> We want to make sure we trully understand what you're struggling with.
The post literally starts with a list of grievances. Maybe ask the AI for an executive summary and the key points.
Mozilla's response should not be limited to clarifying these grievances. But it could have been all the staff member who responded could do.
Looking through that wiki there seems to be a lot of things that ML would get wrong.
It sounds like Mozilla just turned on the machine without consulting the human translators to see if the machine actually worked in a useful manner.
Yes. And someone should make a real apology. But learning what the machine did wrong is part of fixing a machine.
Saving zero dollars and making the product worse is not important, only that there doesn’t seem to be a browser monopoly is.
The person replying is probably not an expert in this. But they want to get more details so they can figure out how to get it to the right people with more information.
This is how it's supposed to work.
Some people just like drama.
Especially when AI is involved, the anti-AI team feels like they need to step up to the plate.
Or maybe an offer to set up a call and talk about the problem and possible solutions in person is not such a bad move after all? Seriously, I don't see how you can be mad at the fact that a representative of an organization wants to discuss the actual problem with an actual member of the community for a change, instead of just writing the usual "sorry but not sorry" corporate bullshit message and call it a day. Maybe it won't solve anything and they won't find a common ground anyway, but still, I cannot imagine a more honest attempt at trying.
Much less crashing in with it in the form of a “SumoBot,” as Mozilla seems to have done to its non-English communities… (with the disclaimer that I have zero insight into Mozilla’s process here outside of this writer’s account).
It puts a name to a considerate consensus-based way to approach change, that seems humane (and effective) in any culture—leave it to the Japanese to have a specific term for it…
I have to say it feels like a really familiar, NGO-flavored disrespect, though: “we’re doing this favor for underrepresented language communities,” regardless of whether they want/need it or not.
“There’s only X number of you having to shoulder the load in XX sub-community, don’t you want us to impose a bunch of ‘help’?”
Well, no, if the choice is between a formidable volume of slop and a smaller but well-executed volume of volunteer labor-of-love…
(…I say as a person very much without all sides of the story, and shooting from the hip a bit. I don’t mean to impugn anybody’s intentions, and I imagine at the end of the day we’re all on the same side here.)
For any RFC, there will be a "comment" after publication from someone who did not take earlier comments seriously enough to read them.
Long time in Japan too, I would not consider newamashi as being Japan's strengths.
The sort of consensus building ultimately involves having to do stuff to make people's opinions feel taken care of, even if their concerns are outright wrong. And you end up having to make some awkward deals.
Like with all this "Japanese business culture" stuff though, I feel like it's pretty universal in some degrees or another everywhere. Who's out there just doing things without getting _any_ form of backchannel checking first? Who wants to be surprised at random announcements from people you're working with? Apart from Musk types.
But of course some people are very comfortable just ripping the band aid off and putting people in awkward spots, because "of course" they have the right opinion and plan already.
Why context matters in judging whether some practice is good or not.
Just that lacking context one really can't make that many blanket statements.
Also, his demanding of not using his work for AI training is nonsense. Because entire articles, this one included is published under a Creative Commons license.
Didn't he agree on that?
Mozilla must reject his further contribution because he stated he don't understand the term of Creative Commons license. His wish granted I guess.
> I request to remove all my translation from learned data of SUMO AIs.
It's Mozilla's data...
> explicit violation to the Mozilla mission
I'm not sure what this is referring to. I don't see any explicit violation of Mozilla.org's mission. If anything it seems consistent with that mission to provide universal translation with quick turnaround.
Really though, all I needed to see was the phrase "jump on a quick call" to form an irrationally strong opinion. That phrase instantly warms my entire body with rage.
The fact that the preceding apology was absolutely awful does not help. "I'm sorry for how you feel" is wrong, since nobody asked them to react to "feelings" but the clearly delineated problems with the automation that Mozilla rolled out.
Asking to discuss something like this over synchronous voice comms is basically asking to go off the record and handle things privately. Sometimes that's appropriate, but if that's what the correspondant wanted they would have asked for it.
