Why exactly was this published? I realize this isn't a professional newspaper, but on what planet is it a good idea to have someone write a whole essay about their own work, and how proud they are of it? Adam, you're well within your rights to be proud of the image restoration work you've done. But how does that translate to this getting put in the Signpost, written by you? That you lampshaded it by saying So how does one write an article about oneself while not appearing completely vain and self-promotional? Well, one doesn't, but let's do it anyway because it'll be at least a couple years until the next milestone really doesn't make it okay. I'm thoroughly disappointed in this month's Signpost. Trainsandotherthings (talk) 19:54, 28 November 2022 (UTC)[reply]
It was originally meant to be by someone else interviewing me, but holidays and timeliness proved an issue with availability. It was do this or get very annoyed at an offer being rescinded. Adam Cuerden(talk)Has about 8.2% of all FPs. Currently celebrating his 600th FP!20:22, 28 November 2022 (UTC)[reply]
I don't see a problem with it, @Trainsandotherthings and Adam Cuerden:. I'd be just as interested to read someone's account of how and why they wrote a selection of featured articles. It would be nice to see the Signpost focus more on the making of the encyclopaedia rather than just being a vehicle for some people's opinions on the WMF. HJ Mitchell | Penny for your thoughts?21:16, 30 November 2022 (UTC)[reply]
It's clearly labeled an "essay". Not sure why you have a problem with it. It just goes to show, people will complain about anything, and when they can't find something to complain about, they will make something up. Viriditas (talk) 21:54, 30 November 2022 (UTC)[reply]
"on what planet is it a good idea to have someone write a whole essay about their own work, and how proud they are of it". On a mostly harmless ocean planet, where an essay is defined as "an interpretative literary composition usually dealing with its subject from a limited or personal point of view". Viriditas (talk) 23:00, 30 November 2022 (UTC)[reply]
Or, alternatively, you could try celebrating the hard work and efforts of others and engage with the subject of the piece. But please, continue to police and gatekeep what other people write instead. That’s so much more constructive… Viriditas (talk) 03:00, 1 December 2022 (UTC)[reply]
May I invite you to wax philosophical about the topic of "image restoration"? Obvious image artifacts aside (coffee stains, creases, etc.), there seems to be a wide latitude in your restorations to "be creative", so to speak. For example, in the Illustrated London News, there's what appears to be a misprint in the long double horizontal line under the heading on the right side that you've "fixed". The misprint likely existed in the original printing---it's not an artifact introduced in the scan, it's part of the "original work". What motivated to you fix that? In undoing creases, such as with the Utopia, Limited poster, you may be forced to invent details under the crease that likely was but may not have been there. And the color correction certainly involves creative/aesthetic choices. Of course, I agree subjectively that the restored images "look better" to my untrained eye. What's your perspective on what you think you're doing? Are you "restoring the image to how it (may have) actually looked like when it was first produced"? Are there cases where you think you may have made the image look better than it ever has in the past? Do we lose anything by presenting restored images as if they were the original? Axem Titanium (talk) 22:21, 29 November 2022 (UTC)[reply]
They should just rename the process something like "Adam Cuerden Memorial Featured Pictures" like how people who donate a lot to a college tend to get halls and buildings named after them. casualdejekyll14:15, 30 November 2022 (UTC)[reply]
I would be interested in hearing more about the tech-heavy side of this in a second essay; details about your tools, workflow, and maybe some speculation about where it's all going with the new AI tools increasingly being talked about, and if you've had a chance to play around with them. For example, lots of people are talking about using deep-fake-like software to bring dead images to life, and to give them voices and movement. It may be the case that we are entering a new era where static images become a thing of the past, and a new generation arises that expects an old image to speak, read poetry, sing a song, or recite the Gettysburg Address, depending on who they were in history. I think it would be an incredible and worthy endeavor, like the famous videographer who restores old films of cityscapes from a century ago and turns them into 4K videos complete with added sound. Viriditas (talk) 22:07, 30 November 2022 (UTC)[reply]
Put simply, I think a lot of the AI things are interesting, but they're hardly encyclopedic. You're adding details that we don't have - we don't have a recording of Lincoln; old photography had constraints, like a long exposure time, that means that the base expressions are going to not be the full range, so you're speculating. We don't have the sound of a cityscape from a century ago, we're doing it in foley. Used well, it's a powerful tool, but it's not that much different than an actor playing Lincoln in a play. It's not going to be a perfect reconstruction. Further, we don't have a lot of the things of the past. We aren't going to have the noise of twelve bathing machines at a beach, or the squelch of people driving through that much horse shit on the road. Adam Cuerden(talk)Has about 8.2% of all FPs. Currently celebrating his 600th FP!16:58, 3 December 2022 (UTC)[reply]
