For the life of me, I don&#x27;t understand how DigitalOcean can&#x27;t afford a $50 a month sponsorship to GNOME and Flatpak but can afford $3M to Omarchy. [0]Not only that, DigitalOcean cancelled their $50 a month sponsorship to GNOME and Flatpak.To put this in perspective, it would take 5,000 years for GNOME and Flatpak to reach the $3M figure DigitalOcean has given to Omarchy.How does this make any sense?[0] https:&#x2F;&#x2F;social.treehouse.systems&#x2F;@barthalion&#x2F;117241534134006167

Ask HN: How can DigitalOcean not afford $50, but can afford $3M to Omarchy?

I use claude for most of what I do, and Fable is largely overkill for me and frequently burns through my Max plan&#x27;s session credits in minutes (!) when just doing an initial mobile app planning with 4 agents. After I waited out the timeout period 6 hours later,  and I picked up again, the cache had timed out so it burned through 2% of the session in less than a minute. Opus 4.8 is now my goto and I will be avoiding 5 until I see a reason to switch back. 4.8 is &#x27;good enough&#x27; for what I need and has been a great value. It mostly gets things right. Most of what I do is web and mobile , largely cloud backend.

Ask HN: What default model do you use and why?

I was recently bit by reddit&#x27;s decision to block usage of old reddit when signed out, which was my normal mode of reading.Now, new this morning, I get this error when trying to sign into Reddit on Firefox w&#x2F; uBlock Origin:&gt; An error occurred. Please disable any extensions or try using a different web browser to continue.Is this real life?

Reddit Now Blocking Firefox for Android with uBlock Origin?

Is anyone else in a panic over the Gemini 2.5 model generation (Pro, Flash) being sunset in October before there&#x27;s even any Pro class model in general availability (with geo restrictions etc.)? Google wants everyone to migrate to 3.x Flash, which beats the older Pro models on the benchmarked tasks, but isn&#x27;t the same thing as the Pro class on reasoning-heavy tasks like complex reasoning on very large documents (my big use case).The Gemini family had a distinct niche in document comprehension, with thousand page input documents taking only 300k tokens. Nothing quite like that in OpenAI or Anthropic world, even at more than 10x the token adjusted price. Should we just give up on Google at this point and engineer around the competitors&#x27; limits and eat the costs? Totally unnecessary own goal by team Google.

Ask HN: Did Google kill its enterprise workhorse model?

I have the automatic &quot;co-authored by&quot; Claude signature turned off for my commits via the config setting.PSA: They&#x27;ve introduced a new &quot;Claude-Session: https:&#x2F;&#x2F;claude.ai&#x2F;code&#x2F;session_xyz&quot; line that&#x27;s appended to the commit now though, on by default.Annoying, feels scummy (of all things, ironic I know)

Tell HN: Claude Code appends new commit signature even if older settings is off

For those of you with enterprise pay-as-you-go plans, what&#x27;s your company doing to keep costs from exploding? Do you have monthly limits? How do those limits spread across vendors? Does everyone have the same limit? What happens if you hit that limit halfway through the month?My company is a medium size startup trying to support multiple vendors and harnesses. We&#x27;re thinking of having tiers of limits (tech vs non tech) and allowing users to decide where they want to spend that.Curious to hear what others are doing as prices and usage get higher and higher.

Ask HN: For Enterprise coding agents, what's your company doing to control cost?

Over past couple months I noticed that HN feed is almost exclusively AI or AI-adjacent news. Meanwhile the legitimately, broadly-hacker stuff gets left out for the most part.I noticed that because the things I find genuinely interesting that I post here now get zero traction, which is the stuff that I believe would previously be met with some discussion.Note that I don&#x27;t care that my submissions were ignored, I don&#x27;t need that validation. Problem is that if my stuff gains no traction, I am myself not seeing similar stuff posted by others, stuff I am genuinely also interested in and would want to hear about from this community.Here are two most recent examples:- Lenovo showed a model that incorporates solid-state air cooling. That technology has been around for a while but it looks like it matured and is production-ready. This allows for super-light and slim designs, which, in turn, should allow for bigger batteries installed and I think it&#x27;s some sort of breakthrough: 
https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=49580229- Brax Industries devices, genuinely interesting, open-soruce and innovative, especially the wall display, with its modularity and on-device processing compatible with Home Assistant. I also appreciate the very clean design: 
https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=49643215As said, I honestly believe this would gain some traction here on HN a year or two ago. I no longer find HN a good source of &quot;what&#x27;s new&quot; in the industry, it&#x27;s almost exclusively AI talk now.So I think the point I am trying to make is that maybe YC could start curating the content somehow to give those easily-drowned-out articles a visibility bump?Or, at the very least, introduce simple tagging, so we can start filtering stuff out.

Ask HN: Can we please limit the AI news flood?

