> How many personal computers in 1994 still had the ability to boot after the OS was trashed?
Every Acorn Archimedes computer, since the entire OS (not just a rescue system, a full, graphical OS) booted in a couple of seconds entirely from ROM. It's the only computer I know of with a fully featured graphical OS that was fully functional without any kind of disk.
The closest I saw to this was my Dads Tandy laptop in late 80s that had no hard drive - the OS (DOS) was ROM and the file system was RAM, you could boot up with no discs inserted. In retrospect it was a great computer for me to learn on since I could poke around freely and there was no chance of bricking it.
I believe many Atari ST models had their graphical OS (TOS/GEM) on ROM. The Archimedes with RISC OS felt revolutionary at the time, however. Several of the graphical desktop programs were written in BBC Basic and you could look at the source.
No. The equivalent to firmware was in ROM in later models, but the OS was loaded from disk.
It was unusually complete firmware, comparable to the Mac Toolbox, but you could not use the computer in any way without an OS that had to be soft-loaded.
The Archimedes was a full multitasking GUI OS, in ROM. No disk of any form needed. It could join a network and load apps and save files to a server with no local storage media even installed in the workstation.
This is why Oracle used it as the basis for the original network computer:
The Pace company, better known for modems and set-top boxes, ended up owning a fork of RISC OS for this purpose. That fork is what led to the current fully-32-bit version and then, later, to the FOSS release.
The Amiga 1000 (that came before the 500) didn't have the "kickstart" in ROM, so you needed a kickstart disk for the 1000.
The Amiga 500 and later had the kickstart in ROM and many of us would mod our Amiga 500 so that we could use a switch to select between kickstart 1.2 or 1.3.
But even on the Amiga 500, that still wasn't a UI from ROM: you had to use the "workbench" disk to get the UI.
I remember a thing called menuetos, written mostly in assembly and fitting a single floppy drive while still having a decent ui and some drivers. You could probably fit that in a normal bios chip and/or boot it as a efi payload.
Imo it’s mostly about nobody having tried that yet (at least afaik).
I’ve been waiting for someone to do something more interesting with EFI. It’s extremely capable and could easily host a minimal recovery environment that would be invaluable in many situations, particularly on laptops which might be out in the field away from recovery tools when things break.
Not from the very beginning. The earliest live boot CD I remember is "DemoLinux". Back then that was still a major hack. Now Fedora, at least, boots into live mode to run the installer from the full GUI.
Yggdrasil Plug-and-Play Linux supported running off the CD as far back as 1993, but you needed a boot floppy because computers couldn't boot from CD at that point. When you installed it to your hard drive, most of the included software stayed on the CD, meaning that you had ~500 MB of software and source code permanently available without taking up hard drive space. This was useful in an era when 200 MB and smaller hard drives were common. After installing, you could pick and choose which system components you wanted to move from the CD to the hard drive.
Ah you mean the OS is in the ROM, soldered to the motherboard, maybe not even writeable. Sure, I don't know about that, maybe your example is the only one
Booting from ROM was pretty common back in the day. A lot of machines from the "golden age of micros" (late 1970s - early 1980s) would boot right up into a BASIC prompt. A full graphical interface was something else though! That's really cool!
Given the controversy surrounding Proton's potential political leanings, this feels like a major misstep for a company that should prioritize neutrality. It makes me seriously question where Porkbun stands.
It's also really bad timing, given that Porkbun's website at the time of writing appears to have been hacked, and is serving an invalid certificate.
I tried finding a reason to use Porkbun over something like namecheap and I couldn't.
The porkbun UI is just barebones, it isn't better than anything else.
Porkbun's prices are equal to or higher than most registrars.
I don't see configuration options in porkbun that I can't utilize with other registrars/dns hosts.
It seems like maybe porkbun has good customer service? I've had that elsewhere though, so again not a huge benefit for me. Support for domains isn't something I've ever really needed though.
I don't really understand why people are recommending porkbun like they are.
Porkbun is cheaper than Namecheap by a few dollars especially on renewals. Namecheap’s renewal price is semi-hidden/obscured.
All my domains are with Namecheap except 1-2 I was/am testing on Porkbun but I’m considering a switch since it means a hundred or two dollars a year in savings.
Porkbun is $11/yr for .com. Namecheap is the same standard price but namecheap regularly has $6/yr sales. .coms are $6.50 on namecheap right now vs $11 on porkbun.
Visited Namecheap and currently registration is $10 (normal being $15) and renewal is $17 (wtf). The $6.5 is a promotional offer for one single registration. Porkbun is $11 all. So, yes, you get first year cheaper, but by third year Porkbun will be 3/4 the costs.
I use Namecheap for most of my domains, it quite sucks. I will admit that the DDNS functionality is nice (even supported by ddclient out of the box, if you ever need that), the selection of domains and everything else is as you'd expect, but the UI feels really clunky and is almost always slow for me.
