Does Someone Know My Name? | Hacker News Reader

Does Someone Know My Name?

blogs.loc.gov

272 pointsTomte89 comments

Very slightly related: One of the most satisfying interactions with a public organization I've ever had was finding an image in the LoC collection of Shasta Dam misidentified as Fontana Dam in North Carolina. A reverse image search revealed that it had apparently been that way for years, and was used in all sorts of contexts by third parties as a picture of Fontana Dam, despite the presence of a 14,000 foot snow-capped volcano in the background, a geographic feature which is notably absent from North Carolina.

I emailed the contact address, and not only did they respond promptly, they corrected the listing and added a note about the previous misidentification and emailed me to let me know that it had been corrected.

https://www.loc.gov/item/2011630316/

I found a similar misidentification (although with less spread of a photo which claimed to be Graham Greene but was some unknown individual:

https://www.dahosek.com/ceci-nest-pas-graham-greene/

https://elpais.com/cultura/2019/10/06/actualidad/1570357128_...

Brilliant! Thanks, the Ransom Center people didn’t know who it was either. They’ll be happy to have this info.

The National Archives is equally incredible and responsive: I made a Twitter bot[1] to automatically post photos from the DOCUMERICA project[2], and found handful of missing scans (just a few dozen in over 10,000) in the process.

I emailed them to ask about it, and they responded in a handful of days[3].

[1]: https://twitter.com/dailydocumerica

[2]: https://en.wikipedia.org/wiki/Documerica

[3]: https://blog.yossarian.net/2021/10/25/A-small-documerica-twi...

I emailed the National Archives about the nature of a quote from an HN comment ages ago. I spent a lot of time digging for sources and got stuck at a “box” somewhere in the Archives.

I emailed the Archives and was politely informed that I would need to visit in person to ascertain the exact document.

A few days later an Archives staffer emailed me under secrecy, saying they were not allowed to be doing this, but then provided exact details for what I needed.

It was truly an excellent internet moment for me. Sometimes the government is cool!

A story with further details desirable.

Trying to honor the unnamed Archives person's request for anonymity :) you can find it in my comment history if you search back far enough

So much of the government relies on staffers discreetly working around the kafkaesque bureaucracy, and everyone involved turning a blind eye and respecting their need for discretion.

One thing that most people don't have an appreciation for is how effective the non-political departments of the US Government are, and also how at risk such jewels are for backsliding into incompetence and corruption if poorly managed.

That’s right, because many departments have their effectiveness undermined by legislation with the intent of undermining the existence of those departments altogether.

The Library of Congress was not doing enough to digitize their library: https://www.theatlantic.com/technology/archive/2015/06/hirin...

The Internet Archive and Google Books have a significant head start over the largest library in the world

Note that a lot has changed since 2015. The LOC's Strategic Plan (https://www.loc.gov/strategic-plan/) and especially the digital strategy (https://www.loc.gov/digital-strategy/) are trying to address that gap — under the “We will throw open the treasure chest” section:

> The Library's content, programs, and expertise are national treasures – we are dedicated to sharing them as broadly as possible. The growth of the Library's digital content, which includes our collections, has increased exponentially every year. We will make that content available and accessible to more people, work carefully to respect the expectations of the Congress and the rights of creators, and support the use of our content in software-enabled research, art, exploration, and learning.

> Exponentially grow our collections

> The Library will continue to build a universal and enduring source of knowledge and creativity. We will expand our digital acquisitions program, as outlined in Collecting Digital Content at the Library of Congress; continue our aggressive digitization program, which prioritizes the Library's unique treasures; and improve search and access services that facilitate discovery of materials in both physical and digital formats.

> We will expedite the availability of newly acquired or created content to the web and on-site access systems. This will mean making improvements to the procedures and tools we use to move content from acquisition or creation to access, which will be critical as we continue to experience exponential growth in the size of our digital collection. We will also improve tools to make it easier for Library staff to enhance content after publication, such as adding additional description or information about the conservation of objects.

You’re comparing two very different things there. Institutional libraries and government archives have records going back centuries, many of which can’t be found anywhere else. Even your local library might have some obscure local newspapers on microfilm from 100 years ago, that haven’t been digitized yet (we’ve admittedly made progress on this problem over the last decade).

