piefed because of the kind of “kumbuya” idealists that claim it’s superior while being blind to its flaws.
I don’t disagree with you, your frustrations, or your decision to step away rather than re-invest somewhere with another shaky foundation
I’m just saying that there isn’t a difference here between the slur filter as is and the vote limiter as is in terms of implementing philosophy into the software. Both are options for admins in the software, not mandatory.
The Lemmy devs, however, initially attempted to make the slur filter mandatory, meaning that their attempt to implement their philosophy into the software was much more heavy-handed, and only walked-back after considerable outcry.
Basically, Rimu’s choice here is immensely shitty, but is fundamentally more a choice of Rimu as an admin than as a dev. As a dev choice, the voting limit is of questionable utility, but not forced on instances - it’s a number that admins can easily (effectively) abolish.
It’s much more a Piefed.social problem than a Piefed (all instances) problem.
I guess I’m just not enough of a “free speech absolutionist” to think that a slur filter, which admittedly I’m still not in support of as I think we’re all adults who can police themselves, is on the same level of nannying and philosophical injection as literal vote restriction.
But I see your point, thank you for clarifying and clearing up my misunderstanding
Of the two, I regard vote restriction as the more inconvenient and troublesome, certainly. But in terms of philosophy-to-code, it seems the same principle - giving admins tools in-line with the devs’ values, rather than hard-coding those values into the software.
The more core issue is that Rimu, as an admin, has used that tool in a shitty way (restricting votes specifically to cut down the ‘top voters’, rather than any of the marginal utility uses).
For instance, yes vote manipulation is arguably worse than slur filtering - both are censorship though, while then again both are changes to the contract of allowable content, rather than silent effects put upon existing submissions.
The Lemmy situation having been implemented in a HARD-CODED manner is generally considered to be overall much worse, forcing instance admins who wanted to change the situation to literally have to edit the sourcecode, whereas on PieFed all someone has to do is define an environmental variable (outside of the docker container iirc). Lemmy’s therefore was mandatory, rather than PieFed at least offering the ability to opt-out. Separately from the content of the respective issues (slurs and quotas), those process matters mean a great deal.
Lemmy also does implement its political ideology into its codebase - perhaps more in line with the features that it does NOT implement more than those that it DOES. e.g. if you recall, while Reddit lacks a modlog, it does offer a modmail, a notification bothering to tell someone that their content has been moderated, and the ability to continue conversation in a deleted post, neither of which Lemmy offers, while PieFed now allows deleted posts to remain up sometimes, and offers notifications for all moderation events. Lemmy is fairly authoritarian then, offering a great deal of power to an instance admin, a bit lesser to a moderator, and least of all to a mere user. A user has very little “rights” or “power” to affect change of their experiences on Lemmy, but PieFed is the polar opposite in that regard.
For example, PieFed offers a much more democratic style of moderation that doesn’t even need someone to decide for you what content is presented to you, by its provision of tools that allow you to decide for yourself. e.g. people bitterly complain about political topics in places like c/lemmyshitpost, which if the users were on PieFed they would not need a daddy mod, and instead could just add a keyword filter to reduce or get closer to eliminating mentions of words like “Musk” or “Trump”, etc. Notice how this is not censorship, bc it is left up to the individual recipient to decide for THEMSELVES what they want to do. Like if someone wants a break for a minute from such (or month), they can decide to do so, and then turn off the filter to resume as normal. Nobody else makes that call: just you, as the user.
This vote quota is fundamentally different though: here the votes are actually censored, not by the choice of end-user but by the instance admin. Which by itself I should add would be fine, except it was not announced almost at all, and the entire roll-out is the complete opposite of transparency. Which PieFed instances will use a vote quota & which will not? Among the former, what values are used? What if that changes later - who will notify the users that their vote contributions are no longer welcome, and on what timescale? The admins seem unprepared for all of this - I wonder if they were even told in advance that it was coming (it was discussed but were they told when a decision had been made?). I’ve seen at least two reports (possibly of the same situation) where admins were caught unawares and when they updated to v1.7 used the vote quota anti-feature without being aware of this fact.