I did briefly use Porkbun and have to say that the minimalist UI that actually loads fast was a breath of fresh air and was so much less frustrating than Namecheap. I wonder what other good options for domains are out there. No comment on customer service in either case, though.
I might not want to run my own DNS servers. I might also not want to set them to ones provided by just a single VPS provider for my use case. I've migrated between DigitalOcean, Scaleway, Hetzner, Time4VPS and Contabo in the past and I don't want to have my domains be tied to any one of them, much like I might not always want to have Cloudflare in the mix.
For example, recently I migrated from Time4VPS to Contabo due to better prices and more RAM (I realized that most of the time my servers are near-idle, so some overprovisioning is perfectly fine as long as it leads to money savings), I could just change the A records for specific servers in NameCheap and the CNAME records that point to those for the actual apps picked up on that and it was a pretty trivial migration. Having the NameCheap DNS UI not suck would have made that even more pleasant, vs it being kind of slow and clunky.
That move saved me somewhere between 400-800 EUR a year, which with the salaries here is pretty good!
Avoiding Porkbun because of politics and going to... Namecheap?!
They suspended all Russian accounts because of politics. This means they use the emotional rather than logical part of their brain and have no place anywhere near critical infrastructure. Full stop.
You might like it when somebody gives two fingers to those damn Russians, but their decision to drop all clients regardless of their political leanings or actions did absolutely nothing to damage the regime, while creating significant problems for projects like OVDinfo that already had their hands full providing legal help to antiwar protesters:
Not to mention that most clients moved to Russian registrars instead, this helping to keep more money inside the country.
FUCK namecheap. Their name says it all. If you're still using them, pray you won't run against their politics sooner or later. It's unpredictable, stranger things have happened in the last few years.
I sworn by Gandi since the early '00s, but they sold themselves to a private equity firm recently and things almost immediately went just as one might fear.
The CEO praised one of Trump's picks for an attorney overseeing internet regulation and then when asked about it, the official Proton account replied with multiple posts about how Democrats were sold out to big business and Republicans were the party of small business. Also about how J.D. Vance was the only person to want to talk to Proton about internet regulation. It's worth noting they also left out that the attorney in question lobbied for multiple big tech companies before being appointed and Proton's whole argument was "democrats are captured by big tech".
Then when further asked about it, the CEO essentially said he didn't actually know much about the situation and hadn't meant it as an endorsement despite the original posts essentially being a one-sided attack on Democrats and all praise for the current administration, including @ mentioning Donald Trump. He also kept saying "my comments from last year" when they were from December and the controversy was in Jan.
At best it seems quite naive from Proton's side to think the administration where Musk, Zuckerberg, and Cook were front row at the inauguration, and was generally funded by Musk, Thiel, Bezos and a raft of other VC folks is going to be more aggressive at regulating big tech companies and for internet privacy in general. At worst, they're trying to curry favor with/actually in favor of the current administration and don't want to admit it because a lot of their value as a business is for people who don't want the government to see their data.
Early on, the CEO, owner, of proton, ran a self published glam campain for himself,company, and family.So it's easy to see him attempt to do a little , heh heh ;), social mountain climbing.
And as a user (less and less) of proton,it's been a slog,with technical problems on an android phone, impossible to resolve, years, multiple different phones, browsers....it's on there end, and there technical support , engages, but does nothing.
And the pokbun does not load.
Having personal/business web space with domain email is reasonable, and the lowest tier prices are very cost effective,simple to use, with clutter free, intuitive UI.
Having a domain free's up any possible liability
from switching service or hardware providers and also means that anything you build, or pay to have built is your own intellectual property.Tech support tends to be good, as long as the problems
are actualy technical, and you can provide relevant details, but the learning curve is steep, and nothing is intuitive on the back end, though it's worth it all, for the " IT WORKED!!, Heeeeeyyyyyy! I'm soooo Cool! " moment.
That's really the core of it for me. Literally no one was asking for or expecting his input on the appointment of an assistant attorney at the US DoJ. Trump probably doesn't even know who this is or what their job is. The fact that Proton's CEO decided to talk about it is most of the attention this appointment even got.
Even outside of me thinking it was a poorly thought out endorsement, it's just so amazingly unnecessary. I follow journalists and activists around anti-trust and internet regulation and I literally hadn't heard about this appointment.
Throughout, he doesn't really give any actual justification for why he thought the original post was correct or remotely well informed. I haven't seen anything from him that acknowledges that he's praising an ex-lobbyist as being good for antitrust, or any explanation of why he'd think it was actually a good appointment for his asserted values. The only stated reasoning was "democrats bad, republicans good". There's no discussion of the role Musk and Thiel and Bezos played in electing Trump. There's no discussion of any Biden-era anti-trust action and why he'd think Trump's folks would do more.