These three things combine to tell anyone who is paying attention that this is damage control, not meaningful engagement, and it's offensive to act this way toward someone who has put this much time into your project.
Instead it is usually a PR tactic. The goal of the call requester is to get your acquiescence. Most people are less likely to be confrontational and stand up for themselves when presented with a human - voice, video, or in person. So, the context of a call makes it much more likely for marsf to backpedal from their strongly presented opinion without gaining anything.
This is a common sleazy sales tactic. The stereotypical overly aggressive car salesman would much rather speak to you in person than via email even though the same information can be conveyed. It is also used in PR and HR situations to grind out dissenters, so it comes off in this context as corporate and impersonal.
It was this exact part of the conversation that touched me negatively too. marsf expresses some very valid criticism that, instead of being publicly addressed, is being handled by "let's discuss it privately". This always means that they don't want to discuss, they just want to shut you down.
If they truly think they're in the right, they can discuss it in public, like the poster already did.
'I'm sorry that our actions caused such distress' come a bit closer to being a true apology.
Importantly, 'if' was changed to 'that'.
It seems like someone who has no awareness of the problem, who wants to learn more about the problem, and the fastest way for both parties is over the phone ASAP rather than through a bunch of emails.
When software goes wrong, you need as much information as possible to figure it how to fix it.
The lead realized that Mozilla doesn't care about their opinion (they did this without discussing with them) nor do they care about the work they were doing (by replacing their work with machine translations). A "quick call" doesn't solve this.
Generally speaking, orgs aren't trying to replace high-quality human translations with lower-quality machine translations. They are often trying to put machine translations in where there are no translations, though. Getting the balance right requires fine-tuning. And fine-tuning requires a quick call to start to better understand the issues in more detail.
that's the problem. stop thinking about the org and think about the person. these are volunteers who feel taken advantage of, being met with corporate jargon
fly out and take him to dinner if you actually give a shit. or write a check. a "quick call" is so insulting
A quick call is a courteous first step. The other person might not have time for a long call, so you want to show you're respecting their time. Then you follow it up with a longer meeting with the relevant engineer and manager, etc. "Taking someone to dinner" is not the first step here. The way to show you care is by trying to understand the situation before anything else.
There is no world in which this is insulting.
> They are often trying to put machine translations in where there are no translations, though.
And at what point are all of the translations done by machines and the work the community is doing no longer needed? At the very least, the nature of their work will change and I think they're not interested in participating anymore.
Let's just assume it is how you say it is. (The only assumption I am not willing to make is that people at Mozilla are already convinced it was a bad idea after all.) What in your opinion would be the right move now, after they rolled this bullshit auto-translator out and pissed off a lot of people in the community, including a major contributor for the last 20 years? Surely they could just ignore him and go on with this auto-translation initiative (BTW, thay don't even have to worry about whatever he wants to "prohibit" to do with his translations, because he waived off his rights by posting them). Would it be better than trying to set up a call and discuss things, try to find some compromise, gather a number of recommendations she may then pass onto people working on the auto-translator initiative (because surely this Kiki person, whoever she is, is not the sole person responsible for this and cannot magically just fix the situation)?
I think it's clear that Mozilla wants machine translation to take a bigger role in producing localized content, and this new process will be a large shift in the way things have been done. I think it's fair for Mozilla to do this, but I also think it's fair for the maintainer to be upset with this decision and no longer want to volunteer his time to clean up slop.
The initial response feels premature and tone deaf which is why people are irked by it.
Given that Mozilla "shot first" so to speak, the onus is on them to take action first e.g. disable the bot, revert changes to articles, etc. Only after doing this can discussion on a path forward happen.
Because a lot of people in this thread are whin… ahem, expressing their discontent with Mozilla, as we all usually do, but I've yet to see anybody to propose anything realistic at all, let alone better than ask an offended community member for a call and at least to try to talk it through and establish what could be some actionable steps to remedy the situation.
In Japan? Sincere appology followed by resignation.
No, the Japanese absolutely do not set up a call to discuss things after you've scerwed and disrespected them. They respectfully give you the cold shoulder.
Mozilla should not be surprised if their market share dwindles in Japan after this.
