I became aware of a huge trove of forgotten 1960s-1970s Atomic Energy Commission videos sitting in a basement at the National Archives on 16mm film. I started filling out forms and personally funded the digitization of one of them [1]. Then, I leveraged how awesome it was to convince a utility in Nebraska to fund the digitization of another 2 (which were relevant to them) [2]. Roughly $900 a pop, but so worth it.
Wow, thanks so much for doing this. And in 4K too!
It's a shame there isn't any govt org that is working on digitising (and making public) all these old govt films.
My pleasure! I think there are just so many obscure ones that they wait for people who care to come along and fund them. I guess it works ok, as long as we come before the films degrade.
Is there really any point to digitizing 16mm film to 4k, instead of 1080p? The initial quality of 16mm film is so bad, I don't see how 1080p is really limiting.
I figured that since the scanners had that option, I'd just take it for starters and find out myself if the FHD version I'd process looked worse or not. It wasn't that much more expensive.
I also got the ProRes lossless format (345 GB) rather than having them compress it for a similar reason. I compressed it with ffmpeg on my computer and am pretty sure it looks fine with h264. If I do more of the I may still ask for ProRes, compress myself, and then discard the ProRes files if it works ok.
Periscope Films does a lot of this, they have a lot of old 16mm government and military films on YouTube. I don't know if they have other stuff behind a paywall.
Yes they're great. They put their watermark on them and post them on youtube, which is huge amount better than nothing. Some of my favorite old vids come from them. You can guy unwatermarked ones from them, which I guess is how they sustain. I'll take it I guess. They hadn't gotten to these extra obscure AEC ones yet though.
@dang: Can we change the link from Aeon to original website? It looks like Aeon re-published/re-shared without the permission, and are just enjoying the ride.
I agree that the original source is probably better and that permission and attribution would be the best thing for Aeon to do (particularly as a sizeable publication), however he did intentionally publish it as CC0.
I suppose things like this are why people hesitate to publish under CC0: it sounds great from an ideological perspective, it's much more difficult when you realize how little control it leaves you with.
Yup, agree that attribution might have been more ethical thing to do, even if not required legally.
As far as I understand the author wanted to release it as CC0 after the summer screenings i.e. few months. Due to which the video isn't really available for download on official website.
Anyways, it was a fun read about CC0 after reading your comment. :)
Isn't MLK's "I have a dream" speech copyrighted and aggressively enforced by his estate? Even today, some 60 years later?
I suppose if something can be commercialized, humans would do it, happily and forever. It doesn't matter how culturally significant it is (like the MLK speech). Money first, I suppose
Or rather, copyright is as long as Disney needs it to be. The length keeps getting pushed back, suspiciously close to when Mickey Mouse would enter public domain.
All this to say, don't expect anything to enter public domain anytime soon.
The last copyright extension was in 1998. With the rise of companies like Google, there’s enough lobbying power in the other direction that I suspect (well, hope) that a new extension wouldn’t pass.
Considering how angry the right wing party is at Disney right now (see their legal battles and backstabbing in Florida) maybe we'll finally see the extensions stop.
One of my very first blog posts (on blogger IIRR) was about me searching for some info on MLK and finding the second or third link on Google (this would have been 2004/5?) - and only after reading it twice in confusion did I work out it was hosted by stormfront.org and basically was ... well emphasising possible plagiarism in phd and possible affairs. You know a mud slinging operation. I thought google could cleverly use attributes in anchor links like "Disapprove".
Anyway the point I am badly making is that I am pretty sure the educational value of his speech on say wikipedia outweighs the benefit of dragging some idiot remixing it to say rude things through the courts.
We had a class the first year of high school that taught various skills every quarter. Things like giving presentations, properly using the scientific method, etc. The writing and researching papers component used that specific site as an example for making sure you dive into the source of "information" and making sure you have corroborating evidence. She would ask the students to check it out, and use the various techniques we knew already to evaluate it. Pretty much every student would fail and call it trust worthy and she would then reveal that it (at the time) was run by one of the largest group of white supremacists on the internet.
Not sure if the librarian who taught the class still uses that site, but as a 13 year old it was a very powerful lesson.
I really don't favor this, although I completely understand that they would want to prevent it being exploited in political advertising. Ironically, keeping King's legacy behind a paywall exacerbates the the very inequality he deplored and limits public access to an understanding of his ideas in the round. So instead we get smarmy politicians that exploit 1 or 2 particularly famous lines of his to shut down rather than open up discussion. By selectively quoting him, they inoculate themselves from criticism, suggesting that attacks on the politician's character or policies are actually attacks on the MLK lapel badge they are (metaphorically) wearing.
> …shut up and be the whore like the contract you signed says you are.
Setting aside your concerns about how poor, helpless politicians are being bullied by all-powerful artists, the fact is that licenses often have carve-outs for conventions and other campaign events. Artists can also prevent use by arguing that it infringes on their Right of Publicity or the Lanham Act.
Not a good thing. Also not working out well; MLK's kids aren't MLK. There's also no reason to believe that whoever they'd decide to sell the copyrights to has any sort of moral authority. They could sell it to Elon Musk or Unilever.
It seems that while it's in the public domain, it's not easily accessible, it really only exists digitized on Getty's servers. It sounds like when you download from Getty, their license restricts you from distributing the copy that you obtained from them (gleaned from the documentary linked in the OP, though I'm not sure how this is legal).