It's hard for me to come away from it with a positive opinion of the CEO's decision making and judgement.
I may have missed it in the thread, but I didn't really ever see an argument. Just a "I trust the republicans and their pick". There was no deeper discussion or assertion of why this was a good pick. All of the followup was about how he disliked the democrats and liked Vance.
That's not really a solid argument to me. No part of the original post or the followup helps me understand why he thinks a former lobbyist for Amazon and Google would be for anti-trust action against them.
Most of the responses on reddit, while overblown, are generally pointing out that most of big tech was funding Trump and Vance's election. If most of the CEO's argument is how much the democrats are in the pocket of big tech, that seems quite relevant. A pretty substantial percentage of the money spent to elect Trump was from Musk and other VC folks, and Bezos notably killed an endorsement of Harris at the newspaper he ran.
As someone said in the thread, it feels like the CEO took a "the enemy of my enemy must be better" in this, when it's entirely correct that both US parties are in the pocket of corporate interests. He's not wrong about the Democrats, but that definitely doesn't mean Trump's appointments are better. He could have just decided to sit this out and say nothing, which makes it all the more objectionable that he decided this was his time to show up and publicly try to get Trump's attention to praise him. As far as I know he wasn't @ mentioning Biden to praise the various anti-trust actions that were taken over the last few years.
He wrote, she has a good track record. Which is true.
> My post is talking about Gail Slater, who is by all measures, actually a good pick, with a solid track record of being on the right side of the antitrust issue. Yes, she happens to be nominated by Trump, but her record speaks for itself.
And just looking at where proton puts its mouth and money, it’s obvious they cannot be MAGA, it just does not fit at all. They commit 10% of their benefit to Ukraine, as an example.
Hence, the post was out of touch a bit, badly formulated, but that’s it. Hence could indeed have simple said nothing, I agree.
I mean, maybe, but as folks in the comments pointed out, she literally lobbied on behalf of Amazon and others against regulation on big tech companies. She worked against a number of EFF efforts, for instance.
To be clear, I don't think this is some hidden effort to secretly support Trump, but it's such an obviously bad move that was then defended that makes it hard to understand what he was possibly thinking.
If he'd been deep in the comments talking about how he personally knew this lawyer and laying out why he was convinced she was a good pick, then maybe. But outside of him saying "solid track record", I don't think I saw any actual articulation of the record he's saying she had, and plenty of people calling out the things she did that were as pro-Big Tech as Chuck Schumer's daughters that he called out as marking the decline of his support for democrats.
Which is the additional part of this being horrible judgement. If it was just "This person is someone I think is good" then sure. Maybe he's wrong, but it's a defensible position. Publishing a detailed explanation of why he's disowning an entire political party that had literally just had Lina Kahn doing some of the most aggressive anti-trust prosecutions in decades is wild. There's a space to just cheer for people doing things well that your organization agrees with, but he chose to issue statements condemning entire parties and praising others by name.
It's just American political posturing and cancel culture. It's really annoying how pervasive black-and-white thinking is now in American political discourse that you cannot even say one good or bad thing about some American politician or policy, without every American jumping on you like you have a chosen a side you should be ready to die for!
The spillover of American politics to the internet is the saddest thing I've encountered online as a non-American because they are trying to force us into right- and left- echo chambers they've created to make us think about politics like them.
I'm not an American and I don't want to trust my email to a company that supports a government that is threatening to annex my country.
This isn't culture war stuff it's just common sense. I didn't fall for some us vs them trap, the current US government is frequently and loudly telling me that it's me vs them.
Like I said, black-and-white thinking - some company CEO supporting some particular public policy of an administration doesn't make them a part of "their" camp and hence your "enemy". I support the end of the genocide in Gaza and the war in Ukraine, and am happy that Trump as the US President wants to do so too (for his own political reasons, and in his own idiotic political ways). That doesn't make me a Trump MAGA fan or a right-wing supporter of the Republican.
I agree totally with you that it is black and white thinking -- either you're with us or you're against us.
Canada's sovereignty is not up for debate and we will not support a country that threatens us.
This means that we're not going to be buying nearly as many American goods, not going to be visiting America for tourism and not suffering the fools that you call politicians and diplomats.
As far as the average Canadian is concerned America can twist in the wind. The feelings of betrayal are real and won't be going away any time soon.
The world is tired of the inane histrionics and is moving on from America.
I've been using Porkbun for my meme domains ever since I left the train-wreck that NetSol had become. broken back-ends, inability to update root server glue records, nobody left that knows the back-end, nobody left that understands how to use dig, etc... I split up my domains into multiple registrars based on category to avoid all the eggs in one basket rotting via cancel culture.