Lemmy software is not ideal, but is manageable with effort. Now PieFed falls into that category as well, but with more features overall to start with. Use whichever you prefer - one is not more ideologically pure than the other.
I need to say though that for some people it is not as simple as using one or the other: some people actually donate to these projects, and some people have a VERY hard time donating to an instance that believes in an ideology that would see everyone in every Western civilization killed, for the crime of not being purely communist enough (although somehow Russia, China, and North Korea are “communist”, despite all the objective evidence to the contrary?). Also, the Lemmy devs have outright refused to separate donations to the further development of the Lemmy codebase vs. the operating costs of Lemmy.ml. That is not me that is tying those two concepts together: that is a firm decision by the actual developers, which they offer no negotiation on. Part of PieFed’s appeal was as an alternative to that, which is why it is so disheartening to now see that it is less “unequivocally better” and more “simply different” in terms of how it interjects its own political ideology into the codebase… although even there, as I noted above, the two are still not “equal” - i.e. Rimu did not have to expend the additional effort to make this an opt-out feature, but he did, thus limiting the amount of effort required to change the code’s behavior in this regard.
Though the ideal would have been to have expended roughly the same amount of effort and make it an opt-in rather than opt-out feature. Both Lemmy and PieFed sometimes inject their respective political ideologies into the code, but not equally so.
It’s much more a Piefed.social problem than a Piefed (all instances) problem.
I agree that in the immediate sense you leaving is related to this change having been implemented on the exact instance that you are using, but overall the outcry on this matter will be much more expansive as a result of the change to the code, rather than just the flagship instance.
That change was implemented at the software level, not as fully mandatory as anything hard-coded, but still as an opt-out rather than opt-in feature. To opt out of something you have to know that it is there, know what it does, know what the different values mean, and know how to disable it (you cannot simply “turn off” this anti-feature, though you can set the environmental variable to a value that effectively equates to that) - in short, forcing every single instance owner to expend TIME and ENERGY and the all-important & valuable ATTENTION to specifically work counter to this offering.
Maybe some instance admins pour through every line of code and do that whole procedure anyway - for all its reputed faults this seems to be what Lemmy.World does. But it’s a LOT to have to stay on top of.
So in the long run, this is going to affect the entire Threadiverse/Fediverse, well beyond a single instance, flagship or no.
For a Lemmy.world example, I was discussing medieval siege warfare (either I or someone else was quoting a primary source, I think?) when I found out that either ‘fags’ or ‘faggots’ was censored.
Had no clue before that.
Essentially, nearly every slur either has other usages (I’m reminded of overzealous word filters censoring niggardly and snigger, or ‘a chink in the armor’) or may still be relevant in the context of quotation (calling someone ‘cunty’ is being Australian, probably; calling someone ‘chink’ is being racist; knowing which 5-letter profanity starting with ‘c’ was used may be relevant in forming an opinion on someone’s behavior).
Slurs should get the ban hammer, not an autocensor; and bans should be handed out by people who can judge context.
The main problem with the slur filter is that it does neither consider context, nor considers delimiters. For example, when someone were to say the word for the purple fruit used to make wine, they’re not intending to use a word for the non-consensual sexual violation, yet the algorithm can’t see the difference when it detects the UTF-8 sequence 0x72 0x61 0x70 0x65, doing a hard-replace with the substring “removed” regardless of the context. Tom Scott once made a video about this phenomenon: it’s called String("The Sc" + "unthorpe problem").
Also, as you can see through this reply of mine, filters are pointless when there’s a plethora of ways to say the same thing without saying the actual word. Naive filters (e.g. RegExp-based match and replace) will just curb those who aren’t creative and/or knowledgeable enough, while affecting the experience for everyone who aren’t intending to do slurs, leading them to start using coded language and, thus, making it even harder to detect slurs as the slurers will eventually learn, through the non-slurers, that they can express the same thing without triggering the filter, until we get to a point in which the entire platform pivots to AI moderation, and even then there’s so many ways to express the thing without the LLM detecting, it’s called “steganography” and curbing this requires technical approaches known to be a Hard NP problem in computer science.
