I'm not sure i understand the ethical argument. The subject of the photo consented to it. As far as i know she was not mislead or coerced.
Like if this was a normal modelling shot, that got sold to an advertizing company to sell some product the model didn't personally approve of, i doubt anyone would consider that unethical. Generally the assumption is that if you sell your likness (assuming no coercion or other factors that would invalidate a contract), you don't get to control the image after that. That is kind of the whole principle the modelling industry is built around.
The reason why i think this point is important, is that our entire industry is built on the notion that licenses are forever and there are no take backs. Like Linus can't go, oh i changed my mind, this GPL thing is overrated, i revoke the license so i can become filthy rich. I do not think we should normalize the notion that renegging on intellectual property licenses just because you changed your mind is ethical.
It seems to match her wishes. You don't need an ethical argument to make a kind gesture to an icon, just make a statement and let everyone move on. There's effectively zero cost.
It feels like a flimsy justification but I'm not sure what kind of person would push back and insist on using that image against both her wishes and the wishes of the leadership of IEEE.
Well the explicit justification is not the real justification. That's all. I find once I realize that I'm content, but perhaps I'm just making up stories.
The rationale for continuing to use it is pretty flimsy as well, so it's not like we lose much by doing away with it. It's poorly sourced, being a low resolution 1970s vintage scan of a magazine print rather than the original film negatives, and the rights holders never really gave permission to redistribute it. Surely there are better test images to use even on purely technical and legal grounds.
> We believe that the history of the Lena image clashes with the extensive efforts to promote women undertaking higher education in science and engineering
This sounds pretty easy to understand to me, and like the main reason for the decision.
Rather effective? There are still issues to work out but the gender balance is much better these days than back when, say, it was considered cool to use porn as the standard test image in the whole field of digital images
The first thing when I open that link is a report on "Why Socialism Fails". I'm not gonna trust that as a non-biased source. Show me a peer-reviewed study rather than a blog post please.
That's an interesting thought. In high school I met many girls that were actually interested in informatics but had a hard time keeping up and thus did not pursue a career in tech. The women who did end up studying with me were either:
* extreme gamers
* close to parent with interest in tech (me)
* mocked by males as too stupid to learn informatics
* generally adventurous and did not care about appearance
* highly ambitious/great work ethics
In general, all did have a good amount of self confidence. Maybe it is just easier to end up in the adult industry if you are taught from an early age you are an sex object rather a human being with interests. This might sound extreme but I met males that pretended to treat me as a friend but never thought of me as a human. It is then freaking scary to be all alone in a course with them and without backup like other women.
Does it? The words are grammatically correct. But the reasoning doesn't make sense. A pretty, successful woman takes a photo, women are somehow belittled because her career choice is sexualizing in nature? Different from theirs? I don't see any sense in the decision or reasoning, though I do acknowledge that the sentence can be read and made sense of.
The issue is the image is a crop from a picture in a Playboy shoot. The issue is not a licensing problem, it’s not a technical problem, it’s that the field for 50 years has used as a reference image a crop of a Playboy centerfold. We have to keep justifying changing it based on licensing and on the technical merits because fundamentally as a field computer science is still about 12 years old and can’t seem to understand why using a crop of a Playboy centerfold is both unprofessional and exclusionary, but the actual problem is the image is a crop of a Playboy centerfold that we’re continuing to use despite it being obvious even back then that it was probably a bad idea.
Modern relativists feign ignorance regarding the cost of what was almost universally regarded as societally harmful. It’s literally regressive behavior making decisions purely on the basest chemical responses and emotion and not higher order thinking, reasoning or logic. And then somehow bundling that up and claiming to be enlightened.
I think its debatable whether the cropped image is pornagraphic.
The original situation when the image was scanned was certainly inapropriate.
But i think its a jump to describe the cropped version as pornagraphic.
Whether something is pornagraphic depends on context. Cropped and removed from the original context i fail to see how it is still porn.
It seems like the argument is because its origin is porn, everything associated with it is porn. This seems like a very dangerous argument and a very slippery slope.
This is what I mean by our field is 12 years old. You know all the answers to those questions, you’ve heard them all told to you in earnest by people desperately trying to get you to understand them, and yet you’re still sitting here arguing in favor of this image because it doesn’t directly affect you (except in the loss of talent available to the field, but people have a hard time mourning things they never knew they had) and it’s fun to argue.
On one hand we‘re not supposed to be prude and judge/shame women who make a career out of onlyfans and prostitution, on the other a picture of a woman‘s head with a hat on, consenting to the usage of said picture, is regressive and exclusionary. Maybe it’s your ideology that needs to figure itself out.