Amazing. I've ranted about the disgusting practices of Getty Images before. They claim ownership of public domain photos and demand payment of $1k/photo for access to them. Utter gutter scum.
Counterpoint: archiving costs money, even for public domain content. Who pays the archivist, if not the users of the archive? And the more obscure a work is, the fewer users of that work the cost can be spread over, so the oldest and most obscure materials would have the highest access cost.
It is tricky, because one could argue government could take on the role of archiver, but do we really trust government to maintain an unfiltered archive indefinitely? Or we may say archives should be privately funded and there should be redundancy, but then how do we guarantee continuity of funding?
Distributing public domain content is a solved problem. You can put it on a government website with a BitTorrent seed, whereupon it'll be mirrored to Wikimedia Commons, Internet Archive, and everywhere else. Archival & digitization is the actually expensive part.
I will note that "who pays" is a problem even before the content hits the public domain. What you're arguing was specifically one of the arguments used to justify the Sonny Bono Copyright Term Extension Act[0]. Unfortunately, copyright is the wrong tool for the job. It merely says that if you produce a work someone wants to buy, then they have to buy a copy of it from the person who made the work.
The problem is, the vast majority of old works are not in any kind of demand. Congress and the EU made sure that copyright would vastly outlive the actual commercial lifespan of the work. So marginal[1] increases in term length do not result in marginal increases in preservation. Because nobody is going to pay for it regardless. In fact, nobody is even going to be aware that the work in question even exists.
The way that archival works nowadays is that you preserve everything you can get your hands on. Because it's the common garbage that nobody cares about that is the most likely to be lost. Any market-based solution like copyright is likely to miss such works.
[0] Derisively called the Mickey Mouse Protection Act, due to Disney lobbying for it, even though it was primarily started by the EU wanting everyone to adopt Germany's copyright term lengths
[1] In the economics sense. e.g. "on the margin, does adding one more year of copyright result in one more work preserved"
You're conflating archiving and archive maintenance here. There are great arguments that can still be made for funding by charging for access the people who search out rare and wonderful things and create fine copies of those things for people to enjoy. Taking an old film from being found in a barn or an attic to a beautifully restored video is sacred, and if people stop doing it films burn and books rot.
Maintenance of a digital archive? Just put it out there with no legal threats or restrictions and let the right people download it once or twice. It'll last forever. Instead, everything is tied up in restricted access, legal knots or just the impression of legal knots. Eventually local hard drives will be as illegal as the RIAA wanted to make blank cassette tapes. All of your data will have to be in clouds managed by state-intimate monopolies to insure future legal compliance.
Datahoarders, academics, filmmakers and others would build communities around peer-to-peer film archives with no legal encumbrances.
IPFS kind of solves this I think. If content is interesting enough to archivers, they can pay to pin that data. If you make it open, then conceivably your costs could be reduced. I don’t know if IPFS supports it but you could image a fractional ownership model.
The amount of money to require you to record your entire waking moment in full high depth video is about 6 tib / year. The cost of storing that on the commercial cloud is relatively inexpensive. Less than a few hundred a month at most.
My point is that the cost is not about what people with money choose to archive, it’s that there’s a curation challenge that is the bigger financial obstacle. Storage costs might also start to skyrocket if everyone starts to record 24/7 video about their lives (already a problem for video cameras).
Are you certain of that? If we say every waking moment in full high-def means 16 hours per day in glorious 1080p, I'd back-of-napkin that to 17 petabytes.
Hmm. Was I off by an order of magnitude? I took 1gib per hour which feels like a reasonable level of quality. 8760 hours per year is 8TiB per year, no? For 17 PiB you’d be using 1TiB per hour which is very very off (uncompressed video?). Is my math off somehow?
1GB per hour of high-def video? Which encoder are you using, that sounds really good.
...Actually, now that I am looking up some actual data it does look like many formats come in lower than I expected for an hour, though I don't know whether those formats are:
A) Free to use, and
B) Full HD (1920x1080, what we in the EU call HD) video losslessly compressed.
My back-of-napkin mistake was looking at a random 1080 video, seeing it was 2.2GB for 24 minutes and extrapolating from there.
Let's see, Apple Animation (which I don't know whether it's free, but it can compress losslessly) comes in at 597.2 GB/h of 1080p video. I can't find numbers for VP9, but it is patent-encumbered. I can't find numbers for Lagarith, but it is GPL, although as a fork of HuffYUV it's optimized for encoding speed rather than storage.
Well, according to Wikipedia, lossless compression can achieve up to 12x compression, so let's call it 80GB/h (I'm back-of-napkining again, bad habit I know), so 640TB/y?
isn't that the purpose of the library of congress in the US? and wasn't the original intention of the US system of registering copyright, that a copy of the registered work was shared with the library?
the problem today is that there is to much content to collect it all, but i believe something like the MLK speech should have a copy in the library. apparently any US publisher is still required to send two copies of each publication to the library.
>... and someone did without his permission. Unfortunate.
Unfortunate, or the result of releasing something as a CC0[1]? Not saying he shouldn't have done that, nor am I particularly blaming him, but this is the very risk inherent with that type of license.
Now I just have to find funders for the next ~90.
[1] https://whatisnuclear.com/zppr.html
[2] https://whatisnuclear.com/news/2023-03-13-hallam-historical-...