Lemmy’s issues are pretty gigantic as well - to the point where many people outright refuse to fund its further development, thus impacting the future of the Threadiverse (one example is the inordinate amount of time spent performing moderation activities rather than actually working on adding features to the codebase).
Anyway, it’s good to have multiple alternative options - having more software implementations of the ActivityPub Protocol is unquestionably a good thing imho.
The social engineering though is NOT a good thing imho - fortunately it’s an option that can be disabled, though unfortunately it is opt-out rather than opt-in, and not transparently handled at all.
Absolutely. My stance is not that Lemmy is not without serious flaw, just that Piefed does not seem to be the solution, to me at least.
PieFed.social is outright censoring votes now, deciding what people are allowed to see or not - and this affects not only people choosing to use PieFed, but also Lemmings (& users of Mbin, Mastodon, Friendica, nodeBB, etc.) as well. The Algorithm has returned to social media, taking away your personal control and instead putting it into the hands of Big Daddy who knows best what’s good for you.
Then again, Lemmy.ml is famous for doing this consistently for literally YEARS. Every single community on there is an echo chamber where certain “undesirable” types of people are not allowed to interact. You could prove me wrong btw by going into any active community and speaking plainly in a negative manner about Russia, China, or North Korea. I’ll wait…? 🤪
Remember that at one time Lemmy devs also implemented a slur word filter, DIRECTLY into the codebase, in a HARD-CODED manner no less, and when the community cried out against that, Nutomic said in response:
We hoped for better from PieFed. Instead we merely got “different”. Though the pull towards authoritarianism is hard to resist - so often it is the quickest and by far least painful path towards a desirable solution (see e.g. Anakin Skywalker in Star Wars), and I can’t fully judge anyone for considering it. Worse, it might be the height of naïveté to even so much as think that anything else could exist outside of pure theory (aka fantasy)? After all, who is offering a “better” platform - YOU? (me? anyone else stepping up to offer?)
So then what is “the solution”, in your mind? Based on your instance, I am guessing you will say Lemmy? In that case, what about PieFed.zip, running its own version of PieFed with the voting quota anti-feature stripped out? Maybe at the end of the day, the software becomes just a starting point, and it’s the instance admins that are the ones who decide which parts of the software will run or not.
(I still cannot see myself donating to the further development of Lemmy software though - if only for the reason that the enormous amount of time spent moderating the Lemmy.ml instance seems to leave little time leftover to actually work on new code changes, though ngl the genocidal attitude towards actively wishing the deaths of everyone living in a Western civilization does put me off a tad bit as well!)
I feel like we’re having two separate conversations.
What the Lemmy devs do with their instance is separate from what they do with their software, imo.
I’ve said this many times before but as I see it we have two bad choices:
Lemmy, whose devs are arguably questionable with worrysome personal politics and philosophy but who do not bake said philosophy into their software.
Or Piefes, whose devs are arguably much more acceptable and well-meaning, but who bakes their social philosophy into their software.
To me, these are both bad options, but I’ll pick the side that develops agnostically without nannying tendencies. That they chose to enact stricter political bounds within their home territory is irrelevant to me because I don’t spend time there and that feels well within their rights without encroaching on the autonomy of others, to me.
Or, in short: I don’t care even a little bit about Lemmy.ml, I only care about Lemmy.
But as OpenStars pointed out, the Lemmy devs did just such a thing in a much more egregious (hard-coded) way previously. The slur filter is now optional after considerable time and outcry.
Piefed’s vote restriction is an option that admins of any instance can fiddle with, or, effectively, disable entirely. Like the slur filter currently in Lemmy, and thus less forced than the slur filter as it was initially rolled out on Lemmy. The issue, for me, is that I made a home on Piefed.social, specifically, and now it’s… not home for me.
If you don’t care about what admins do on their own instance, this shouldn’t bother you. I care - both in the abstract and insofar as it affects me, and so am… winding down my participation.
If it sounded like what I was trying to say was “PugJesus should have just picked and stuck with Lemmy!”, then I apologize, because that is enormously not what I’m trying to say.
If anything I’m saying I understand your frustration and was just lamenting a lack of a good solution, and attempting to point to the insufficiency of all available options.