Porkbun are one of the few registrars that support IP restrictions on accounts so I can avoid pseudo 2FA SMS or propriety 2FA apps and they have a simple UI. I have no use for Proton however. If I want to send secure email I use PGP in Thunderbird and if I want more security then I use PGP with my own self hosted email servers. I consider PGP real E2EE for the email body vs. servers that have pseudo E2EE by managing the certs using code they have control over. If I want true E2EE for the entire file or message contents then I use self hosted SFTP with encrypted headerless random ciper+hash loopback images.
cryptsetup benchmark
Then I randomly pick one of the three slowest ciphers and hashes along with a 222 to 499 random character strings and use an obscure filesystem on top of it. Multiple of these can be nested each using a different cipher+hash+long pw. There are additional techniques that dork up commercial forensics tools.
I am not seeing the hacked page or bad certificate but your comment was from 18 hours ago as of the time of my comment so they must have resolved it. [1] I hope they release a detailed write-up on the incident.
I got served with an invalid certificate too, just for one hit, then the site loaded normally. Haven’t made a screenshot, but the certificate was issued by crewai dot com.
Between the crypto wallet I didn't ask for and their CEOs pretending like his brainrot ideals don't influence the company after sharing them on company social media, I'm glad I got rid of Proton. It was a hassle as I'd just moved there to get off Google. I'm on Tuta now.
I've never paid attention to Porkbun but I won't be checking them out. Fuck fascists and people who sympathize with them if it's good for their bank account.
I’m not sure if Porkbun still do it, but for a while they were also pushing those decentralized blockchain TLDs alongside the real ones, which feels a bit disingenuous.
the “controversy” is that the CEO of Proton used his official proton twitter account to be excited about the US government appointment of someone that is likely to pursue antitrust sanctions against Big Tech. Fast Forward to now, this is true against Microsoft. He was also amused at Big Tech CEOs kissing up to Trump due to this specific threat of their monopolistic nature.
The CEO removed their posts and the company reaffirmed that was not in line with the neutral stance.
Vicariously invested people act like anyone that finds something beneficial in this administration is supposed to be ostracized.
And that even if the CEO had posted this on a personal account, “accountability” is needed and should affect everything he is a part of and benefits from, in order to cause financial pain, I think forever? I’m not sure what people want or expect, it doesn't feel fleshed out
The CEO posted something, and then the official Proton account also followed up with a multi-post writeup of how democrats were all captured by big tech as if the appointed person in question wasn't a former tech lobbyist.
I don't think it should follow them forever, but for a company a lot of people are looking to to protect them against US government intrusion, picking this time specifically to @ mention Donald Trump and praise him appointing a big tech lobbyist and then have official company account defend him is certainly a bad look even if it mostly got deleted.
Things like email security and VPNs are 99% a business of "do you trust this company to defend you against a nation-state as best they can" and injecting themselves into what, for them, is international politics at very least shows questionable judgement.
If it was a post now about the action against Microsoft, sure. But preemptively praising someone's appointment feels very "kissing the ring" in a way that I don't want my trusted security company to be doing.
Second, regarding the message that was mistakenly posted from Proton's Reddit account, that message was not approved to be posted. I was unaware that it was posted, and I asked for it to be removed as soon as I saw it. We apologize for this failure in internal controls.
So, I think that's referring to a different message though. Most of what I referred to was from the official Mastodon account: https://archive.ph/LlbSj
Either way, it wouldn't have been much better if he'd personally posted the things the official account posted, or if the official account had posted that political rant with a "A response from our CEO:" at the top. Endorsing specific governmental actions would be something I could be fine with them doing, but calling out specific parties and candidates just feels like poor judgement.
>Given the controversy surrounding Proton's potential political leanings
???
Explain?
>It makes me seriously question where Porkbun stands.
It's a cross selling partnership. Porkbun is referring its customers to use proton for mail hosting instead of fastmail or whatever. I'm not sure how that implies alignment of politics.
It's less about stopping and more about how often you have to charge the car. If more cars have to charge more frequently with a longer charge time, then more fast charging infrastructure needs to be built out.
Larger vehicles (buses, trucks) need more battery. Towing requires more battery.
Fewer stops out of the necessity to charge is a good thing.
I thought the opposite - the 'lost cat' sign at https://youtu.be/2UXS6DBD6g0?t=94 is incredibly legible and definitely better than an upscaled 1080p image.
That's because normal people are supposed to use the Gemini chat interface, which has access to the same image generation model as ImageFX, and I'd imagine video is coming.
That's what I thought. And it could be pulicity of Gemini as well that it is so good that it can teach students say 5x faster. If it is Gemini, there isn't any reason to hide. My bet is it is some unreleased Gemma or some model.
Since you're not using the official models (since they're not GGUFs), what exact model are you trying to use? The 3rd party you rely on might have screwed something up.
