lima

5 ms·

1.
by lima 8d ago
They may still train on it if you submit feedback or flag a safeguard. The terms are a bit fuzzy on this.
2.
by lima 11d ago
Inside the JS sandbox, not the browser's outer containment sandbox.
3.
by lima 11d ago
That blocks all sorts of abuse, but a targeted exploit would be delivered on a clean domain.
4.
by lima 11d ago
It would be chained with a separate 0day sandbox escape.
5.
by lima 11d ago
uBlock Origin won't help with this kind of targeted exploit, and Firefox has a much worse security track record.
6.
by lima 12d ago
Which browser has a better security track record?
7.
by lima 13d ago
eID PKIs have very little in common with the web PKI. There's a national root of trust with strong attestation. It's a very simple trust relationship. You already trust the respective government to issue IDs. Plenty of European co
8.
by lima 1mo ago
Yes, if you're unable to be polite about it, just don't comment and/or email the moderators. Your comment would have been informative and well-received without the snark. You were right to call out the other person for trivia
9.
by lima 1mo ago
TIL about Ticktogs. Thanks!
10.
by lima 1mo ago
Be careful with the dismantling, if it's a used test, you will likely end up with DNA amplicon contamination which is extremely hard to get rid of.
11.
by lima 1mo ago
HN has very specific rules for this: https://news.ycombinator.com/newsguidelines.html You don't respond to them at all, flag the comment, and/or email the moderators. Being snarky does not contribute to the discus
12.
by lima 1mo ago
Lucira did that, but IMO it's quite wasteful. LAMP reactions don't need much, so it's best to make a cheap reader like Metrix or Pluslife so the tests can be less expensive.
13.
by lima 1mo ago
Antibiotics aren't free of risks to the patient either, including doxycycline. There's always a risk tradeoff.
14.
by lima 1mo ago
There's also the more fundamental issue that Postgres - unlike something like Yugabyte - does not use a distributed consensus algorithm for writes and can lose committed writes during network partitions. But of course, neither does Tai
15.
by lima 1mo ago
And tick-borne encephalitis, in regions where that is endemic. No need to be rude, though!
16.
by lima 1mo ago
Ticks climb low vegetation (not just grass) and wait for a host to pass by ("questing"). If you're going to disagree with basic textbook biology knowledge, the burden of proof is on you.
17.
by lima 1mo ago
CDC recommends prophylaxis only if the tick is engorged. In my country, guidelines do not recommend prophylaxis at all (for better or worse - personally, I would take a single doxycyclin dose for a >24h high-risk tick bite, Lyme is no jo
18.
by lima 1mo ago
Ticks do not drop from trees.
19.
by lima 1mo ago
> But you would know whether you needed to even book an appointment. Only if the test is sufficiently accurate to rule out transmission. Even molecular tests can't reliably predict that.
20.
by lima 1mo ago
Yes, but this test cannot diagnose Lyme in humans.
21.
by lima 1mo ago
Unfortunately, there are major issues with this approach. The vendor claims "lab-level accuracy" (conveniently omitting the actual accuracy numbers), but the test is a lateral flow test, which means its Limit of Detection is going
22.
by lima 2mo ago
Bandwidth is really cheap when you run your own infra
23.
by lima 2mo ago
It's a competitive market. Kimi K3 has almost reached parity with Opus.
24.
by lima 2mo ago
It always worked like that
25.
by lima 2mo ago
Even if you can't authenticate at the application level, it's still much better to encrypt/authenticate traffic on the wire (using a VPN or something like Tailscale) instead of 802.1x auth.
26.
by lima 2mo ago
Non commercial licenses are not generally considered open source
27.
by lima 3mo ago
Google solved credential phishing a long time ago using hardware tokens (gnubby - the predecessor to FIDO/U2F/Passkeys...). There's other types of social engineering, but phishing is mostly an engineering issue.
28.
by lima 3mo ago
Probably 802.1x, but it's easy to bypass if you have access to an authorized device. This kind of authentication has to be done at the application level, treating the network as a perimeter doesn't work.
29.
by lima 3mo ago
The company also should have restricted network access to the port in the conference room so that an unknown device like a Raspberry Pi could not make an Ethernet connection from that spot Bad take - the actual problem is that there was a
30.
by lima 3mo ago
This. It's easy to forget that Postgres is fundamentally a single-node database without distributed transactions. It won't pass the Jepsen test suite with multiple nodes. DBOS, Temporal and friends inherit this limitation. Somethi

More ›