To answer these “who am I” questions posed in the article, of unidentified historical photos, we need institutions like the Library of Congress digitizing their records, uniting them with other institutional datasets, and organizing them in a way that we can run facial recognition algorithms on them. The answer to “who are these musicians circa 1930” probably comes from a local newspaper that ended publication 50 years ago.

In case it isn’t abundantly clear, this is going to take decades to accomplish. These institutions operate on timelines measured in centuries, not weeks. They care deeply about problems like bit rot when replacing physical archives. If it is lost to the world in mere decades, it is not useful to them.

The Internet Archive is of little help here, beyond sharing technical information for efficient archival and retrieval of truly massive public datasets.

> You’re comparing two very different things there.

This is such an important point. Governments have longevity and universal service requirements that others just don't. I am personally a charge-ahead-and-try-shit kind of person. But I recognize that works for me because I can easily say "fuck it" and move on from my experiments. Anything the LOC does they're stuck with, possibly for centuries.

So I'm glad that they proceed at a pace that seems positively glacial to me. That's what success on those timescales often requires.

Furthermore, there's digitizing things and there's providing unlimited open access to those digitized things. The LoC needs to tread lightly on the latter to a degree that other institutions may choose not to do. Just because they're the LoC doesn't mean they can throw open digital access to anything they feel like.

Why not?

Because the LoC can't just commandeer the book you just wrote or the film you just made and release it for anyone to see for free.

As the agency that administers copywrite in the US, they can in fact do pretty much what they want to on the subject.

Is it really your contention that the US government can do whatever they want with your property?

Just because they administer copyright doesn't mean they simply get to ignore it themselves.

This is true because the non-political departments of US government have ultimately very limited power, so there is little incentive to either capture them or destroy them by partisan actors. Political operatives would much rather control something like FBI or EEOC instead of LoC or USFS. This should motivate us to ensure that each department’s capabilities are as restricted as possible to achieve its mandate, so that Library of Congress can focus on archiving materials, and not, say, drafting rules on what is required and banned from public libraries across the country (a very contentious topic these days, apparently).

Of course, the political operatives want the opposite: they want to concentrate and extend the power of government agencies as much as possible. The incentive to do so for administrative agencies is strong: elected politicians can be voted out, so they have to be more careful, but career bureaucrats are effectively not removable. The result is a fight for the control of the agency, and then, if one side wins the control, to diminish the power/funding of the agency, which obviously is not conducive to agency being effective at its stated goals.

Good description of the essence of democracy and the path of it sliding into authoritarianism.

In a democracy all of the institutions of government and society are relatively independent, with power spread out and relatively balanced between them. This goes for the Executive, Legislative, and Judicial branches of govt, as well as the civil service, the press, the academy, industry.

In contrast, as politicians start controlling "something like FBI or EEOC", and start forcing the justice department, the press, academic institutions and other government and societal structures to bend to the will of the leader, you soon have authoritarianism.

Once it goes that way, it is very difficult to restore a democracy.

I would argue that's an inevitability with democracy, since short term thinking tends to dominate.

You are certainly right, it is not constantly and actively maintained.

After the Constitutional Convention in 1787, it's said that when Ben Franklin was walking out of Independence Hall, someone yelled: “Doctor, what have we got? A republic or a monarchy?”

To which Franklin responded: “A republic, if you can keep it.”

Wise Words.

And, as Plato said: "The penalty for good men not participating in politics is to be ruled by evil men.".

*oops, "...+IF+ it is not constantly and actively maintained."

Huh I always thought of the Library of Congress as having pretty strong political power -- under the DMCA Section 1201, every 3 years the Librarian of Congress enacts new rules proposed by the Copyright Office (part of the LOC) and can broadly determine that entire categories of software are not illegal DRM-circumvention devices.

They see that power as very, very narrow. For one, they have to renew exceptions every 3 years, for another, they've shied away from doing anything at all to some of the broader categories that have been proposed.

Note the recent mission creep of the CDC where they tried to extend eviction moratoriums.