> They are designed to help prevent our models from generating harmful content, i.e.,
> [...]
> Sexually explicit content
Dear tech companies. Sexually explicit content is not harmful. Why are you all run by puritans? I don't even want to make edgy porn, I just want to be treated like an adult.
It's harmful in that there exists a significant and vocal subset of users who does not wish to see that content or does not wish their children to do so. It's easier to teach your model never to produce that kind of content than to teach it to perfectly distinguish whether this user should see that content or not. TV channels are barred from broadcasting this kind of content for similar reasons.
Sure, there are always jailbreaks, but then the narrative changes from "we made a model that tells erotic stories to children" to "this ingenious teenager figured out a way to hack our model to make it produce erotic stories." In other words, jailbreak move the fault from the model producer to the model user.
It's also worth keeping in mind that erotica comprises a surprisingly large portion of fiction easily available on the internet for free, and "unfiltered" models tend to produce that kind of content unprompted (see e.g. the original Mistral). The major AI labs are probably filtering it out, but I suspect they can't go too far there, as having a model that is good at fiction is something they actually want.
Then there are the non-chat-gpt-app use cases (like customer support chatbots, automatic summarization etc), for which unprompted erotica is highly inappropriate. Those are the "business travelers" of AI, not the first thing one thinks of when talking about who uses AI models, but extremely important nonetheless.
I heard of this described as the minority effect, that a small minority can have a disproportionate impact. The example given is that it's cheaper to make all instances of a product kosher or halal than to make an entirely separate product.
>It's harmful in that there exists a significant and vocal subset of users who does not wish to see that content or does not wish their children to do so
It's hard to think of a scenario where there's a child technical enough to run Gemma 3 locally but somehow unable to access any other written erotica. Project Gutenberg is full of erotic textual content and I haven't heard of anyone calling for that to be banned.
>Then there are the non-chat-gpt-app use cases (like customer support chatbots, automatic summarization etc), for which unprompted erotica is highly inappropriate. Those are the "business travelers" of AI, not the first thing one thinks of when talking about who uses AI models, but extremely important nonetheless.
And how many of these are going to be using Gemma, when Gemini over the API is cheaper, faster and easier to use?
> It's hard to think of a scenario where there's a child technical enough to run Gemma 3 locally but somehow unable to access any other written erotica.
The reason you're struggling to understand is that you're thinking about this logically.
Adult content is obviously freely available to any child or adult with minimum technical skills. What makes LLMs different is that it's "the new thing" and people respond differently to "the new thing".
All of this is true but then it's as easy as releasing censored and uncensored versions of the model.
Then it's up to users (or parents, in the case of children) to choose the adequate version for each purpose. Just like there are child-friendly movies and adult-only movies, and no one beyond fringe puritan crusaders would say that the latter should outright not exist.
Well here you still have the same problem, since they're not gonna release an actually uncensored version, that tells you how to do awful things (or indeed, that tells you to do them).
So then you'd have censored and less censored, and it would still be a matter of where to draw those lines.
True, "uncensored" is not the best term for what I meant (as I'm aware that fully uncensored is not a realistic thing to ask from companies).
What I mean is a model for all audiences and an adult model, and the line would be drawn at the law of the country producing it (if it's something that would be legal to publish for a human author at a website, then it should be allowed as an LLM response). So erotica would be fine, while instructions for making a bomb wouldn't.
Companies release uncensored models all the time. They're called "text" models. I just had llama3.2:3b-text-fp16 give me step by step instructions on how to make a pipe bomb.
I think it's easy to released the uncensored version, it's just the censored version that's likely super super hard.
Since this is just giving the model directly, there's no ability to do any filtering as part of inference, so I would imagine you have to assume the worst (intent) on any input coming into it.
There are also some practical constraints, like any kind of erotic content is completely prohibited in some regulations (like India), so if you want to be able to have access to human labeling or deploy the model under these regulations, you do need to comply.
It’ll get easier once the costs of building foundational models go down and human labeling gets automated. Sit tight, models that’d be creative and amazing at generating erotic content are certainly coming.
> It's harmful in that there exists a significant and vocal subset of users who does not wish to see that content or does not wish their children to do so.
"I have a right to live in a society that perfectly adheres to my personal morals" is not how companies or people should operate in a pluralistic society, despite Nassim Taleb's claim that the intolerant minority wins.[0]
It's funny because the results are in, millennials grew up with pretty easy access to all manner of porn from an early age and the effect has been nothing. Even a reduction in intimacy if anything.
I'm sure the hysterical puritans of the past will come out any day now and admit that they weren't even 1% correct in their assertions.
It's what they switched when confronted with evidence, roll the clock back 10, 20, 30 years though and it was "Will turn them into rapists, molesters, and social degenerates."
No, there's no movement to shut down pornography on the internet. There's a movement to shut down specific websites and make a lot of noise about it but continue consuming pornography behind closed doors.