On the one hand we're not supposed to shame people for their career choices, but on the other hand we're not supposed to create an atmosphere which makes scientific fields feel like "boy's clubs" by e.g using pictures from porn shoots in journals
What‘s boy‘s club about a woman deciding to sell pictures for an erotic, not even some hardcore porn, magazine? That would imply she was forced into it by men, rather than using her privilege of being born beautiful to make a ton of money
Are you playing dumb? Nothing wrong with a woman deciding to sell pictures for an erotic pictures, I've been pretty clear about that. The "boy's club" part is the part where the (almost exclusively male) researchers then take the erotic image of a woman from a porn shoot and decide to make it the standard test image for the whole field
"Oh my how can we ever tackle this level of nuance", "But in case it helps" makes it clear you're arguing in bad faith instea dof looking for a discussion.
You ask "Are you playing dumb?", but then pretend " I don't know who David is." as if that's possible. It's pushed down high school students' throats. And if you googled it, then surely you'd remember that one time you saw it printed on cheap toilet paper in that hotel in Bangok.
When it suits your point it's "Nothing wrong with a woman deciding to sell pictures for an erotic pictures"
When it doesn't suit your point it's "from a porn shoot" and "THE standard test image" (instead of one of many)
I'm arguing in good faith, I just don't think you are.
I know many people called David. I didn't go to American high school, so no, one particular "David" was never pushed down my throat as the one canonical "David". I've also never been to Bangok, wherever that is (did you mean Bangkok? I haven't been there either)
I did finally google it though and I guess I've seen a picture of that statue before, didn't know it was called Statue. Cool beans
I don't understand what you're trying to say with that last part. There's nothing wrong with a woman selling erotic pictures from a porn shoot. I've been consistent on this from the beginning.
Its literally one of the most important sculptures in western civilization. It is kind of hard to imagine someone being ignorant of it. Sort of like someone saying that they have never heard of Shakespeare.
I don't think they're playing dumb, I think they genuinely don't get the reasoning. I don't get the reasoning either. Good faith discussion and all that.
The reasoning is literally as simple as: there's nothing wrong with being in the porn business, but that doesn't mean porn is appropriate in all contexts.
If a woman does porn (or a man for that matter), that's fine. They didn't do anything wrong.
If a company CEO decides to display porn in a company-wide meeting, there's probably something inappropriate and unprofessional about that.
Where is the contradiction between these things???
Well, for one, the image used in the field is not pornographic at all, it's a woman's face. Second, its been used for over 50 years. If this were really about showing porn in the office I think the old guard would've definitely put a stop to it. It's not about showing porn because the image is not porn. Finally, it is used primarily because it is technically fantastic for its use cases.
It's not some hardcore thing, it's obviously not as inappropriate as the "CEO showing porn in a company meeting" example, but it 100% creates a "boy's club" type atmosphere.
There's this website I use a lot, called rawpixels.net (seems to be down now?). It's an excellent tool for peeking at raw picture data when I work on debugging video decoding stuff at work. However, the default loaded picture is the Lena picture. It feels inappropriate every time I open the website at work, in our open office space where my screen is visible. Because it clearly looks like a picture that's intended to be erotic/suggestive. And if it feels inappropriate to put up on my large screen at work, it's clearly not appropriate as the standard test image for a whole field.
So what? The image used is of just her face. We are all naked under our clothes, is it inappropriate to be naked under our clothes at work? You said it yourself, it was cropped, so as to remove the inappropriate nudity. Problem solved.
Attractive women who are a vast minority should not be made to feel excluded by the majority.
This just goes to further stigmatize women by saying that being attractive and having your picture taken rises to the level of being unethical to publish in a scientific paper.
Attractive women’s should be made to feel welcome in all STEM fields and not shamed for their bodies or clothing choices.
Maybe we should just not use pornographic images for testing purposes. There surely exist tens of thousands of viable test images, which work just as well, but just aren't pornographic.
Not to pick on op personally, but this is sort of a perfect example of the stereotypical techy type just not getting it.
Yes, in the legal framework, everyone would probably be well within their rights to keep using it.
But this stuff is made of people. No one is harmed, mostly everyone sort of feels better in this collective recognition of "this isn't the sort of thing we're into, in this kind of way, let's do it differently."
Now if we could just rename a certain vaguely ableist photo manipulation program...
> Not to pick on op personally, but this is sort of a perfect example of the stereotypical techy type just not getting it.
No worries I appreciate your comment. After reading some of the pro-lena image replies that were kind of icky, i can see why you have this view.
> But this stuff is made of people. No one is harmed, mostly everyone sort of feels better in this collective recognition of "this isn't the sort of thing we're into, in this kind of way, let's do it differently."
To be clear, i understand there are other reasons to object to this image (many have been mentioned in sibling comments), and if those were the only reasons i wouldn't object.
However I respectfully disagree that there is no harm here.