What a great example. Only reason this did not happen is because the Supreme Court managed to call foul, belatedly though as it had. Can we always depend on Supreme Court doing the right thing and limiting the government’s overreach? The answer is, clearly no, as history gives plenty examples of SCOTUS rubber-stamping clear overreach. The best example is, I think, Wickard v. Filburn, which opened the floodgates, and gave the federal government powers that the writers of the constitutions would have never believed. There are more recent examples as well, for example Kelo v. City of New London, which ruled that government can kick you out of your home and give it to someone else, if it believes that it will create jobs and bring more tax revenues, even if it doesn’t end up happening.

I've been to both these dams. Searching DDG for "Fontana Dam" still brings up a bunch of pictures of Shasta Dam. I wonder how long the correction will take (if ever?) to work its way through the Internet? How long ago did LoC make the correction?

The LoC is a great way to fill up your RSS reader with above average quality content: https://www.loc.gov/subscribe/#blogs

When I saw that this was from the LoC I really hoped to see some photography from the Prokudin-Gorskii collection [1] but this seems to be mostly “iconic” photos from Western media. I’d be a lot more interested to know who the two cossacks were [2] and what they thought about at the dawn of the first world war or literally anything about the shepherd’s boy that randomly stopped for a photograph in 1910 [3], not knowing he’d be immortalized as the first shepherd caught on a color camera.

The whole collection consists of thousands of (mostly) full color photographs of Russia taken on color plates between 1905 and 1915 by a single man and its well worth a look. When the plates align perfectly, the detail in the photographs is nothing short of stunning

[1] http://www.loc.gov/pictures/search/?st=grid&co=prok

[2] https://www.loc.gov/pictures/resource/ppmsc.03940/?co=prok

[3] https://www.loc.gov/pictures/resource/ppem.01541/?co=prok

Definitely don't know the people in the photos, but I do love the work the LoC does - it's so delightfully broad and wonky.

I learned recently that the LoC has a whole bunch fascinating blogs: https://blogs.loc.gov/ I find the "Preservation" one most interesting personally, but they've got a huge variety of topics.

For asking for help identifying the people, those are very small resolution images.

You can remove resolution from the image URLs to get a slightly better version.

However you're right. People hosting this blog clearly are still in the dial-up era.

But then again, archival/historical institutions do this all the time for some reason, that is not providing original scans.

I wouldn't say a resolution from 300x244[0] to 3023x2456[1] is "slightly better." All things considered, it's pretty standard to use thumbnails in an article, but it definitely would be nice if they made the thumbnails directly link to the original resolution image.

[0] - https://blogs.loc.gov/now-see-hear/files/2021/11/48-300x244....

[1] - https://blogs.loc.gov/now-see-hear/files/2021/11/48.jpg

Very nice! You can clearly see something like "Sesostris Temple" on the background award thing and maybe "Honorary Big Brother?" Honorary Elk Member"?

That may be an angle for tracking it down, the man seems to be behind "his desk" which would mean the award applies to him.

https://www.sesostrisshrine.org/ looks like Nebraska area? Perhaps someone at that group might have some idea who he is.

Just finding out who made the scary tie-pin might be enough to identify this person!

The caption in the blog says this:

We’ve been all across the country and back trying to ID this man at his desk. Yes, that’s a Shriner’s certificate on the wall behind him. It was issued from the Sesostris Shrine located in Lincoln, NE. And his stationary says “Memo to Recorders.” The folks at the Lincoln Shrine have been very helpful to us be we still have not yet been able to ID this man.

Trying to decode the letter on the desk... The resolution is such that it's all guesswork, but it's clearly a "Memo to Records" or Recorders[0] which I think of as something for a "permanent record". Gov't records, medical records, etc.

And possibly the image of the person on the letterhead is the signer. So, his company? Gov't Position? Law/Medical Practice? I'm bad at this.

[0] https://imgur.com/a/YBFmPTL

Looks like "Honorary Life Member" if my squinting is working.

Interesting how around his right ear and along the top of his head it looks like his picture was cut out at some point.

I don't think that's cut & paste, I think it's dodge & burn. It's something you do while printing a negative that emphasizes the subject, in this case the man, and de-emphasizes the background. I was a news photographer in the 1970s and we did this all the time.

This is a fascinating process I’d never heard of before. Thank you for sharing. If there’s any good sources for descriptions of the photo processes like this you’re aware of, I’d be hugely interested as well (maybe write a book :)

Check out Ansel Adams’ series of books on photography, particularly “The Print”

I think Technology Connections on YouTube has a recent series on photo development process that mentions this.