Hello HN ,I built a knowledge graph engine for codebases- Records why each change was made — written back by agents as they work, so it gets better the longer you use it
- Dashboard stays in sync with commits and updates from GitHub and GitLab
- Flags stale files&#x2F;code and deranks them internally, so the graph stays current
- Token savings in the range of 78 to 99.74%.
- Connects to claude code, codex , opencode etc
I will be happy to talk about Graph , sych updates ,persistent memory and also how it detects stale files and token savings ?Visit :
https:&#x2F;&#x2F;github.com&#x2F;Novaya-AI&#x2F;novgraph
https:&#x2F;&#x2F;trynovaya.com&#x2F;

Novgraph: Persistent Knowledge Graph for Codebases

I&#x27;ve reset this more than once and the last time I made a careful note of when I did it and to my surprise I found it re-enabled when I checked just now. Make sure you check this thing to see if it hasn&#x27;t been re-enabled if you believe it to be off right now.

Tell HN: OpenAI keeps re-enabling the 'allow training' setting

Namecheap is having hosting issues again despite their promise to make things right after the incident a month ago.

Namecheap Hosting Is Down

Hard req: has to have cell connectivity &amp; support roaming (US &lt;&gt; EU). Not for daily use, but for comms when they&#x27;re out of the house in case they need to reach out. I looked at apple watch family and it&#x27;s great, except that it doesn&#x27;t allow roaming or plans abroad (it seems that BetterRoaming might work but it&#x27;s unclear). The Garmin Bounce 2 looks like it fits the bill, but it has ZERO reviews and some comments make it look like it&#x27;s crap.The Light Phone III seems like an awesome idea, but at ~$900 it&#x27;s a strong pass.The Punkt MP02 is so far my top contender.Any suggestions?

Ask HN: What's a good "semi-dumb" phone/watch for kids?

GoatCode is a single 85 MB binary that runs in your terminal. It brings 180+ LLM providers (or your existing Claude&#x2F;ChatGPT&#x2F;Gemini&#x2F;Copilot subscriptions via OAuth), and when your quota dies mid-session it automatically retries then walks your fallback chain live — switches provider mid-turn, says so, and finishes the answer. No crash, no lost context.What’s new in v2.1.10:
- Prompt caching on Anthropic (up to 90% cheaper long sessions; savings shown in &#x2F;cost)
- Inline unified diffs after every write&#x2F;edit
- &#x2F;rewind — jump transcript + files back to any turn
- Parallel subagents (up to 3 task calls fan out concurrently)
- &#x2F;search + goat search — full-text across every saved session
- @file Tab completion, &#x2F;model sonnet aliases
- True context meter (real API tokens vs model window)
- GOAT MODE (bypass) that actually works: mid-turn switch, prompt resolves YES
- Noir theme: deep slate + black-green surfacesInstall: npm install -g goatcode-cli (or grab a static binary from the releases page)Repo: https:&#x2F;&#x2F;github.com&#x2F;Arhan-w&#x2F;GoatCode
Landing: https:&#x2F;&#x2F;goatcode.vercel.app

GoatCode – open-source terminal AI agent with provider failover

I am curious what actually happens between two generations of AI models.For example, how do you go from Sonnet to Opus? Is Opus trained from scratch, built on Sonnet, or mostly the same model with more compute and training?And how do models like Astra suddenly make a big jump in some capabilities? What is stopping Anthropic, Mistral, or others from doing the same thing?
Is the main difference just more compute and money, or are there training methods, data, architecture, and research breakthroughs that competitors may not know about?I can&#x27;t think of a better place to ask this. I am guessing there are people here who actually work on these models and know what goes on behind the scenes.

Ask HN: How are much smarter AI models made?

As of Sept, 11th: https:&#x2F;&#x2F;openai.com&#x2F;index&#x2F;navier-stokes-solution&#x2F;Compare to https:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20260908235305&#x2F;https:&#x2F;&#x2F;openai.com&#x2F;index&#x2F;navier-stokes-solution&#x2F;Also, see GitHub - git history: https:&#x2F;&#x2F;github.com&#x2F;openai&#x2F;NavierStokesAndEuler

OpenAI changed Navier-Stokes press release and Lean4 code on GH

Per https:&#x2F;&#x2F;support.google.com&#x2F;mail&#x2F;answer&#x2F;17101213, starting in January 2027, you may no longer send email as you@x.com if you use a you@gmail.com account. You can only send as you@gmail.com.For those who like the convenience of having all of their external email commingling with their Gmail email, and taking advantage of the integration, such as calendar integration, Google Drive integration, and more, what is our best path forward?At this point, it seems like doing a Google Workspace paid account may be the best solution for continuity of experience, but you break all the commingling and integration. Also, your you@x.com account is a separate window from your you@gmail.com account.You could also consider a service like Fastmail. While it is reasonable quality, you don’t get the quality of the app or of the web experience, and its spam catching features don’t seem to be as good. Your access to you@x.com will be via Fastmail&#x27;s own interface or some IMAP solution. The Gmail mobile app appears to support IMAP, but it&#x27;s just an IMAP tool at that point, not strongly differentiated from any other IMAP tool. The Gmail web client has no support for IMAP.