I only targeted towards Piefed because of the kind of “kumbuya” idealists that claim it’s superior while being blind to its flaws.
I don’t disagree with you, your frustrations, or your decision to step away rather than re-invest somewhere with another shaky foundation
I didn’t take you as saying that.
I’m just saying that there isn’t a difference here between the slur filter as is and the vote limiter as is in terms of implementing philosophy into the software. Both are options for admins in the software, not mandatory.
The Lemmy devs, however, initially attempted to make the slur filter mandatory, meaning that their attempt to implement their philosophy into the software was much more heavy-handed, and only walked-back after considerable outcry.
Basically, Rimu’s choice here is immensely shitty, but is fundamentally more a choice of Rimu as an admin than as a dev. As a dev choice, the voting limit is of questionable utility, but not forced on instances - it’s a number that admins can easily (effectively) abolish.
It’s much more a Piefed.social problem than a Piefed (all instances) problem.
I guess I’m just not enough of a “free speech absolutionist” to think that a slur filter, which admittedly I’m still not in support of as I think we’re all adults who can police themselves, is on the same level of nannying and philosophical injection as literal vote restriction.
But I see your point, thank you for clarifying and clearing up my misunderstanding
Of the two, I regard vote restriction as the more inconvenient and troublesome, certainly. But in terms of philosophy-to-code, it seems the same principle - giving admins tools in-line with the devs’ values, rather than hard-coding those values into the software.
The more core issue is that Rimu, as an admin, has used that tool in a shitty way (restricting votes specifically to cut down the ‘top voters’, rather than any of the marginal utility uses).
This is all so terribly complex.
For instance, yes vote manipulation is arguably worse than slur filtering - both are censorship though, while then again both are changes to the contract of allowable content, rather than silent effects put upon existing submissions.
The Lemmy situation having been implemented in a HARD-CODED manner is generally considered to be overall much worse, forcing instance admins who wanted to change the situation to literally have to edit the sourcecode, whereas on PieFed all someone has to do is define an environmental variable (outside of the docker container iirc). Lemmy’s therefore was mandatory, rather than PieFed at least offering the ability to opt-out. Separately from the content of the respective issues (slurs and quotas), those process matters mean a great deal.
Lemmy also does implement its political ideology into its codebase - perhaps more in line with the features that it does NOT implement more than those that it DOES. e.g. if you recall, while Reddit lacks a modlog, it does offer a modmail, a notification bothering to tell someone that their content has been moderated, and the ability to continue conversation in a deleted post, neither of which Lemmy offers, while PieFed now allows deleted posts to remain up sometimes, and offers notifications for all moderation events. Lemmy is fairly authoritarian then, offering a great deal of power to an instance admin, a bit lesser to a moderator, and least of all to a mere user. A user has very little “rights” or “power” to affect change of their experiences on Lemmy, but PieFed is the polar opposite in that regard.
For example, PieFed offers a much more democratic style of moderation that doesn’t even need someone to decide for you what content is presented to you, by its provision of tools that allow you to decide for yourself. e.g. people bitterly complain about political topics in places like c/lemmyshitpost, which if the users were on PieFed they would not need a
daddymod, and instead could just add a keyword filter to reduce or get closer to eliminating mentions of words like “Musk” or “Trump”, etc. Notice how this is not censorship, bc it is left up to the individual recipient to decide for THEMSELVES what they want to do. Like if someone wants a break for a minute from such (or month), they can decide to do so, and then turn off the filter to resume as normal. Nobody else makes that call: just you, as the user.This vote quota is fundamentally different though: here the votes are actually censored, not by the choice of end-user but by the instance admin. Which by itself I should add would be fine, except it was not announced almost at all, and the entire roll-out is the complete opposite of transparency. Which PieFed instances will use a vote quota & which will not? Among the former, what values are used? What if that changes later - who will notify the users that their vote contributions are no longer welcome, and on what timescale? The admins seem unprepared for all of this - I wonder if they were even told in advance that it was coming (it was discussed but were they told when a decision had been made?). I’ve seen at least two reports (possibly of the same situation) where admins were caught unawares and when they updated to v1.7 used the vote quota anti-feature without being aware of this fact.