Seems that this is exactly what Mozilla did? And Microsoft, and Reddit, etc.
Companies are absolutely falling over themselves to replace high quality human translations with lower quality machine translation. I’m not sure how a hacker news poster could miss this trend.
I don't disagree with your statement, but I read the sentence: "Would you be interested to hop on a call with us to talk about this further?" with a similar gross reaction as the OP comment did.
Reading that in response to Marsf's original message of airing grievances and feelings of disrespect towards his work felt entirely tone-deaf and corporate in nature. Especially in context of this being in response to the Japanese team, where Japanese business communication norms are often at odds with the American standard.
You might think that this method of communication is inefficient, but the heart of the matter seems that the Japanese team finds the very emphasis on efficiency as disrespectful when it comes at the cost of the human element of respect.
The person is a "Support Community Manager" in Indonesia if you click on their link. They're not the CEO of Mozilla who is supposed to be an expert in intercultural communication. I think you're being kind of harsh on someone who is presumably not high-level and just trying to do their job and get more information to be helpful.
This is completely backwards. The CEO is not expected to manage intercultural communication. You know whose job that is? The community manager.
The community manager for Indonesia wouldn't be expected to manage communication with Japan, but managing local contributors is absolutely a job for the community manager and not the CEO.
Very bill lumbergh energy.
No, he didn't. I'll repeat a comment I made elsewhere:
The problems are nowhere near actionable. A lot more information is needed. E.g. literally the first bullet: "It doesn't follow our translation guidelines". OK -- where are those guidelines? Is there a way to get it to follow them, like another commenter says works? Does the person need help following the process for that? Or is there a bug? Etc.
These are the things a call can clarify. It's the necessary first step, so why are people complaining?
> Calls require focused attention which if you are used to multi-tasking are a huge drain.
Solving important problems requires focused attention. Which is why you get on calls to solve them when they're urgent and important, and not something that can be multitasked.
- No apology
- No "we stopped the bot for now"
"We're sorry for how you feel" is enterprise for "we think you're whining". Maybe not what the person meant but how anyone is going to read it.
The original sin here is Mozilla just enabling this without any input from the active translation community.
This isn't a new problem, loads of Japanese translations from tech companies have been garbage for a while. People sticking things into machine translation, translators missing context so having absolutely nothing to go on. Circle CI, when they announced their Japan office, put out a statement that was _clearly_ written in English first, then translated without any effort of localization. Plenty of UIs just have "wrong text" in actions. etc etc.
Anyways the point is just that one side of this relationship here clearly cares about the problem way less, and _even when presented with that fact_, does not even pretend to be actually sorry for the damage they are causing.
- No "we stopped the bot for now"
"We're sorry for how you feel" is enterprise for "we think you're whining".
Anyways the point is just that one side of this relationship here clearly cares about the problem way less, and _even when presented with that fact_, does not even pretend to be actually sorry for the damage they are causing.
This is just a single initial reply from a "community support manager" in Indonesia. It's not from the Mozilla CEO or the leader of the project. They surely don't have the power to stop the bot. But what they can do is find it more over a call, and then who to escalate it to. Then maybe it does get turned off before it's fixed or changed.
You seem to be confusing someone in customer support with someone who holds power over entire projects. I don't understand how you think a customer support person should be able to just turn off software across the globe in response to a single short message on a forum with few details.
I do get what you're saying, and it's not like I think the CSM should be fired for the message. I just think it's bad comms.
Here are some alternative choices:
- post nothing, figure out more internally (community support is also about vouching for people!)
- post something more personal like "Thank you for posting this. I'm looking into who is working on this bot to get this information in front of them". Perhaps not allowed by Mozilla's policies
- Do some DMing (again, more personal, allowing for something direct)
But to your point... it's one person's message, and on both sides these are likely people where English isn't their native language. I'm assuming that community support managers are paid roles at mozilla, but maybe not.
And like... yeah, at one point you go into whatever company chat and you start barking up the chain. That's the work
Regardless they are representing the company. If they aren't the right person to respond - they should not have responded and kicked it up the chain/over the fence to the right person - instead of responding by offering to waste the complainants time on a call with someone you are asserting is not the right person to be handling this. Supposing you are correct about their position, it makes their response far worse, not better.