I'd recommend "The practical photographer" by Ernst A. Weber

There are dozens, if not hundreds, of books about photo techniques, it was extremely common back in the days

Looks more like digital artefacts from a bad scan

Wait... You were given that 300x244 version before mucking about? I was given 1024x832[2] using latest Chrome for Android 13. Your point remains, but to a far lesser degree, in my case.

[2] https://blogs.loc.gov/now-see-hear/files/2021/11/48-1024x832...

> clearly are still in the dial-up era

Modern web devs should be required to spend a weekend with 56Kbps just to experience how terribly bloated the web is today.

If one were to transport these modern web devs to the time of 56Kbs they'd find things were just as bloated given the capacities of the time. The only question is what novel use of <blink> they'd find.

As example, here is A List Apart in 1999 "CODE: Broadband, schmoadband. In the real world, we must trick our pages into loading faster."

https://web.archive.org/web/19981212012511/http://alistapart...

A List Apart is one of those sites that just impresses me with their content. I'm not a UI guy, and from time to time have to remember how to do something I had done somewhere else usually something to do with CSS. There's a specific example on Taming Lists from them that I refer to any time I'm trying to do inline lists that makes them play/feel nice other than display:inline.

If I'm ever searching for a CSS something and receive A List Apart link, it gets clicked

Yes, but... the best practice today is lazy-loaded thumbnails linking to the real deal. Or opening a lightbox with a 800-1000px or so image and a full-sized download link below it.

Meta did this on Tuesdays:

https://www.theverge.com/2015/10/28/9625062/facebook-2g-tues...

> You can remove resolution from the image URLs to get a slightly better version.

> However you're right. People hosting this blog clearly are still in the dial-up era.

Or do they just have their blog software misconfigured? I have a feeling the scaling is being done automatically.

I found the stories of the solved pictures in 2018 to be a great read: https://blogs.loc.gov/now-see-hear/2018/05/hard-won-victorie...

I would have expected the Library of Congress to have thousands or even millions of unidentified photographs, so I'm curious what the reasoning is behind these 17 receiving such special attention.

You have to start somewhere?

> “Our crowd-sourcing via this and other LC blogs has proved to be a spectacular success. We started with well over 200 “unknowns” and, now, we have only 18 we don’t know! Thank you all!”

They've been doing this for a while. This post is headlined "One Last Time".

In this 2018 link, 23rd in a series, the LoC says it has "exhausted its currently slate of unidentified stills"

So, they have very few unidentified stills in the Moving Image and Recorded Sound division of the Library of Congress.

https://blogs.loc.gov/now-see-hear/2018/01/photo-blog-23-fin...

If shown out of context, I would totally believe that #16 was generated by one of the recently-popular text-to-image AIs.

"Baby C-3PO in the fetal position, floating in outer space, wearing Beats By Dre headphones. Retro-futurism. Black and white photograph"

Some attempts, started with your prompt then started chasing the theme. Image generation process in the captions. Stable Diffusion and DALL-E did the best in general (the uploaded examples cheated by going two stage, do the starfield first then erase the middle and infill). Midjourney thinks of stars like you're drawing a children's book:

https://imgur.com/a/pO6JFWH

Makes me think AI image golf would be fun. Start with an image, then see who can get to something approximating the finished image with minimum prompts. Nearly impossible to accurately score but it's a fun process.

> ...AI image golf would be fun. Start with an image, then see who can get to something approximating the finished image with minimum prompts....

That would be such a fun game indeed!!!

Beeple ran a little mini-contest on Twitter for people to recreate one of his daily works with AI. The thread has people's results as responses.

https://twitter.com/beeple/status/1559182964587651072

I'm gonna bet that the pic is an interpretation of an image from ‘2001 Space Odyssey’. Basically that child plus Sorayama.

I'm still trying to figure out what "obviously inspired by the word of artist Hajime Sorayama" means.

I wonder if it'll be possible to do a reverse genealogy search from photos in the future. We have huge databases of faces, and I'd wager both Facebook and the department of transportation probably have sufficient parts of this tree to work with.

Facial phenotype correlates to genes, though stresses of life take their own independent toll.

Obviously some branches will wind up not producing children, but we might be able to look back quite a bit.