Lemmy software is not ideal, but is manageable with effort. Now PieFed falls into that category as well, but with more features overall to start with. Use whichever you prefer - one is not more ideologically pure than the other.
I need to say though that for some people it is not as simple as using one or the other: some people actually donate to these projects, and some people have a VERY hard time donating to an instance that believes in an ideology that would see everyone in every Western civilization killed, for the crime of not being purely communist enough (although somehow Russia, China, and North Korea are “communist”, despite all the objective evidence to the contrary?). Also, the Lemmy devs have outright refused to separate donations to the further development of the Lemmy codebase vs. the operating costs of Lemmy.ml. That is not me that is tying those two concepts together: that is a firm decision by the actual developers, which they offer no negotiation on. Part of PieFed’s appeal was as an alternative to that, which is why it is so disheartening to now see that it is less “unequivocally better” and more “simply different” in terms of how it interjects its own political ideology into the codebase… although even there, as I noted above, the two are still not “equal” - i.e. Rimu did not have to expend the additional effort to make this an opt-out feature, but he did, thus limiting the amount of effort required to change the code’s behavior in this regard.
Though the ideal would have been to have expended roughly the same amount of effort and make it an opt-in rather than opt-out feature. Both Lemmy and PieFed sometimes inject their respective political ideologies into the code, but not equally so.
I agree that in the immediate sense you leaving is related to this change having been implemented on the exact instance that you are using, but overall the outcry on this matter will be much more expansive as a result of the change to the code, rather than just the flagship instance.
That change was implemented at the software level, not as fully mandatory as anything hard-coded, but still as an opt-out rather than opt-in feature. To opt out of something you have to know that it is there, know what it does, know what the different values mean, and know how to disable it (you cannot simply “turn off” this anti-feature, though you can set the environmental variable to a value that effectively equates to that) - in short, forcing every single instance owner to expend TIME and ENERGY and the all-important & valuable ATTENTION to specifically work counter to this offering.
Maybe some instance admins pour through every line of code and do that whole procedure anyway - for all its reputed faults this seems to be what Lemmy.World does. But it’s a LOT to have to stay on top of.
So in the long run, this is going to affect the entire Threadiverse/Fediverse, well beyond a single instance, flagship or no.
I agreed with you up until the slur filter:
Seems perfectly reasonable to me
For a Lemmy.world example, I was discussing medieval siege warfare (either I or someone else was quoting a primary source, I think?) when I found out that either ‘fags’ or ‘faggots’ was censored.
Had no clue before that.
Essentially, nearly every slur either has other usages (I’m reminded of overzealous word filters censoring niggardly and snigger, or ‘a chink in the armor’) or may still be relevant in the context of quotation (calling someone ‘cunty’ is being Australian, probably; calling someone ‘chink’ is being racist; knowing which 5-letter profanity starting with ‘c’ was used may be relevant in forming an opinion on someone’s behavior).
Slurs should get the ban hammer, not an autocensor; and bans should be handed out by people who can judge context.
Good points, and also good thing I’m not a moderator
The main problem with the slur filter is that it does neither consider context, nor considers delimiters. For example, when someone were to say the word for the purple fruit used to make wine, they’re not intending to use a word for the non-consensual sexual violation, yet the algorithm can’t see the difference when it detects the UTF-8 sequence 0x72 0x61 0x70 0x65, doing a hard-replace with the substring “removed” regardless of the context. Tom Scott once made a video about this phenomenon: it’s called
String("The Sc" + "unthorpe problem").Also, as you can see through this reply of mine, filters are pointless when there’s a plethora of ways to say the same thing without saying the actual word. Naive filters (e.g. RegExp-based match and replace) will just curb those who aren’t creative and/or knowledgeable enough, while affecting the experience for everyone who aren’t intending to do slurs, leading them to start using coded language and, thus, making it even harder to detect slurs as the slurers will eventually learn, through the non-slurers, that they can express the same thing without triggering the filter, until we get to a point in which the entire platform pivots to AI moderation, and even then there’s so many ways to express the thing without the LLM detecting, it’s called “steganography” and curbing this requires technical approaches known to be a Hard NP problem in computer science.
!fedimemes@feddit.uk