Can Data Die? Tracking the Lenna Image | Hacker News Reader

Can Data Die? Tracking the Lenna Image

pudding.cool

19 pointsBerislavLopac26 comments

Lenna's image is an historical artifact. It is an historical fact that it was used as one of the first test image and has the same historical importance as the Utah teapot for 3d and Tom's Dinner for mp3.

So at the risk of having an unpopular opinion, does it really make sense to censor ourselves by hiding that history? I would agree that we don't necessarily need to continue using it as a test image for further research and publication but when talking about the history of image research and processing I don't see why we should hide or censor it.

It would be one thing if the picture was taken against her will or she was coerced, or even if she didn't intend for the picture to be public. But that doesn't seem to be the case here.

I feel like describing this image as being used without consent is kind of demeaning to people who actually have their images shared without consent.

I think its pretty clear you're consenting to have your photo shared widely if you agree to model for a magazine.

I sort of agree with you, but this particular case is much trickier.

She agreed to the photos in an era when the magazine would be printed once, and reproduction of those pictures was pretty limited. There’s something of an expiry date attached with the whole process.

Instead, she got caught up in the whole computer graphics revolution. She’s the subject of one of the very first widely distributed digitised photos, published in decades of academic journals, and so on. It’s a pretty big difference from what she thought she was consenting to.

This is a good point. Its easy to forget how much more ephemeral the pre-internet world was.

However, at the time, playboy had a circulation of over 7 million, was known for publishing literature of note, etc. Maybe the computer graphics aspect was out of left field, but the magazine itself was notable enough that being an object of academic study wasn't exactly that surprising, and similarly reproducing it in the form of an anthology was probably likely.

She was unaware of the spread for decades (obviously), and then later was fine with it. Since she has apparently changed her mind about that, I'm certainly fine with no longer using it as a sample, but that's a more recent phenomenon.

> To make sure we weren’t contributing to the spread of the Lenna image, the large Lenna “image” you see on this page is actually a collection of colored pixels.

collection of colored pixels? This sounds exactly like they are spreading the image, but in low resolution and obfuscated form.

Proper form spreading image without actually spreading it (mostly because of copyright) is to hire a painter and let them create sketch of that image.

> Proper form spreading image without actually spreading it (mostly because of copyright) is to hire a painter and let them create sketch of that image.

That doesn't neccesarily evade copyright.

The point isn’t evading copyright, it’s avoiding propagating the image any further.

I mentioned copyright because the gp mentioned it as a reason, but unless the rendetion is extremely abstract, i doubt it prevents propagating the image either, except in the most literal sense.

Lol, I assumed that was a joke but it appears the article says this in all seriousness.

This "text" you see in this comment is actually a collection of rendered glyphs.

I think the idea is that it's not a file to be crawled, saved and archived as a distinct entity the same way the image is.

Didn’t the model sign a release to the magazine? The magazine then owns the rights and chose not to attempt enforcement of copyright. If all the images to pick to illustrate personal rights over one’s image… this seems like a poor choice.

Yes, in Nemesis, although a little of him survived until the end of Picard Season 1.

I realise this is mostly about the Lena image but to answer the question "Can data die?", I think the answer is Yes, but only if it is permanently erased or contained in a medium that is rendered verifiably and permanently inaccessible. For as long as it remains accessible, data remains 'alive'. There are technical measures that can ensure data is permanently erased or made permanently inaccessible, but both need the backing of good legislation, otherwise they will never happen. I agree with sersi that this image is a historical artefact and a case can be made that it should be treated differently but overall, legislation mandating time limits on data would be a good thing. The problem is how would you enforce such legislation? It's so easy to make copies of data and so easy to hide said copies. Perhaps the real issue is the public or commercial use of unconsented data. Putting an end to that would be a good thing. The conundrum about this particular case is that Lena, the data subject, does not consent to the use of her image but the owner of the image (presumably) does. Is this something a judge should resolve?

Dupe. https://news.ycombinator.com/item?id=29084492

The real question is - HOW can we control data that can be screen-shotted like images? Is it really a "genie out of the bottle" or is there some technical way that we can evolve file formats to ensure the owners of an image retain control of the rights to use and reproduce that image?

Well she sold the rights to the image. Even if technical restrictions were possible, it wouldn't matter because she isn't the owner of the image.

As far as technically - given this was published in a paper magazine, it would be pretty impossible to prevent duplication - the so-called "analog hole"

ignore the specific case here - I am talking about the general case...

Fair enough - DRM is a thing. Its not typically the most robust thing, but people have certainly tried and modern cpus have secure enclaves that make implementing such a thing sort of fesible (see for example widevine).

However ultimately at some point, you have to convert the image to photons to look at it, and its pretty hard to prevent someone from taking a photo of those photons. I guess you could require every camera to auto recognize copyrighted work and refuse (think EURion constellation or Cinavia) but that is pretty orwellian and you still have plain old analog cameras.

Centralized kill switches for information? Sounds like the dream of a tyrannical government, pretty dystopian and not something anyone should want.

no - a file format (or something else) that enforces rights of the owner of the data.

It’s all the same. For it to be possible for an image to stop you from taking a photo of it, your camera would need to be complicit in the scheme. That is just fodder for tyranny.

Nobody publishes RAW files, you edit them and publish a jpeg. That means you don’t need the camera to be complicit, you just need to add that DRM step to your editing process.

I don’t think you understand what I’m saying. As long as you can capture an image of something, be that by capturing your screen buffer a.k.a. screenshoting, or just pulling out your camera and taking a picture, any DRM scheme for images is circumventable by a child unless your device being complicit in the censorship.

The core point of the article is an image that was scanned in the 1970s. Preventing the spread of something like that requires a system where your devices don’t do what you want them to.

Who said the owner has rights to stop me taking a picture?

Just because you have copyright does not mean I cannot make a copy for purposes of journalism, critique, inveatigation, law enforcement, etc.

Thr idea that my liberty should be limited to help copyright holder make money is obscene.

Reply on news.ycombinator.com