We use artist's conceptions all the time on the encyclopedia and elsewhere (astronomy and maths articles, for example) to show the "extrapolated appearance of something for which the actual appearance is not known or cannot be seen". The same would be true for recreating the past. Philosophically, this argument is just one of purism (or conservatism) versus pluralism (or liberalism), and while the purists will always have their place, technology promotes pluralism by its very nature. You can see this everywhere. The latest example in the news this last week was the reissue of Revolver (1966) as Revolver: Special Edition (2022), which was only made possible by creating a new stereo remix using de-mixing technology, creating an entirely new recording in the process. And there are entire fields and disciplines that exist solely to recreate the past and predict and forecast the future, such that the content and the images they create are nothing but artist's conceptions. I don't see the process of photographic restoration wholly removed from these issues. History itself is nothing but an artist's conception of the past, and the same can be said about human memory, which reconstructs events in a crude and often fictional way. While it's nice to think that there's an objective, encyclopedic reality we can lean on for stability and guidance, no such singular reality can truly be found. We are always recreating the past in some form or another, even when we document it with cameras, since we choose to highlight and frame one thing over another. The very act of observing something is a form of bias, and there's no escape from it. Viriditas (talk) 19:50, 3 December 2022 (UTC)[reply]
At the same time, we need to be clear about level of knowledge. And some tools aren't ready for primetime. The further we leave the sources, the less accuracy we're going to have. This doesn't mean animating paintings or photos is bad, but it's not good for us. Adam Cuerden(talk)Has about 8.2% of all FPs. Currently celebrating his 600th FP!21:33, 3 December 2022 (UTC)[reply]
Forgive me, but I think something might have been lost in translation. I’m not talking about animating paintings or photos. I’m talking another bringing dead people back to life and recreating their identities using our sources to inform their avatars, which can then be used for education and entertainment. In the near future, 2D images will be perceived as we view cave paintings today. "Why don’t they move? Why don’t they speak?" kids will say. That’s where the technology is going. Nobody is going to sit down at a desktop computer to study these things. It will be created as needed and presented as if they are sitting on a chair in front of me. I’m not talking about restoring images, I’m talking about recreating the past. This is already happening, of course. It’s called digital immortality. And despite your objections, we are already doing it. We do it every time we write about the past, and every time we upload media illustrating that past. The next step is to simulate it, or in other words, recreate it, beyond the written word and static images. I don’t expect you to agree. But it’s going to happen. Viriditas (talk) 04:25, 4 December 2022 (UTC)[reply]
Well, under those grounds, we've been doing that for millennia. What's, say, Aeschulus's The Persians if not that with the tools of the time? But such things are corruptible. There's plenty of examples of people bending famous people's views to support them; A digital avatar is no less capable of being used for an agenda.
But this doesn't make things bad by default. Thucydides isn't bad because he made up speeches in the style of the people he portrayed to try and convey the impact despite not having exact words available. Gettysburg doesn't cease to be educational for having reenactments. Inherit the Wind isn't a 100% accurate portrayal of the Scopes monkey trial, but it's hardly bad or uneducational for taking dramatic liberties. Hell, for that matter, Roots doesn't lose its importance to encouraging the study of African-American history for being based on a novel.
But there's a difference between a primary source and a dramatisation. The latter might well be highly educational, but it can also be used to push an agenda. For example, in America, there's a lot of looking to the intent of the founding fathers when interpreting laws. A malicious avatar creator can make sure those founding fathers, or whoever they want, all agree with them on the points they care to. We see things like this being done today already. All it takes is not caring about actual facts, or using biased sources to inform your avatar. Or it might get scrubbed of elements. Do parents object to avatars Cole Porter or Tchaikovsky being open about their sexuality?
Avatars aren't some new concept. Having dead people come to life is as old as theatre. There's nothing inherently wrong with them, but, for purposes of an encyclopedia, they're not necessarily any better than theatre. That's not to say there's no educational merit in them, but, by their nature, where records conflict, they're going to have to choose one of the two sides, and that smooths over the roughness of the historical record. They're not bad, they're just not a final word. Adam Cuerden(talk)Has about 8.3% of all FPs. Currently celebrating his 600th FP!21:16, 4 December 2022 (UTC)[reply]
Great work, Adam, and obviously this is great as a Signpost article where the criteria are quite rightly nothing to do with the criteria for Wiki-articles. Of course it's right that we can read and celebrate editorial accomplishments of all kinds. More, please! Chiswick Chap (talk) 19:33, 11 December 2022 (UTC)[reply]
← Back to Essay