croemer

7 ms·

91.
by croemer 4mo ago
Fellow T1D here. The same pump tslim X2 also failed on me 3.5y in, also on holiday (I was in Morocco). I had backup pens with me so no big issue. I've now switched pump to my Ypso mylife with CamAPS FX algorithm. It's so much bett
92.
by croemer 5mo ago
TLDR: Products affected: "All dedicated servers and cloud plans at all locations are affected by this adjustment. However, web hosting products, managed servers, servers from the Server Auction, IPs, storage products, Load Balancers, V
93.
by croemer 5mo ago
More than 15 min after it started an incident is finally posted: https://www.githubstatus.com/incidents/xy1tt3hs572m
94.
Tell HN: GitHub is partially down again
12 points by croemer 5mo ago|5 comments
95.
by croemer 5mo ago
Your website bio and LinkedIn don't match at all. Is the LinkedIn link on your website wrong? Update: yes it is. This is the correct one: https://www.linkedin.com/in/logan-nye You're doing too much vibe codin
96.
by croemer 5mo ago
We rewrote Nextclade in Rust and are very happy. Works nicely both for CLI and client side browser with wasm. https://github.com/nextstrain/nextclade
97.
by croemer 5mo ago
Those are all the tests for alignment. They don't even check the alignment is correct. Just that there are no errors. This is a joke: https://github.com/logannye/rosalind/blob/main/tests/alignme
98.
by croemer 5mo ago
> If it is open-source, you will see a meaningful activity - stars, forks, contributors. That's not true, I'm quite sure most repos on GitHub have neither many stars, nor forks, nor multiple contributors.
99.
by croemer 5mo ago
Issuance was stopped almost 2 hours ago: May 8, 2026 18:37 UTC.
100.
by croemer 5mo ago
Reads at least partially like LLM writing, for example: > When code production gets cheap, the cost doesn't disappear. It migrates. > It was true then. It is unavoidably true now.
101.
by croemer 5mo ago
Staccato sentences, em dashes, word choice, the headings, "X is not Y. It's Z.": > The legacy of the book has almost nothing to do with the satire. It has to do with the language. Or this also reads very LLM like to me: &g
102.
by croemer 5mo ago
You can suggest changes on the talk page. > Wikipedia provides a structured process for subjects to suggest changes without directly editing articles. The mechanism centers on the article's "talk page", a discussion area a
103.
by croemer 5mo ago
Someone should try to replace the neurons with urand and see if the chip can still play Doom, in the spirit of the qday prize winner.
104.
by croemer 5mo ago
At this point, it looks just like an experiment. It's not a definitive "were going to switch". I think people here are reading too much into it.
105.
by croemer 5mo ago
Presumably stock market valuation is grounding? Also, eBay shareholders can vote down the acquisition if they don't think the deal is good for them.
106.
by croemer 5mo ago
The stock part is more like a merger than a buyout.
107.
by croemer 5mo ago
One step at a time? Quick back of the envelope: Assuming 30k body hair follicles, at 5s/follicle that's 42 hours. Now if you just want to do your arms this is like down to a few hours. Not so bad?
108.
by croemer 5mo ago
What matters is where the headquarters and key employees are. SE can be based anywhere in EU and even then legal entity stuff isn't what matters. Origin: German, HQ: German, Accounting regime: German, Main stock listing: German, Execut
109.
by croemer 5mo ago
That's not what the Pull Request intends, though. There's a difference between "This PR is buggy and enables attribution on everything" vs "VS Code enables attribution on everything". The title is fuzzy on inte
110.
by croemer 5mo ago
This should be a "Show HN:" given it's author submitted and quite promotion heavy. AFAICT, this just combined two chips on a board. And the 100x bandwidth is due to using a higher frequency chip. Nothing revolutionary.
111.
by croemer 5mo ago
The "regardless of usage" in the HN title isn't correct. If you dig into the source you see that the attribution line is only added when some changes in the commit come from Copilot, either inline completion or agent. @dang
112.
by croemer 5mo ago
Discussed 2 days ago (28 points, 14 comments) https://news.ycombinator.com/item?id=47962775
113.
by croemer 5mo ago
Doesn't read like LLM writing to me and my detector usually fires very quickly. Don't see any obvious signs - want to share what made you think of LLM writing?
114.
Agentic coding is burning me out(0xsid.com)
2 points by croemer 5mo ago|1 comments
115.
by croemer 5mo ago
None of the links to Github work because you're pointing at the main branch instead of your default branch ghosts-only
116.
by croemer 6mo ago
> For Wikipedia itself: The “reliable sources” policy needs to grapple with a new world where LLM assisted vandalism can produce plausible press releases at the click of a button. Citation only to a single source registered within an edi
117.
by croemer 6mo ago
Since you're here, to calibrate my LLM detector, it'd be great if you could say whether or not you used an LLM to write parts of the article.
118.
by croemer 6mo ago
I think it's even slow for high schoolers. I didn't practice that much and ran 100m in 12.5s from rest at my peak. 4s slower is snail pace. I think most in my class could run that fast (or slow).
119.
by croemer 6mo ago
No, it's slower than most people's sprints. It's 17 seconds per 100 metres which is slow. Most teenagers can do this starting from rest.
120.
by croemer 6mo ago
Good explanation of the flatpak sandbox escape. For those allergic to LLM writing: Some sentences read very LLM-like, e.g.: > The fix wasn’t “change one function” — it was “audit the entire call chain from portal request to bubblewrap ex

More ›