JUNK DRAWER GALORE

Ada Wong

Robbed from Leon saying this in resident evil requiem,,

Rookiebug AU

[ danteleon ] managed to beat up art block by drawing summertime gay people🏖️

Noodleleon

Noodleleon AU

When Grace arrives to confront Zeno and discover the truth about Elpis, the last thing she thought she was gonna see was a DRAGON sleeping in front of the door. She goes through the whole gambit of emotions in the span of a few seconds after seeing it. Fear (what is that?) wonder (OMG It's a dragon!) sadness (Oh no the dragon is sick...) confusion (..wait, why is the dragon wearing Leon's pants and shirt-?) and finally Shock (...HOLY SHIT LEON IS A DRAGON!!!)

That's how I'd do it but given that her 'I do what I want' evasion of copyright might not extend to making everything accessible to the entire online population and iirc Stratt being kind of controlling and particular about who she works with, I'm not sure that's how she'd do it. Unfortunately since she made everyone speak English as the lingua franca (authorial bias showing imo) she might just put everything in English to shrink the amount of data. That would simplify the project but you lose a lot in translation, so that's better if her goal was simply to aid the mission crew vs preserve Earth's cultural heritage.

Actually I have 7 minutes before I have to sign on. If our priority is just mission effectiveness rather than cultural preservation, the strategy I'd probably take off the top of my head is:

Qualitative interviews with the crew and relevant subject matter experts. Go through that data to identify themes/topics most likely to be relevant to the mission.

Have LIS or related experts transform those themes into a controlled vocabulary.

Develop an automated tool that can ingest your content, generate a brief abstract, and assign index terms from the controlled vocabulary.

Have human expert review of the tool's output. (This will probably need to be spotchecking if you've got a small team of experts; again we are talking about all of human cultural output here.)

This is basically the other side of the scale from the first suggestion - it would give you a faster, more targeted search that would help you zero in on mission-critical content. (After all you probably don't need your search tool running through the scripts for 50 years of General Hospital when you're trying to find emergency treatments for a crewmember experiencing arrhythmia.) The tradeoff of course is that if you end up needing something you weren't expecting (like, say, linguistic theory because you just met an alien) it might not be as effective.

(None of this is touching on how this much data is being STORED of course. I have to imagine at least the audiovisual content is compressed and only the index is being actively searched, and content is unzipped if it needs to be accessed. Which again points to needing a really good indexing system. But I'm not a computer scientist.)

Would it make sense then to have a two-tiered system:

One for scholarly work and relevant nonfiction overseen by a team of specifically selected specialists focused on mission-critical topics, probably first prioritizing medicine, engineering, biology, genetics, astronomy, chemistry, physics, computers, and psychology (though I do think translation and linguistics would go in a high priority tier—if the paper they really need is in French or Japanese or isiZulu they don’t want to leave their astronauts SOL) and then as time allows branching out into food and nutrition science, ecosystem and environmental science, geology, mathematics, history, archaeology (look. you don’t KNOW what you’ll find out there), with academic work on topics like economics and music and literature probably being at the end of this priority scale or put into…

The folksonomy/citizen science initiative specifically for fiction, popular media, and entertainment topics, on a separate indexing track so if you’re looking for actual medical information it won’t turn up House M.D. but if you want to see what nonsense the House writers came up with it’ll still have been fully tagged and searchable by dedicated show fans. If Stratt expects to go to prison anyway and the world is dying, I can easily imagine her concluding that either no one is going to be paying money for media in a few years as global economies collapse, or certain softwares will become globally critical and the owners of it don’t retain the right to profit off the end of the world, so sure, release the archive of pop media to everyone and give them a crack at it.

They can be going in tandem, with the categorizational whims of the world not affecting the organization of mission-critical data but also the specialist team not having to spend their time also going through 16,000 episodes of General Hospital.

Folksonomies focusing on fiction/cultural work would make sense, especially for webnovels/webcomics/youtube videos and other stuff that doesn't already have a MARC record. The tough one for me is the open web since it's such a mishmash of format and genre. Some professor's janky web 1.0 HTML webpage might have valuable information but it's not going to be indexed in Web of Science. Could you automated a webcrawler to identify academic/nonfiction vs fiction/entertainment webpages? (Beyond just registering .edu domains.) Probably, with human QC followup, and then the sites get sorted into the appropriate bucket.