It is true both they and the person they are responding to are marked as "SUMO Locale Leaders"... but it seems rather clear from the context that is not the role they are inhabiting in their (non) apology and request for a "quick call" with the complainant.
The language they use is certainly not the language a peer would be expected to use either.
In a company you should never ever "quick call" someone (especially on a group forum) who has presented a genuine list of grievances against whatever you're doing, unless you're subtly trying to pull rank to override those grievances.
It suggests a decision can be reversed with a quick call, which questions one’s choices or conviction. As if to suggest the choice was made without considerable thought and care. It’s such an unserious tone to a moment that’s very serious to the other.
I guess it acts as a mirror of sorts though, because that's precisely how this decision appears to have been made in the first place. But it's clear that whoever represents Mozilla there is already assuming the fault lies with the person that just got kicked.
> I'm sorry for how you and the Japanese community feel ...
That may seem like an apology, but it's more a dismissing their issue as "that's a you problem".
It would be such irony if they asked GPT to reword it to a more polite tone though...
It's almost always (like this time) "I'm sorry you feel that way, please spend more of your free time<EOF>", and sometimes (like this time) "[we're doing it anyway but maybe we'll make some changes]".
It feels insulting because it is insulting. The decision has been made, they just want to not feel bad about you being insulted.
So, even in the best "sorry we screwed up" scenario, the quick call covers their butt and let them leeway to backtrack as needed. That's also part of why we viscerally react to opaque meetings IMHO.
“Sorry for your feelings” comes off as dismissive and avoiding taking ownership for the lost work and years of volunteer contributions.
In the end it may boil down to some strong hatred for AI, this seems to be very common recently and "I prohibit to use all my translation as learning data for SUMO bot and AIs" certainly points that way. If that is the root cause then it may be impossible to resolve to the satisfaction of both sides.
Mozilla is painted bad here, but who knows if the automated translations do not help more people than it hurts the translators.
What if the reduced financial pressure allows Mozilla to focus more on privacy and less on ads.
Unfortunately these things are really gray, but you really can’t expect a company to keep you paying in good will.
Mozilla should have discussed this with the translators in advance at least.
> What if the reduced financial pressure allows Mozilla to focus more on privacy and less on ads.
My impression was marsf was a volunteer.
- No prior communications.
- No discussion about what uses the contributed information was being put to.
- No discussion about the release and the parameters around the operation of the bot.
- No discussion about whether or not this was a desirable in the first place (with the community, not just internally).
- Flippant tone to someone who is clearly severely insulted.
If it was a paid job and you treated the person who did it like this it would already be beyond rude, if it is a volunteer group then it is more than enough to throw in the towel. This isn't gray.
There was a period where every company was trying to "crowd source" free labor. It died off because people didn't like working for corporations for free.
I can see why they have it under Mozilla.org. And lots of companies have community support.
But I do think we should ask ourselves whether companies have some sort of moral obligation to continue relying on unpaid labor because it might make the unpaid laborers feel a sense of meaning. I'm very sympathetic to the need to have a sense of meaning. But I'm less sympathetic to for-profit companies relying on unpaid labor and especially to the idea that we should encourage more of it.
I too would be upset if an organization threw out a decade of translation work without any warning or discussion, in favor of a robot pretending to understand my language and failing.
Sure we had machine translation before, but it was still a little off. Now the latest language models get us 99.9% there, so they are judged good enough to deploy at scale. What results is a weird twilight zone where everything is in your language, except it feels kind of wrong and doesn’t really communicate in ways specific to the culture from which the language is.
You’re in France, you search for something, a thread pops up with everyone interacting in French - seems reasonable enough, but it just reads kind of weird? Then a message is entirely out of place, and you realize that you’re reading an English language thread translated to French.
Or your mom sends you a screenshot of a Facebook thread in her native language that has her worked up - and reading it, you realize it’s an LLM translation of something that should have no bearing on her.