Google to discontinue allowing send as an external account in Jan. 2027

I just replied to somebody on here with these links, thought I should share them. If any one has something to add, dont be shiesty:) It was a pain to find these.https:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20241003101005&#x2F;https:&#x2F;&#x2F;www.angelfire.com&#x2F;pa3&#x2F;VAMPMAN&#x2F;krypt_dcinfo.htmlhttps:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20031205061600&#x2F;http:&#x2F;&#x2F;dreamcast.planetweb.com&#x2F;community&#x2F;homepages&#x2F;dcdesign.htmlhttps:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20181029115530&#x2F;https:&#x2F;&#x2F;dreamcastarena.com&#x2F;how-to-create-dreamcast-website-friendlyhttps:&#x2F;&#x2F;www.dreamcast-talk.com&#x2F;forum&#x2F;viewtopic.php?t=16378http:&#x2F;&#x2F;kamui.cool&#x2F;zz_lab&#x2F;infinitedream_HTTP_USER_AGENT.htmlhttp:&#x2F;&#x2F;kamui.cool&#x2F;zz_lab&#x2F;jeffma_useragent.htmlhttp:&#x2F;&#x2F;kamui.cool&#x2F;zz_lab&#x2F;browsermatrix.htmlhttps:&#x2F;&#x2F;dreamcast.wiki&#x2F;Developmenthttps:&#x2F;&#x2F;dcemulation.org&#x2F;phpBB&#x2F;viewtopic.php?t=47902https:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20250713015510&#x2F;http:&#x2F;&#x2F;pc.psilocybindreams.com&#x2F;https:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20000521015744&#x2F;http:&#x2F;&#x2F;tools.planetweb.com&#x2F;wmhelp.html

Dreamcast Web Development Links

The software industry is at a standstill while everyone is busy buying Claude and chatgpt licenses to reduce operations cost.The news is full of AI or AI - adjacent articles to the extent it&#x27;s now a tedious topic.Where are the exciting innovations happening in the world? What great products are being built we&#x27;re not hearing about?

Ask HN: Where are the innovations happening today?

Comments on social media (when discounting the ironic posts) are mixed whether Astra is worth the cost increase, similar to the Fable-Opus transition.For the novel projects I&#x27;m working on, Astra did unblock me where I was stuck with GPT 5.6 Sol, which in itself gives it value for myself, but outside of that it&#x27;s not a significant improvement if at all. I have noticed Astra is more efficient so for my use case it may not be 2.5x cost.

Ask HN: Is GPT-6 Astra worth the 2.5x cost increase over GPT-5.6 Sol?

Since AI is getting smarter by the day,would it not be more profitable for both man and machine for  it to come up with more,harder Millenium problems?

Ask HN: AI and Math

Hi HN,
I built Stardust Store, an independent Android app marketplace as part of my ZISHU TRON ecosystem.
Website:
https:&#x2F;&#x2F;stardustapkstore.zishuai.cloud
The main idea is to focus on app discovery, rather than simply providing a large list of APKs.
The current platform has:
Editor&#x27;s Choice
Trending and New Releases
Popular Apps and Games
Search by name, developer and category
Top Free &#x2F; Paid &#x2F; Grossing &#x2F; Trending &#x2F; Rated rankings
App categories
User profiles
Wishlist
Download history
Reviews
I built the project independently and I&#x27;m currently focusing on improving the marketplace based on feedback from users and developers.
The Android version is also available here:
Download APK: https:&#x2F;&#x2F;github.com&#x2F;zishutron&#x2F;Stardust-store&#x2F;releases&#x2F;download&#x2F;v12.0.1&#x2F;starduststore.apkPrivacy: https:&#x2F;&#x2F;stardustapkstore.zishuai.cloud&#x2F;privacyTerms: https:&#x2F;&#x2F;stardustapkstore.zishuai.cloud&#x2F;termsI&#x27;d especially appreciate feedback on the marketplace UX, discovery model, rankings, and what developers would expect from an alternative Android app marketplace.

Stardust Store – An independent Android app marketplace

What is the average number of hops in your social network to someone living who knew someone who was killed in the 9-11 attack? I am particularly interested in how the answer would vary depending on the age of the first person in the social network graph?

Ask HN: What is the average number of hops in your social network to 9-11 death?

There are thousands of pairs of source-executable pairs that could be used for training, and the compiler is a parameter that AI can easily identify and account for as a heavily weighted factor.So why isn&#x27;t there (or maybe there is and I don&#x27;t know about it) an AI based decompiler that can take almost any machine code, and transform it into highly readable source code (complete with meaningful guesses for variable names and comments)? It seems like it should be even easier when the executable could be run and inspected at runtime by the AI model.What is the state of the art with AI decompilation, and are there tools I can play with today?

Why isn't decompilation a solved problem in the AI era?

I&#x27;m planning to release an open source software library soon and was wondering - are there any licenses that would prevent LLMs from ingesting it into their training? Do any of the labs say that they don&#x27;t ingest software code that uses certain licenses?I realize this would be something that would be nearly impossible to discover and they could always say that they don&#x27;t but they end up ingesting anyway.