Hacker News

Article URL: https://renewablesnow.com/news/reflect-orbital-gets-fcc-nod-for-inaugural-solar-reflector-1298071/

Comments URL: https://news.ycombinator.com/item?id=48919954

Points: 1

# Comments: 0

Article URL: https://www.youtube.com/watch?v=Ng_Bj7tVw78

Comments URL: https://news.ycombinator.com/item?id=48919945

Points: 1

# Comments: 0

Article URL: https://lore.kernel.org/linux-media/CAHk-=wi4zC+Ze8e+p3tMv8TtG_80KzsZ1syL9anBtmEh5Z40vg@mail.gmail.com/

Comments URL: https://news.ycombinator.com/item?id=48919931

Points: 1

# Comments: 0

Article URL: https://github.com/emergingrobotics/gorai

Comments URL: https://news.ycombinator.com/item?id=48919913

Points: 1

# Comments: 0

Article URL: https://www.ft.com/content/c5a880f1-aa6a-4336-b499-3ab95cf37857

Comments URL: https://news.ycombinator.com/item?id=48919907

Points: 2

# Comments: 0

Article URL: https://github.com/emergingrobotics/gowyoming

Comments URL: https://news.ycombinator.com/item?id=48919894

Points: 1

# Comments: 0

Article URL: https://github.com/datopian/portaljs

Comments URL: https://news.ycombinator.com/item?id=48919882

Points: 3

# Comments: 0

Article URL: https://www.schweizerbart.de/papers/zdgg/detail/prepub/108503/Geological_and_geophysical_characterisation_of_the_exploration_boreholes_EB1_and_EB2_Inde_Syncline_Rhenohercynian_Fold_and_Thrust_Belt_and_Weisweiler_Horst_Lower_Rhine_Embayment_Germany

Comments URL: https://news.ycombinator.com/item?id=48919874

Points: 1

# Comments: 0

Article URL: https://briarproject.org/news/2026-maintenance-mode/

Comments URL: https://news.ycombinator.com/item?id=48919869

Points: 2

# Comments: 0

Article URL: https://www.cnbc.com/2026/07/13/spacex-spcx-stock-ipo-price.html

Comments URL: https://news.ycombinator.com/item?id=48919861

Points: 2

# Comments: 0

Article URL: https://metro.co.uk/2026/07/15/twitter-x-a-history-tweets-mark-20-years-platform-29127604/

Comments URL: https://news.ycombinator.com/item?id=48919851

Points: 1

# Comments: 0

Article URL: https://github.com/MultiArrayQueue/WaitFreeMultiArrayQueue

Comments URL: https://news.ycombinator.com/item?id=48919845

Points: 1

# Comments: 0

Article URL: https://uptimemonitoring.com/

Comments URL: https://news.ycombinator.com/item?id=48919840

Points: 2

# Comments: 0

Article URL: https://andywidjaja.com/blog/110-pipeline

Comments URL: https://news.ycombinator.com/item?id=48919832

Points: 1

# Comments: 0

Article URL: https://sublimeinternet.substack.com/p/how-to-create-a-distinct-visual-identity

Comments URL: https://news.ycombinator.com/item?id=48919826

Points: 1

# Comments: 0

Article URL: https://github.com/google/GNM

Comments URL: https://news.ycombinator.com/item?id=48919820

Points: 1

# Comments: 0

Article URL: https://torrentfreak.com/eus-top-court-geo-blocking-protects-publishers-in-copyright-disputes-vpns-not-liable/

Comments URL: https://news.ycombinator.com/item?id=48919810

Points: 1

# Comments: 0

I spent weeks reading about how Cognee, Graphiti, and Neo4j's `agent-memory` build their agent memory architectures. They converged on the same heavy knowledge-graph design: an ontology, LLM extraction pipelines, deduplication, the works.

I really wanted to use them for my personal use case, but that looks like such a heavy setup that adds a lot of friction and silos. Plus, it feels like I just get my data trapped in their service, for not a ton of value.

That's why my "long-term memory" still lives in Obsidian, Readwise, and Google Drive, with per-project LLM wikis as the agent's memory. No infrastructure. And I'm fine with it.

They ship memory as a product, which, in my opinion, at a personal or small scale, is overkill. You can build the same "knowledge graph" experience via plain old `.md` files within an LLM wiki memory.

But still, graphs are strong, so I adapted the same architecture from the Cognee, Graphiti, and Neo4j `agent-memory` stacks to build a data-mining tool with just MongoDB, VoyageAI, and Gemini Flash. But I scoped it to a very particular problem and ontology domain to avoid the KG noise.

On the other end of the spectrum, if you want to ship a product at medium-to-large scale, it makes sense to start using monsters such as Neo4j, Zep, or HydraDB.

But I am curious: what is your long-term memory setup? Obsidian + LLM wikis vs. Cognee/Graphiti/Zep? Do you actually use tools such as Cognee or Zep?

In case you are curious about how Cognee, Graphiti, and Neo4j's `agent-memory` work under the hood, I wrote a full breakdown here: https://www.decodingai.com/p/unified-memory-from-scratch-knowledge-graphs

Comments URL: https://news.ycombinator.com/item?id=48919162

Points: 1

# Comments: 0

hi HN,

5 months ago i posted ( https://news.ycombinator.com/item?id=46959456 ) an early stage of a video editor in the browser, not relying on a cloud backend. I wanted a solution for myself to not only edit videos, but bring my full adobe, ableton, blender workflows inside one browser application. Important for me is that almost everything is running in realtime, leveraging the GPU of the users system efficiently. Although i have to say that every change is hard when you used to work a certain way for 16 years, there are more and more situations where i am able to create what i want wit this solution, faster than in the old ways.

so i didn't stop building on it, vibing you say, but i prefer the term "kindergarten teacher for highly gifted toddlers"

since feb i implemented much more stuff, and a few things make me really happy and confuse me at the same time. 3d capabilities with cameras and lights. Fbx, objs, even splat sequences on the timeline with a, for me, usable preview system. i started with three js as base, but quickly realised that a custom approach is much more versatile.

i also reused the whole UI to add a full blown audio editor with many effects and even midi piano roll. so to switch to a sounddesign workflow or even creating musik is just one click away. there is also a complete node-based layer which presents an ai agent of choice all meta data of the media, so the agent can build a fast programmatic solution for the user needs directly inside the app. like an effect with keyframes, but created on the fly.

i also implemented a complete workflow for handling AI generated content like videos, images, speech, and musik. i just dont want to leave this one solution anymore in order to create my videos, no matter if Ai gen is part of it or not.

you can also import adobe premiere projects, but this is not that well tested, worked one time on one project.

features get added quicker than i can make them stable, but i spent well over 1000 hours doing this, and its the best time of my life. Everything is free open source MIT (except the ai generation stuff obviously)

Iam developing on win 11 (needed adobe). chrome is my main target, but firefox and safarie on mac are less tested (firefox needs native helper for file system acces). Linux is more tested

https://www.masterselects.com https://github.com/Sportinger/MasterSelects

Comments URL: https://news.ycombinator.com/item?id=48919147

Points: 1

# Comments: 0

Article URL: https://www.nytimes.com/2026/07/14/science/reverse-sprinkler-physics-problem.html

Comments URL: https://news.ycombinator.com/item?id=48919113

Points: 1

# Comments: 0