Same with various support pages on websites - it all reads mostly fine until you hit a weird sentence where the LLM messed up and then you’re transported back to the reality that what you’re reading was not authored by anyone who can actually operate in that language/culture.
There’s a lot of nuance in language beyond the words - how you express disagreement in English is not how you express disagreement in Japanese, how you address the reader in French is not the same as in Korean, etc. Machine translation flattens all modes of expression into a weird culturally en-US biased soup (because that’s where the companies are headquartered and where the language models are trained).
I have no illusions that this trend will reverse - high quality translation work is skill and time consuming, and thanks to LLMs anyone on Earth can now localize anything they want in any language they want for ~free in ~0 time.
The weirdest part is seeing this bubble up to the real world. I’ve been hearing young people use turns of phrases/expressions that I recognize as distinctly American, except not in English.
The classic linguist response to this, which I subscribe to, is “no language is fixed, language is ever evolving in response to various external cultural pressures“. Which is true. But it doesn’t make our post-LLM language landscape any less weird.
I can read both just fine. Platforms defaulting to always showing one or things like youtube auto-translating titles all to English or all to Spanish is frustrating because I always have to do the math in my head as to "Why does this thing I'm reading sound weird as hell" and realize its because it was lost in translation.
Hell, I watch creators/consume content where the creator or writer themself speaks/writes interchangeably in both languages often within the same sentence because Spanglish is very common, and that just destroys most of these automated generators brains.
Calquing has been a common thing since long before AI translators, and it's not notable that it now happens for modern memes. It happens whenever a language is notable and nearby; English has a lot of calques from Greek/Latin/German/French as a result.
Ironically, "calque" is a loanword, but "loanword" is a calque.
It probably made a lot of sense in certain contexts, and certain side effects weren't predicted, or it just has bugs that need fixing.
Presumably nothing malicious or stupid. But just ironing out the kinks.
Would you like to hop on a quick call to chat about this further?
Just a quick lil call.
Quick lil ol' callerino.
Hoppity hip hop.
This issue is specifically centred around the human element of the work and organisation. The translators were doing good work, they wanted to continue that work. Why it's important that the work done is by a human is probably only partially about quality of output and likely more about authenticity of output. The human element is not recorded in the final translation output, but it is important to people that they know something was processed by a human who had heart and the right intentions.
Not that I entirely disagree with the conclusion here, but…
It feels like that same sentiment can be used to justify all sorts of shitty translation output, like a dialog saying cutesy “let’s get you signed in”, or having dialogs with “got it” on the button label. Sure, it’s so “human” and has “heart”, but also enrages me to my very core and makes me want to find whoever wrote it and punch them in the face as hard as I can.
I would like much less “human” in my software translations, to be honest. Give me dry, clear, unambiguous descriptions of what’s happening please. If an LLM can do that and strike a consistent tone, I don’t really care much at all about the human element going into it.
His demand of not using his existing work for AI training is nonsense. Because the entire article is stated:
> Portions of this content are ©1998–2025 by individual mozilla.org contributors. Content available under a Creative Commons license.
Didn't he agree on that?
So, this contributor revealed he doesn't understand the license his work is published under. As such, Mozilla must refuse his contribution because he don't understand the idea behind Creative Commons license. His wish granted I guess.
If mozilla wants to tell him that his work was valuable and therefore has grounds to sue him for rescinding the license, they will have a lot of difficulty proving that after their sumobot summarily deleted years of it for no good reason at a whim.
Good for him. He should probably consider suing them for destruction of his work.
You can make updated version of your work to non-CC, but the version you published under CC is CC.
But I'm not a lawyer so I don't know and in real business, they casually use the word "license" in Japan. But in my opinion, everything is contract under Japanese law.
My problem with this type of gate keeping is that machine learning does open up translations that are accurate to the masses. It is quaint having a real human do your translations though. Kind of like having a real human drive your car or do your housework. Not everyone can afford that luxury. But, on the other hand, having a singular organization own the training data and the model and not publishing the model itself is where the gatekeeping continues.
And the fact 20+ years Mozilla contributor didn't understand it too. You can't restrict the usage to things you don't like it under CC.
