If you’re still interested in ChromeOS (but not shit), you might like this
https://www.zdnet.com/article/fydeos-chromeos-clone-revive-aging-pc/
Can confirm: run .apk’s natively when I tried (ran Newpipe and Redreader on it, as well as some native linux apps).
I think you’ve answered a different question to the one I was making. I’m not claiming chatbots are a sound business. I’m saying the pricing path follows the Uber/Airbnb playbook: subsidise hard to build the user base, then tighten the consumer tier once the real money is elsewhere. That’s the “as intended” bit.
Uber’s also a shaky example for “virtually no OpEx”. It lost money for the better part of a decade, running on investor-funded rides and driver incentives, before it squeezed either side. That’s the pattern I was pointing at: generous now, worse later.
Whether the underlying business holds up is a separate argument, and you’ve got a fair point there (Air Canada is a good example of the liability risk). It just isn’t the argument I was making.
Just to add a soupcon of detail: both Codex (OpenAI) and Claude Code (Anthropic) have had significant price increases this year. Eg:
The goal (clearly) has always been B2B sales… with consumer tiers slowly being tightened or transitioned. At the moment, OpenAI is stupidly, unsustainably generous with their $20/month tier (which includes separate pools for chat and Codex). As we’ve seen with Anthropic, that will slowly winnow down to the point of non-utility.
IOW, the plan is proceeding as “intended”. See: Airbnb, Uber, etc.
Thanks.
It’s Lemmy. Everything is apparently AI slop now.
Again: coded by meatbag ™; documented and poked by clanker.
Fork it call it and call it Chazzwazzers? 🤣
Kidding aside, it was designed to run on a low memory device, specifically the raspberry pi. It displays maps. So…
The power of ASD compelled me.
Boom motherfukka! I stayed up all night but POI population is DONE! I now have 1,809, dynamically generated, ZIM Wikipedia-linked POIs for Western Australia running entirely locally on my Raspberry Pi map server.
The POIs themselves come from the WA OSM dataset. This works because a lot of OSM features already carry wikipedia=* and wikidata=* tags…which gives me a link from a mapped place to its corresponding Wikipedia article.
I can then resolve those article titles against my locally hosted Kiwix Wikipedia ZIM and only publish POIs that actually exist in that local copy. I used wikipedia_en_all_mini_2026-06/Wikipedia ZIM (~13GB) as the test.
Meaning, clicking a marker now opens the article lead inside the map, fetched from Kiwix, with a “Read full article” link to the full locally hosted article (see pic)
And just to confirm - this is all local. No calls out to Wikipedia or Wikidata.
Seeing folks want this in their hot little hands, I’ll delay fucking around adding / solving those 4 outstanding ticket items and yeet this onto Github (after I get a few hours of sleep).
Quite a lot, actually.
Coding, document analysis, STT, home assistant, shopping assistant, gaming, journalling, image and video generation, OCR, language translation, recipe/meal / workout planning, study/flashcard generation, email drafting, adversarial review, search engine on steroids, hardware troubleshooter, companion for elder care, music curator and DJ …
All of that without creepy ass cloud shit from Big AI.
I can go on, but “a lot” probably covers it.
EDIT: asked, answered and…down voted. Classic Lemmy anti-ai knee jerk. FWIW I work with AI in healthcare settings as well as code review for my own personal projects.
What I said are actual use cases, not a wishlist generated by Jippity. I can elaborate on any and all of them with actual real life experience.
This any good?
https://www.azuracast.com/