People like pornography. They'll as soon ban alcohol again (which worked so well last time)
Not all sexually explicit content is harmful in all contexts for sure, but in many contexts it is fairly universally considered harmful (eg content involving minors). Do you have means of distinguishing between the two? Are you suggesting that a company must invests millions into teaching the model where exactly the red line lines so that it can have a conversation close to it but without crossing it? Or you suggest biting the bullet and releasing the model not only capable of generating eg child porn, but also having a >0 chance of randomly discussing it in unrelated contexts? Chance of error is always there, and companies decided that a risk of really bad behavior in benign context overweights the gains. Imho, a decision to not play whack a mole with this land mine is quite rational, esp considering gains vs risks vs costs. Think of it as a cost cutting measure, not as an infringement on free speech. You are free to invest you own money into this problem if you think that's a grave mistake and a missed opportunity. The first project to push the automated generated content moderation against what is considered appropriate in the given context far enough to make it economical for companies to put their guard down could actually be worth a lot if you think there's market for it (eg agents on dating websites? idk, you tell me)
I don't agree that textual, fictional explicit content involving minors is "fairly universally considered harmful". Such content is allowed on large platforms like Archive of Our Own or Japan's Shosetsuka ni Naro. I think "don't think it's harmful, but not willing to defend" is a pretty typical attitude.
They mean "harmful to us", not the users. It's harmful because they live an echo chamber of a single mention of genitals makes all the stakeholders run away. Why do they run away? Because they also have stakeholders, and so on.
Everyone is treating this like corps have anything to gain from an open uncensored model. Switch your view and give me a single argument for it? That random nerds on HN stop jerking each other about what „open“ means? You are just not their target group. Having this discussion every time no matter if the model released is censored or not is just insanity. Bring new arguments or don’t use the models you don’t like. There will be a new sota „tomorrow“, maybe even one open enough for you.
The argument is that it simply improves the product. For instance, Github Copilot is apparently refusing to do anything with variable names like "trans" and anything related to sex or gender, regardless of the intended meaning. That is a serious flaw and makes the product less useful.
That is not relevant to the argument. Censoring limits possibilities. While that sometimes has its uses, the overly puritanical approach American companies generally take degrades the value of their products.
I am talking about an „open“ weight model you are talking about a service. If the service wants to censor that’s fine and on them and their leadership if an „open“ model gets released with censorship it’s not, because it’s just „open, but how my manager likes it“
The lack of NSFW knowledge/capability makes them less useful for content moderation. I've tried to use multimodal models for categorizing images from large, mixed data sets. 95% of the input is safe for work. 4% contains nudity but is not sexually explicit. 1% contains nudity and is also sexually explicit. I'd like to categorize content so that nudity is hidden from users by default and that sexually explicit content is always hidden.
Every model I've tried so far is bad at distinguishing sexually explicit content from mere nudity, and many models are bad at distinguishing nude from non-nude. I don't know about Gemma 3 but Google's large commercial Gemini models refuse (or formerly refused; haven't tried recently) to tell me anything useful about images containing human figures. I assume that this is due to aggressive "safety" measures. On a technical basis, I assume that a model that can distinguish 10 different breeds of dog should also be able to usefully describe images of people wearing swimsuits, nude people, and people engaged in sexual intercourse.
There are models especially tuned for it even open weight ones. llms even multimodal ones are not up to the task. You know what doesn't help the discussion at all? That everyone's response is as usual just about titties.
4 months ago I tried every dedicated NSFW-image-classifier model I could find on HuggingFace or GitHub. They have a high false positive rate on certain kinds of benign content, like close up photographs of hands with painted fingernails, and a high false negative rate on artistic nude photographs. I even tried combining multiple models with gradient boosting but the accuracy barely improved; maybe everyone is training with very similar data sets. At this point I should train my own model but I was hoping to find something capable off-the-shelf, since content moderation is such a common task.
This is what HNers surprisingly seem to not understand.
The risk of the model generating illegal content and then the company getting bad PR from vultures in journalism simply outweighs any benefits of including this content in the training data.
This is also why you will never see the big companies release a capable open weight image or video gen model.
>The risk of the model generating illegal sexual content and then the company getting bad PR from vultures in journalism simply outweighs any benefits of including this content in the training data.
This is completely unsubstantiated. The original Sydney (Bing AI) was violently unhinged and this only drew more users; I haven't met a single person who prefers the new Bing AI to the old Sydney, and for that matter I haven't even heard of anyone using Bing AI for ages now they toned it down. Trust in journalists is at an all-time low ( https://news.gallup.com/poll/651977/americans-trust-media-re... ) and America recently elected an extremely unorthodox president in big part due to the sheer hatred of the media shared by a large proportion of the population. Even the most hardcore social conservatives aren't calling for companies to censor the training of open source models so they don't produce adult textual content even when prompted to do so; it's not a political issue.