CC licenses (and some other foss licenses, e.g. Apache 2.0) are explicitly irrevocable... which is probably enough for US law though I still wonder to some degree if there isn't some country that would take issue with that term... especially a country which recognizes "Moral rights".
Some other FOSS licenses (GPL for instance) contain explicit terms allowing revocation under certain circumstances (but otherwise claim to be irrevocable).
In particular, the primary purpose of AI as we know it is to strip off attribution, which is explicitly forbidden by basically every license in existence.
To nitpick "explicitly forbidden" isn't quite right. Licenses basically only grant more permissions, they can't remove them. It's explicitly excluded from the rights granted by the license, but it's not explicitly forbidden because it is the law that might or might not forbid the activity, not the license.
Talk about being tone deaf. This was so incredibly rude. No consult, no request whether they wanted this or not. Mozilla keeps finding new ways to shoot itself in the foot, these are probably some of the most loyal people that you could wish for, that's a precious resource if there ever was one. And to add insult to injury they want to them 'hop on a call' and to 'trully[sic] understand what you're struggling with' even though they just spelled it out as clear as day.
Free tier Gemini CLI literally writes Android app for me by just endlessly wondering in English. AGI's here. And it struggles with Japanese. How!?
It doesn't mention mistranslating, so it's difficult to know the root of the problem is AI "struggling".
> It doesn't follow our translation guidelines. > It doesn't respect current localization for Japanese users, so they were lost.
I believe this is the root of the problem. There are define processes and guidelines, and LLM isn't following it. Whether these guidelines were prompted or not is unclear but regardless it should've been verified by the community leaders before it's GA'ed
barbazoo•3h ago
Could this have been a mistake rather than a malicious act?
alcide•3h ago
benatkin•2h ago
Incipient•2h ago
Is there a slightly different phrasing that would make this better, or is it the sentiment that's crap?
"I'm sorry for how these changes impacted you"? Personally just the sentiment feels insincere to me haha.
pseudalopex•1h ago
mewse-hn•1h ago
However, "let's hop on a call" is just additionally dismissive.
4bpp•1h ago
* The infantile corporate-cutesy wording "hop on a call" is not appropriate when talking to somebody who feels that you deeply wronged them. It has the same vibes as cheery "Remember: At Juicero, we are all one big family!" signatures on termination notices, and Corporate Memphis.
* In the first sentence, Kiki says "about the MT workflow that we just recently introduced". Why is this level of detail shoehorned in? Everyone in that conversation already knows what it is about. It's as if Kiki can't resist the temptation to inject an ad/brag about their recently introduced workflow for any drive-by readers. "I'm sorry you were dissatisfied with your Apple(R) iPlunger X(TM), which is now available at major retailers for only $599!"
petre•49m ago
kentm•31m ago
The change did not fall out of thin air. It was something they did. If they do not own it explicitly then it’s insincere full stop.
crazygringo•2h ago
They don't know what exactly has gone wrong. All they can say sorry for is for how the person is feeling. Then they want to get on a call to learn more. Which is the start of helping.
The response is as sincere and helpful as it could be for an initial response from someone who wants to figure out what the problem is.
rileymat2•1h ago
crazygringo•1h ago
E.g. literally the first bullet: "It doesn't follow our translation guidelines". OK -- where are those guidelines? Is there a way to get it to follow them, like another commenter says works? Does the person need help following the process for that? Or is there a bug? Etc.
These are the things a call can clarify. It's the necessary first step, so why are people complaining?
handoflixue•1h ago
It's entirely possible that such information is well-known to everyone involved in the translation community.
I would consider it outright insulting if someone who ostensibly "wants to help" doesn't know basic information like that - if the people making decisions about SumoBot are NOT aware of basic information like "where to find the local translation guidelines" then they are presumably not qualified to release a tool like SumoBot in the first place.
kentm•25m ago
They should have understood the guidelines before turning on their machine translation in a given locality.
kentm•29m ago
BrenBarn•29m ago
layer8•2h ago
omoikane•2h ago
move-on-by•1h ago