[gwern]

> I searched for services which offered to digitize Video8 tapes. Most services cost about $20 per tape. Even with discounts for bulk amounts, it would likely have cost about $2k! I considered paying it (how exactly do you value a few hundred hours of childhood video?) but then I noticed how they delivered the videos - a private media hosting solution for 60 days. I knew this would be a huge amount of data, and only giving me and my siblings two months wasn’t sufficient.

I'm not following here. Even if it was several terabytes of video (digitized at high resolution and minimal lossiness for archival purposes), that's plenty of time to download. Especially if you're a developer who can casually spin up a cloud or dedicated server to proxy through if need be? (And $2k sounds reasonable once you start going through "hundreds of hours" at a bare minimum, and again especially if you're a developer with real opportunity cost.)

Also, as far as the video analysis goes, Gemini might've been a better idea?

[dsr_]

You're looking at the technical capability, not the human behavior.

A random non-techie human without particular urgency will not download a hundred old family videos in 60 days. They will watch some of the first one, try to fast forward, stop when it gets boring, and think about downloading them sometime in the future and then forget about it until six months later.

(Except your uncle who is the family historian, who tries to download all of them but runs out of room on day three.)

[SamPatt]

Yup, you nailed it. This was my exact concern.

[ekropotin]

As I techie human I’d probably do the same thing, thanks to ADHD

[SamPatt]

Hi gwern!

My siblings are very much not developers. That's a lot of data for them to download, store, and figure out a way to view.

I was worried they'd just see a list of filenames and not put in the effort. By creating a streaming experience, I thought they'd actually watch them.

You might be correct that Gemini could have helped, I didn't test it, but much of the knowledge of who was in a scene, where it was, and why it would matter is inside my head. I doubt any model could effectively label locations and people over 20 years of video.

As to the opportunity cost - I'm currently looking for work, so mine is undoubtedly lower than yours!

[gwern]

> My siblings are very much not developers. That's a lot of data for them to download, store, and figure out a way to view.

I wasn't suggesting anything about your siblings, but you, who are a developer. I was just talking about the actual download step, not what you did after that. (Obviously you were going to host them somewhere else in some other form. Probably not DVDs but a little quickie website or maybe just a Flash drive with a HTML file index, say, I don't know, lots of options here to make it user-friendly for your siblings on Christmas Day. The hard drive or Flash drive idea has the benefit of LOCKSS, especially if you use up the spare space providing PAR2 FEC.)

> I doubt any model could effectively label locations and people over 20 years of video.

Actually, Gemini is highly promptable with a large context window and a single still image only takes up ~300 tokens IIRC, so I think that you could probably do so! Just include, say, 3 photos of each person over time with a natural language description, and 1 photo of each location, and that might be enough to get back useful labels. Gemini can even do bounding boxes. (Google is quite proud of its vision and video analysis capabilities.) And you can run multiple passes or split up videos etc.

[SamPatt]

Ah I understand you now. Yes I could have had a service do the digitizing then only done delivery myself. And given the time investment that probably would have been more sound. I don't think I'd do it all myself if I did it again.

I didn't know Gemini models were that capable. I admit I'm still skeptical about this approach though - even if it were capable of accurately labeling people and locations across decades, there's no way it could know when a scene is of personal interest. I kept a running log for each sibling as I was manually doing the labeling, knowing what they'd want to see, which presumably is only possible for me and my siblings to do with any accuracy.

If AI could ever do that then we've definitely hit ASI!

[gwern]

> I kept a running log for each sibling as I was manually doing the labeling, knowing what they'd want to see, which presumably is only possible for me and my siblings to do with any accuracy.

But you could feed that back in! Just write it down. It's all tokens. As you read over descriptions and note down key pieces of family history or per-sibling details, that provides information about better annotating the next video for possible points of interest. And you can chat with the LLM and write down more general principles. It's not like a LLM like Gemini doesn't know an enormous amount about family life and things of sentimental value, and can't make good initial guesses. And when you do this, you still haven't used up more than a small fraction of the context window with these image references and text profiles and principles...

[mrguyorama]

It's also just really really weird.

Every single digitization service I have ever known is a small business of one guy who sends you back DVDs full of files. It's not even a consideration by anyone that they could possibly hold your videos hostage like that.

[SamPatt]

It wasn't about having the videos held hostage, but about the ease of accessing and viewing them. I didn't want to do DVDs at all - frankly, that's just a bad experience to go through 200+ hours of unlabeled video.

[IncreasePosts]

You don't even need to spin up anything - eg AWS datasync will mirror the remote server for you directly into glacier which can be extremely cheap if you're okay with access-a-file-12-hours-after-you-request-it

[gwern]

I'm not familiar with 'AWS datasync', but isn't the point of Glacier that it's really expensive and slow if you retrieve everything? It sounds like a bad idea for videos he planned to download in order to catalogue them. (He might want to delete them or could safely infer that 'no, neither me nor my siblings will want to watch this on Christmas Day', yes, but he still has to look at them first since he doesn't know what's on them at all.)