The entire free software/free culture movement is built around the idea of someone releases something then someone else builds on it. This can only happen because you can be sure that the thing you are building on won't go away. The foundation has to be solid for such a system to work.
Any normalization of the idea that people can retract work that others have built on way after the fact, is a harm. Its a threat to the whole ecosystem.
I have no problem with ceasing to use the image for other reasons, but i do believe allowing any normalization of the idea that authors can retract work way after the fact, is a real and significant harm.
Edit: that said, my argument is probably stronger through the lens of the categorical imperative than a harms framework.
Yeah I think the ethical arguments are nonsense. Seems to be men creating offence on women's behalf, similar to when white people were offended on black people's behalf about the word `master`.
But it's also a seriously over-used photo with weird colours. It would be a bad idea to use it for a paper anyway for the same reason that you shouldn't benchmark AI on MNIST.
That quote is from a documentary called "Losing Lenna". Pretty clear that they asked for it and she agreed. It wasn't unprompted.
In any case, she's clearly not offended by it. Can we hear from any actual women who were offended by this image and decided not to go into the computer vision/graphics field because of it?
IMO it's quite patronising of men to think that women are so fragile they can't handle a photo of an attractive women being used for benchmarking.
Same as how it's insulting of white people to think that the racism that black people care about is innocent use of the word "master".
I don't know why people can't concentrate on actual issues. I guess this change is easy to do and makes people think they've achieved something.
As someone who used to work in an imaging research group, we quit using Lena years ago specifically because the women researchers in our group brought it up. We came to the decision to retire it and move on. It's a small change, but it's important to promote an environment where everyone feels comfortable and is able to do their best work. I can't quite understand why people waste their time debating a technically inferior test image.
Fair enough. I don't think anyone is really arguing to keep it any more than they're arguing `master` is a better name than `main`. Just debating the reason for getting rid of it.
If it really offends women then I think it's fine to retire it for that reason (in addition to the technical reasons).
I don't understand the backlash outside some sort of weird "anti-woke" attitude. Even ignoring all the problematic history, the Lena image hasn't been a good test image for years. It's a 512x512 scan of printed CMYK dots and not particularly representative of the types of files and images we share today. This was true even twenty years ago when I myself was using this test image to do wavelet transforms.
75 comments
[ 6.9 ms ] story [ 158 ms ] threadhttps://journals.ieeeauthorcenter.ieee.org/create-your-ieee-...
[0]: https://news.ycombinator.com/item?id=39869439
[1]: https://news.ycombinator.com/item?id=39872242
Like if this was a normal modelling shot, that got sold to an advertizing company to sell some product the model didn't personally approve of, i doubt anyone would consider that unethical. Generally the assumption is that if you sell your likness (assuming no coercion or other factors that would invalidate a contract), you don't get to control the image after that. That is kind of the whole principle the modelling industry is built around.
The reason why i think this point is important, is that our entire industry is built on the notion that licenses are forever and there are no take backs. Like Linus can't go, oh i changed my mind, this GPL thing is overrated, i revoke the license so i can become filthy rich. I do not think we should normalize the notion that renegging on intellectual property licenses just because you changed your mind is ethical.
I don't have any actual objection to no longer using it. I just don't like the justification.
This sounds pretty easy to understand to me, and like the main reason for the decision.
* extreme gamers
* close to parent with interest in tech (me)
* mocked by males as too stupid to learn informatics
* generally adventurous and did not care about appearance
* highly ambitious/great work ethics
In general, all did have a good amount of self confidence. Maybe it is just easier to end up in the adult industry if you are taught from an early age you are an sex object rather a human being with interests. This might sound extreme but I met males that pretended to treat me as a friend but never thought of me as a human. It is then freaking scary to be all alone in a course with them and without backup like other women.
Obviously whoever picked it was a forward thinking person who wanted women to know they would be able to work in the nascent field of digital imaging.
Would you have an issue if it was a picture of Michaelangelos David?
It's wild that this even has to be spelled out but here we are.
But in case it helps: not all nudity is pornographic, and not all non-nudity is non-pornographic.
The original situation when the image was scanned was certainly inapropriate.
But i think its a jump to describe the cropped version as pornagraphic.
Whether something is pornagraphic depends on context. Cropped and removed from the original context i fail to see how it is still porn.
It seems like the argument is because its origin is porn, everything associated with it is porn. This seems like a very dangerous argument and a very slippery slope.
Make up your own god damn mind, but it's pretty clearly pornographic to me.
Oh my how can we ever tackle this level of nuance
You ask "Are you playing dumb?", but then pretend " I don't know who David is." as if that's possible. It's pushed down high school students' throats. And if you googled it, then surely you'd remember that one time you saw it printed on cheap toilet paper in that hotel in Bangok.
