Project Nomad — ネットが落ちても知識は手放さない。オフラインファーストの知恵袋

Project Nomadは、インターネット接続に依存せず、いつでもどこでもアクセス可能なナレッジベースの構築を目指すプロジェクトです。「知識をオフラインにしない」ことをコンセプトに、たとえネットワークから切断された環境下であっても、重要な情報へのアクセスを永続的に保証する革新的な取り組みを紹介しています。

I like the idea of an LLM that acts as a public knowledge base. But that doomsday framing on the site is pretty annoying.

So this thing is based on Kiwix, which is based on the ZIM file format.

In the meanwhile, wikipedia ships wikidata, which uses RDF dumps (and probably 8x less compressed than it should be).

https://www.wikidata.org/wiki/Wikidata:Database_download (https://www.wikidata.org/wiki/Wikidata:Database_download)

There is room for a third option leveraging commercial columnar database research.

https://adsharma.github.io/duckdb-wikidata-compression/ (https://adsharma.github.io/duckdb-wikidata-compression/)

In a world where this is useful, you aren't going to be spending your precious battery on running an LLM...

I'm a fan of "civilization in a box" kinds of projects. However the ZIM file format leaves a lot to be desired in 2026. I've been exploring a refreshed, alternative approach: https://github.com/stazelabs/oza (https://github.com/stazelabs/oza)

I do think having an LLM as an optional "sidecar" is a useful approach. If you can run a meaningful Ollama instance alongside your content, great!

For anyone wanting the video explanation from the creator, watch: https://youtu.be/P_wt-2P-WBk (https://youtu.be/P_wt-2P-WBk)

There's a company which sells something like this, as "Prepper Disk".[1]

In the 1950s, US Civil Defense had a set of microfilms on how to rebuild society. These were packaged with a sunlight reader and stored in larger fallout shelters. Someone should find one of those.

[1] https://www.prepperdisk.com/ (https://www.prepperdisk.com/)

Normally I cringe at doomsday preppers but given how many dictators out there love the idea to cut their country off Internet whenever anything starts going not in their favor, I imagine a lot of people may find this useful.

I wouldn’t want to lose access to knowledge how to fix a sink or which medication is better, just because the local kingface currently feels that free exchange of opinions about him threatens his kingship.

I come from a time when internet connectivity was not permanent.
It was only available a few times per day when you connected via the phone line. My first ISP gave me an allowance of 20 hours of internet per month.
You would dial-up, check the news, check your email, read a page or two, download what you had to download, and then disconnect.
The internet was very slow by today's standards, and the connection would get lost very often.
It was during that time when it was drilled into my head that the network access comes and goes.
That it should not be taken for granted.
So a lot of the stuff that I use nowadays, I also have in an offline format.
I keep offline docs either in pdf or in html format of most of the programming languages and frameworks that I use.
I keep the source code of various projects that are essential to me.
I keep a local wiki with notes on various things that are useful to me.
Obviously it's not enough for a major catastrophe but it's better than nothing.
I'm by no means a prepper, but I also believe that each of us should be prepared for short term disruptions of various kinds. The network should not be taken for granted.

Found a click or two in looking for storage and other system requirements:

  What About Raspberry Pi?
  Project NOMAD is designed for more capable hardware to support local AI. 
  If you're looking for a Raspberry Pi-based solution, check out Internet 
  in a Box — it's a great lightweight option for basic offline content.
  Project NOMAD is for when you want the full experience: GPU-accelerated 
  AI, comprehensive content libraries, and a professional management 
  interface.

Sounds like I should look at one of the other options mentioned there and in this thread, assuming their libraries and maps are basically the same. I'd like the “comprehensive content libraries” while travelling or when otherwise away from a reliable connection, perhaps with a useful management interface for easy updates and such when on good connectivity, but just in a format I can click or grep through. While I'm assuming I could just turn off, or otherwise ignore, the LLM side, just not having it in the first place would be more efficient.