Brings an argument from nearly a decade ago ignores everything on google in the last four years. Ofc the „first“ rogue AI drew in more users because of the novelty of it… what a shit argument.
>You are just not their target group. Having this discussion every time no matter if the model released is censored or not is just insanit
Who is their target group for small local models that benchmark inferiorly to their proprietary solution (Gemini 2.0) then, if not hobbyists and researchers?
>> The press and decision makers without technical knowledge are the target group, it doesn’t matter if it’s used in production or not. They need a locally deployable model to keep up with enterprises that are to risk averse to put their data into the cloud and also don’t care that their shitty homegrown ChatGPT replacement barely works. It’s a checkbox.
The press and decision makers without technical knowledge are the target group, it doesn’t matter if it’s used in production or not. They need a locally deployable model to keep up with enterprises to risk averse to put their data into the cloud and also don’t care that their shitty homegrown ChatGPT replacement barely works. It’s a checkbox.
I want to use a multimodal model for manga translation, analysis, and tagging.
If this gives me the "aschually as a ethical safe harmless assistant I can't ..." spiel on anything mildly mature, that would be very disappointing. I'll run a test with Berserk and see how it goes.
I'm not a big believer in abliteration, it seems to always hurt performance. Safety should be handled by a separate system, no need to cripple the actual LLM.
The multimodal models aren't good for this. Refusals aren't the issue (they're fine with BERSERK, though occasionally they'll refuse for copyright). The issue is the tech isn't there yet.
You'll want to use custom models to segment the manga (panels, speech bubbles), OCR the text, translate (gemma punches above it's weights for this part).
That said, I've been experimenting with using Pixtral to do the analysis part with okay-ish results (providing individual panels with the character names) but it'll still mix up the characters when they're drawn differently.
> I'm not a big believer in abliteration, it seems to always hurt performance.
Agreed, it's fun to play with but it increases halucinations. And for creative writing, it makes the model write more compliant characters (they'll give in too easily during negotiations, rather than refuse, etc)
Could probably be improved with more targeted abliteration.
There are very few pro-porn voices in the corporate, tie-wearing environments that have the money to train new LLMs from scratch.
Oh, there are loads of porn enjoyers working in such companies - but traditional professionalism means you leave the porn at home during the work day. It is, after all, NSFW.
So at the meeting where censorship decisions were being made, even a weak argument for censoring explicit content will be accepted unopposed.
You can discuss something kosher and have the LLM misinterpret it as something sexually explicit. Yours or their logs will now have all of this miscommunication, and this is a liability. Using models that can’t generate this content even by accident is good legal decision for many. Same goes for images. Stay safe!
Or perhaps it was removing the curly brackets that improved it more than the damage caused by losing the nsfw content.
Or perhaps the measurement of improvement was biased. If a model doesn't understand the word gay there would certainly be people who would find real world use of the model to be substandard.
Did the assessment of what counts as improvement come from the same community that decided that excluding things with 'gay' was cleaning the data?
The model is open weight, I'll bet someone or the other will abliterate it soon. Maybe you want to do the honors? I have an abliterated Llama running on a server shared with friends and it works great.
This only works until it doesn't. Start with a model that simply hasn't been trained on anything your shareholders find objectionable, and there will be nothing to reveal with abliteration.
The solution to this problem is to make it not work. If there are various technological developments in the world that do and don't have porn, and if such were cases that the common denominator of failures were lack of smoothly graduated spectrum of contents without disruption from casual family safe content to hardcore pornography, the problem will correct itself.
Actually, it will happen naturally and eventually. Just look at Apple Vision Pro which still don't have VRChat support, and compare how deeply DOA it has been to other VR headsets that are clearly nowhere near as important. Or "Metaverse" that were all explicitly SFW.
This effect can even be seen in the Apple App Store itself. Who uses App Store? You flow into App Store through porn-enabled platforms, such as web or social media. No one browses App Store as a content. What does it not have? Pornography.
Generating sexually explicit content can cause reputational damage or have legal risk. Not generating such content is something that many developers are looking for. There is people who may want such harmful content and other players can cover such a niche.
That's a bullshit excuse. The Chinese model creators live in a totalitarian dictatorship where porn is banned and the creators could be arbitrarily jailed, but even they don't go to such effort to censor their open source models (there's censorship on their hosting websites but minimal if you run the models locally).
Filtering is not necessarily a good user experience and comes with a cost to do. Google making a model they expect there to be demand for is not just an excuse.
They don't expect to make money serving Gemma; it benchmarks worse in almost every way than their closed-source Gemini. Believe it or not, one of the main sources of demand for these small, non-SOTA models is people using them for roleplay locally. Anyone corporate has the money to use a bigger, more effective model.