When it suits your point it's "Nothing wrong with a woman deciding to sell pictures for an erotic pictures"
When it doesn't suit your point it's "from a porn shoot" and "THE standard test image" (instead of one of many)
I know many people called David. I didn't go to American high school, so no, one particular "David" was never pushed down my throat as the one canonical "David". I've also never been to Bangok, wherever that is (did you mean Bangkok? I haven't been there either)
I did finally google it though and I guess I've seen a picture of that statue before, didn't know it was called Statue. Cool beans
I don't understand what you're trying to say with that last part. There's nothing wrong with a woman selling erotic pictures from a porn shoot. I've been consistent on this from the beginning.
If a woman does porn (or a man for that matter), that's fine. They didn't do anything wrong.
If a company CEO decides to display porn in a company-wide meeting, there's probably something inappropriate and unprofessional about that.
Where is the contradiction between these things???
It's not some hardcore thing, it's obviously not as inappropriate as the "CEO showing porn in a company meeting" example, but it 100% creates a "boy's club" type atmosphere.
There's this website I use a lot, called rawpixels.net (seems to be down now?). It's an excellent tool for peeking at raw picture data when I work on debugging video decoding stuff at work. However, the default loaded picture is the Lena picture. It feels inappropriate every time I open the website at work, in our open office space where my screen is visible. Because it clearly looks like a picture that's intended to be erotic/suggestive. And if it feels inappropriate to put up on my large screen at work, it's clearly not appropriate as the standard test image for a whole field.
That's the picture someone cropped, scanned, and used as the standard test image for all of computer imagery.
So what? The image used is of just her face. We are all naked under our clothes, is it inappropriate to be naked under our clothes at work? You said it yourself, it was cropped, so as to remove the inappropriate nudity. Problem solved.
Your arguments are not very compelling.
This just goes to further stigmatize women by saying that being attractive and having your picture taken rises to the level of being unethical to publish in a scientific paper.
Attractive women’s should be made to feel welcome in all STEM fields and not shamed for their bodies or clothing choices.
I think it is pretty obvious why using pornographic images in scientific papers for no legitimate reason is unprofessionell.
>Would you have an issue if it was a picture of Michelangelos David?
It isn't a pornographic image, but depending on the photo it might be appropriate or not. It is also an awful test image for color images.
Morality police don't.
That's all you need. IEEE no longer consents to it.
Yes, in the legal framework, everyone would probably be well within their rights to keep using it.
But this stuff is made of people. No one is harmed, mostly everyone sort of feels better in this collective recognition of "this isn't the sort of thing we're into, in this kind of way, let's do it differently."
Now if we could just rename a certain vaguely ableist photo manipulation program...
No worries I appreciate your comment. After reading some of the pro-lena image replies that were kind of icky, i can see why you have this view.
> But this stuff is made of people. No one is harmed, mostly everyone sort of feels better in this collective recognition of "this isn't the sort of thing we're into, in this kind of way, let's do it differently."
To be clear, i understand there are other reasons to object to this image (many have been mentioned in sibling comments), and if those were the only reasons i wouldn't object.
However I respectfully disagree that there is no harm here.
The entire free software/free culture movement is built around the idea of someone releases something then someone else builds on it. This can only happen because you can be sure that the thing you are building on won't go away. The foundation has to be solid for such a system to work.
Any normalization of the idea that people can retract work that others have built on way after the fact, is a harm. Its a threat to the whole ecosystem.
I have no problem with ceasing to use the image for other reasons, but i do believe allowing any normalization of the idea that authors can retract work way after the fact, is a real and significant harm.
Edit: that said, my argument is probably stronger through the lens of the categorical imperative than a harms framework.
> ...with respect to the wishes of the subject of the image, Lena Forsén, IEEE will no longer accept submitted papers which include the ‘Lena image’
I imagine its a combination of factors, but it seems like a significant component of this was due to Lena's statements on the matter.
https://arstechnica.com/information-technology/2024/03/playb...
But it's also a seriously over-used photo with weird colours. It would be a bad idea to use it for a paper anyway for the same reason that you shouldn't benchmark AI on MNIST.
In any case, she's clearly not offended by it. Can we hear from any actual women who were offended by this image and decided not to go into the computer vision/graphics field because of it?
IMO it's quite patronising of men to think that women are so fragile they can't handle a photo of an attractive women being used for benchmarking.
Same as how it's insulting of white people to think that the racism that black people care about is innocent use of the word "master".
I don't know why people can't concentrate on actual issues. I guess this change is easy to do and makes people think they've achieved something.
If it really offends women then I think it's fine to retire it for that reason (in addition to the technical reasons).
https://news.ycombinator.com/item?id=25706788
It's time we moved on.
The argument they’ve made is that the image is unethical.
I believe this also does not abridge her moral rights.
If there’s something I’ve overlooked I’m all ears.
Discussion: https://news.ycombinator.com/item?id=39872242