comfy

I want to build a small site which acts as a broad, searchable FAQ for a certain topic.

Consider I have the FAQ:

What is the approximate mass of Earth?

It's 5.9722 × 10^24 kilograms, wow!

I want the user to have a chance at finding this FAQ by asking How heavy is our planet

Looking at this basically, the two similar questions have only one shared word, "is", which is an extremely common word. So using something really simple like word comparison or even stemming/lemmatization alone won't help.

On the very other end of the spectrum, a search engine's AI feature can interpret this effectively, rephrase the question and give a similar answer. So, what strategies are are in-between these two extremes?

1. A few people will be adding questions to the site regularly.

2. If possible, no external services, just self-hosting on an affordable server.

3. Simpler and lighter solutions are preferred.

Are any of the features in OpenSearch (ElasticSearch/Lucene fork) able to do this? Is it overkill?

Since the site will have new questions to match regularly, will a solution require the repeated, wasteful retraining of NLP models to to create weights? Or is training so efficient for small-scale text datasets that it's responsible and reasonable to do on a cheap low-end server?

edit: Just spitballing here, I could try a solution which does the bulk work at insert-time rather than runtime, by asking a general pre-trained language model to rephrase the question many different ways, or generate keywords, then use those responses to generate tags for a basic keyword search to match. This would avoid making a heavy search function or retraining any model on the server.

GPT-4o mini

Here’s a list of synonyms for the keywords in "What is the approximate mass of Earth?" formatted as an array of strings:

json

[
  "weight",
  "heaviness",
  "bulk",
  "load",
  "volume",
  "estimated",
  "rough",
  "approximal",
  "near",
  "close to",
  "planet Earth",
  "the globe",
  "the world",
  "Terra",
  "our planet"
]

Finding a wholesome community archive with tagging done well is a rare treat, so I think this place is worth a special mention.

The linked page has clips posted in a 2015 thread, along with links to the full détourned Aiura episodes most of the clips are from.

Fifteen are from Aiura, three are from The Devil Is a Part-Timer!, and one is from Eden of the East.

The how-to guide (although, being a decade later, there are probably now improved ways to do this)

Bonus: Inspired by those, a nukechan user made three Shrek 2 détournement clips:

• https://lefty.pictures/post/view/23419
• https://lefty.pictures/post/view/23420
• https://lefty.pictures/post/view/23421

PSA is a public service announcement, an awareness campaign.

It could be as simple as teaching everyone to walk on the same side of the footpath in each direction, to demonstrating how quickly a fire spreads and ways to prevent and react.

Alright, I'll save you from typing: https://en.wikipedia.org/wiki/Israel_Epstein

I'm asking this out of curiosity; I don't need to host anything that can't already be done in the West

Lots of countries have very relaxed or non-existent enforcement of torrent filesharing. That's not what I'm asking.

I'm asking about what place one could openly host every known commercial pop song and every Hollywood film without any worry about being shutdown or sued.

For a reference, according to a one-minute check of Wikipedia, the only countries which haven't ratified the Berne Convention or the TRIPS Agreement in any way are Eritrea, Kosovo, Palau and Palestine. While joining these agreements doesn't imply they're enforced, it gives an idea of how widely governments do agree to intellectual property rights.

~~how much would it cost to launch an independent server into orbit?~~

I'm going to make a compilation, so specifically mention the work which answered the question. Bonus points for a direct link or quotation of a paragraph or two which directly refute the argument.

I'm going to have access to a 3D printer for a few days. I know two friends who've used them, but it's only been for art and figurines, or professional purposes.

Are there any other cases you can think of where a custom-printed item is better than the myriad of mass-produced plastic items?

Zootopia is a 2016 animated crime film centered around themes of corruption, pride, and systemic prejudice. It follows Officer Judy Hopps, whose idealistic worldview rationalizes their reckless endangerment, abuse and torture of Zootopia's citizens as Hopps struggles to uncover a conspiracy. Despite career success, Hopps's spirit progressively degrades as they begin to realize the reality of their violence, power, police and the state.

• Means TV: https://means.tv/programs/pln
• YewTube (invidious): https://yewtu.be/channel/UC2jXeGiElzyMmW8NJt2ajfg
• YouTube 🐷🎩: https://www.youtube.com/channel/UC2jXeGiElzyMmW8NJt2ajfg

A monthly video summarizing some bloomcore news from around the world. It's refreshing to see.

Join Mexie for your monthly dose of hope as we celebrate our wins and let them strengthen our resolve to fight.

Together, we will win!

Send positive leftist news stories to veganvanguardpodcast@gmail.com

What is something you can sense that few-if-any people you know can sense? Literal answers only.