Does anyone know of research ongoing in this area?

Seems like an aspect of the field was in the news last month:

https://www.smithsonianmag.com/smart-news/doppelgangers-dont...

It would be wonderful to use a search engine that could find historical photos of my ancestors, if they exist.

familysearch.org has a lot of old photos that people have uploaded. I was able to find quite a few photos of my ancestors there.

As someone who has gotten deep into genealogy, this is why I am of two minds of facial recognition in photos. I'd love a place to find photos of my ancestors that others upload or hell even photos of myself I just happen to be in. On the other hand that is super creepy.

There is one place I have found that does it for civil war photos. https://www.civilwarphotosleuth.com/

I guess with enough of a time gap, it is less creepy? Maybe WWI photos next?

Seems like something the big ancestry website would be able to do. People have been plugging data into that website for decades. I would totally be shocked if they didn't have a way of viewing a literal family tree with all of the images. Probably even make a Sirius Black wall for you if you pay enough and subject yourself to all of the shady website games they play

Their already solved picture puzzles are easier to read: https://blogs.loc.gov/now-see-hear/2018/05/hard-won-victorie...

Thank you, this is fantastic.

I wonder if you could attack this from another angle.

Let's assume we can narrow down a photo to "probably somewhere near LA because movies" - then if we can roughly estimate the age of the person, and when it was taken, we should be able to determine the current age of the contemporaries, or current age of the youngest person at the time of the photo who would recognize them.

So if the photo was taken in the 1960s, and the youngest person "on set" would have been 20 at the time, they would now be around 80.

So where would LA 80 year olds be who might recognize the photo? Put it in front of them (grocery stores posters? HAVE YOU SEEN THIS MAN banners at nursing homes?)

If they could provide a bit higher resolution scans that would help out tremendously

You can remove the resolution from the end of the urls, though they should really provide links to that..

If you like this kind of thing, you might like an investigation Pablo Maurer did for The Athletic last year. He dug up one of the three actors pictured in the former logo of the Columbus Crew, an American professional soccer team. https://theathletic.com/2682819/2021/07/01/columbus-crew-mls...

I can say definitively that the last photo is of 2 women. They're both female. To mee it's plainly obvious, just look closely at them, they're women. That should help narrow it down quite a bit, after reading the article about that photo they said they can't figure that out.

Something about the person in the foreground made me think of Gene Simmons of the band Kiss. After looking through pictures of him, some of which also showed Paul Stanley (also of Kiss), I felt the latter was a possible match. Four people replying to the article linked from below this picture shared this opinion.

Tat I would have the original thought seems rather odd to me, as I know almost nothing about Kiss and I'm not sure I have ever heard of Stanley before. Almost everything I know about Simmons comes from his rather infamous interview by Terry Gross of WHYY.

The man in the ‘Baba Yaga’ photo looks very Mongol, Yakut or thereabouts.

(Coincidentally, apparently there's currently a thriving scene of low-budget cinema and documentary in Yakutia—if my memory is not mistaking it with another one of those eastern republics.)

That's clearly a very yellowface costume.

Are you saying that he has a rubber mask on his face? Because the face is what I'm talking about. The eyes shape, the cheekbones, the nose. Even the hand plays into this hypothesis. I don't need a costume to deduce that someone looks Mongol.

However, speaking of costumes: while ‘a wrinkled old hag as a villain witch’ says Baba Yaga, the long hat is not a feature of cultures in and around Russia, or in Asia in general to my knowledge. Neither old hags nor witches need such hats here.

So all you need to do to identify strangers is snap a photo and mail it to the LoC?

I would expect the valve bald head there too. Was that ever identified?

He is probably in Gabe's basement playing l4d2.

I'd have thought the Valve head was 3D art, not from a photographic source. I guess maybe that level of detail wasn't possible when it was created though?

It was a real guy. There were actually a couple Valve mascots, a guy with a valve coming from his eye, and one with it coming from the back of his head (the more famous one)

Apparently they just found a couple guys off the street to take photos of, but didn't keep their names or anything so nobody knows who they are. Though they did actually end up doing a 3D model recreation of the second guy for Half Life: Alyx (the VR title)

https://half-life.fandom.com/wiki/Mr._Valve

Reply on news.ycombinator.com