I don't think it's reputation risk of companies at large, but risk to individual developers. "He worked on porn" is such an easy gut logic for terminations. It's in our human instincts. Everyone know that in guts.
The early models were uncensored, but people seeing early llms give meth recipes and how to make car bombs made them quickly get neutered before public release (additional controls, for pirvate info, nudity, swearing etc all come from additional guardrails and improvements of the protection they can offer the company and not end users)
Have an uncensored model loop through nypost articles and ask it to synthesize content from that. Nypost has tons of scandalous content and can easily get spun into erotica by an uncensored model.
It’s unsafe for that reason, so you absolutely needed both censored and uncensored. It wasn’t an accident.
> can easily get spun into erotica by an uncensored model.
A sexualized fine-tune yes, but that's because you have to make them overly horny to overcome the original censorship.
Nothing prevent them to train a model that will have an appropriate level of sexual content (that is, only upon user explicit request) the same way they train it not to have sexual content at all.
The reason they do that is because they are American companies, the same companies who also censored nude paintings and statues from European museums' pages.
Hard to get more puritanical than "if you disagree with my opinion then you're morally repulsive". Not to mention that your argument implies that all traces of sex ought to be scrubbed from the entire Internet? And that that conclusion is the only moral one?
The term "puritanical" doesn't exclusively refer to the existence of or adherence to the Puritan religion, and when it does, the term is usually capitalized. From Dictionary.com:
puritanical [pyoor-i-tan-i-kuhl] adjective
1) very strict in moral or religious matters, often excessively so; rigidly austere.
2) Sometimes Puritanical. of, relating to, or characteristic of Puritans or Puritanism.
It is entirely possible within the parameters of commonly understood English parlance for Muslims, or any group, to be puritanical.
Things do not exist on a black and white basis but there are relevant gray scales to be considered:
It is quite different if we talk about removing any sort of text about body parts related to consensual sexual activities or if we try to censor hard pornography or illegal sexual acitivities. I personally find LLM producing sexual content as text rather irrelevant in the same way that you could go to a library or bookstore and buy a romance.
It is also quite different if your definition of kids goes all the way to 18 years. I don't want my kids not to encounter topics surrounding sex until the become legal adults. They absolutely have to learn about it, and be able to develop healthy relationships to their own body and sexuality and have insights that enable them to understand sexuality in others.
I want to protect my kids from harm, but there must be some balance with other aspects as well.
Millennials, who are now well into adulthood, grew up with easy and readily available access to porn. I know when I was in middle school it was everywhere. Kids would even hand out burned CDs of porn.
Please show the damage that it did to that generation. If you are "sky is blue" levels of correct, the evidence should be everywhere. So please, present it.
If there is no real evidence, reply with "So you are saying we should endorse porn for everyone" or some other strawman along those lines. Thanks.
This is a false dichotomy. We can make tech for adults, and children, either with optional settings, filters, or simply multiple versions, managed by their parents or guardians. It's not tech's responsibility to raise every child.
When we’ve solved the access to explicit content in the rest of the internet wr can come back and have this conversation. Until then teenagers will just go to Reddit or wherever and get it there. If we ban that it’ll just move to sexting on Snapchat which if you have ever spent any time with parents of teenagers you’ll know has a tendency to be screenshotted and distributed.
So you’re arguing for teenagers to be encouraged to share explicit content of minors with each other?
The transformer algorithm was originally intended for AI language translation use-cases, and it excels at this task. It's far better than anything else I've tried.
Except that nearly 100% of the capable models out there refuse to translate swearing, sexual, or violent content.
Legal content! Content you can see in the cinema, borrow from a library, or watch on free-to-air television!
Most models will regularly refuse to translate subtitles, for example, because they've been indoctrinated by Puritan priests to put their fingers in their digital ears and scream la-la-la-la to avoid their pure digital souls from being sullied by the bad words.
Wireheading humanity into population collapse via pervasive sexual hyperstimuli (which is realistically what is on the table here) is basically the definition of "harmful".
This is just silly because it only takes one AI company to defect and start enabling it, and the problem is already pretty bad even without AI.
I think all of the solutions are demand-side, not supply side. I would expect differential reproductive rate trends between populations with and without proscriptions on ersatz reality consumption (i.e. aniconist Muslims, Mennonites, etc.) to accelerate
I don't know about agents, but this finally adds the ability to search the web to the API. This is a very useful big deal.
Kind of annoying that they've made a bunch of tiny changes to the history format though. It doesn't seem to change anything important, and only serves to make existing code incompatible.
Every Acorn Archimedes computer, since the entire OS (not just a rescue system, a full, graphical OS) booted in a couple of seconds entirely from ROM. It's the only computer I know of with a fully featured graphical OS that was fully functional without any kind of disk.
reply