penagwin

To be clear they said it “still works” and “ ok thinking more about this, there's no way this is the same vuln, this is another one.”

Meaning this specific bug the parent comment is talking about is not new in just these patched versions from today.

So delaying your updates today would mean you’re exposed to an actively exploited RCE and still exposed to the other memory corruption bug (which may or may not be an easily exploitable RCE) for an entire 2 or 3 days.

And from the sounds of it people are very actively exploiting this so waiting 24+ hours will practically ensure you’re hit.

I do agree that it’s generally matched my impressions of models for general SWE at least until we had Opus 5 and 5.6-Sol beating Fable 5. Which coincidentally is when they hit 74%.

Models like Deepseek 4.1 flash, Muse 1.3, and MiniMax 3.1 flash are certainly capable and very cost effective, but I’m not sure anyone would say they’re better than Fable 5/5.1 in capability

> [Space Bunny] was an early version of MiniMax M3.1 Flash.

They mention that space bunny was an earlier checkpoints, so it may have improved (or may it could be benchmaxxed)

I find it interesting that so many models seem to be at almost exactly 74% specifically which makes me suspicious of the benchmark. I want to say 5.6-sol, Astra, deepseek flash 4.1, opus 5 (and 5.5?), Gemini 3.8-flash and a few others all ended up there?

The FAQ on their website says Minimax 3.1 flash is 428B total, 23B active

Which is roughly similar to deepseek 4.1 flash.

I find it interesting that we’re calling these “flash” models now, even though they aren’t “small or light” and in many cases aren’t particularly that fast.

I suppose it’s more so that the standard for “regular models” is becoming 1T+

https://agent.minimax.io/tools/m3-1-flash-preview

Would there be any point in 2.5gbps beyond installing system updates a few seconds faster?

AFAIK you’d be hard pressed to hit 150mbps with your own local media.

And any streaming services will use way less - Netflix and Hulu recommend a mere 16Mbps at 4k for example.

I assume it’s largely a side effect from the final RL in post training?

That’s the step that causes the most significant gains in agentic performance.

But the RL doesn’t care about anything except maximizing the score, so if you only score based on coding benchmarks, anything can happen to the writing style (as long as it doesn’t hurt the coding performance).

That’s why it often gets worse on models that simply had more RL post training from the same base.

Reasoning models are the same speed. They’re just post trained with RL to do CoT inside tags like <thinking></thinking> before a tag like <response></response>

There’s no difference in the inference implementation, parameter count, or speed.

You can find a list of satellites here: https://www.ucsusa.org/resources/satellite-database

According to the image on Google maps it was taken by Maxar. Maxar appears to have a few satellites, looks like 5 in geostationary orbit (~35,700km), 3 at about 400km and 1 at about 500km.

Yes, as well as applying for jobs (or at minimum when hired), renting an apartment, and lots of financial things including any type of KYC crypto exchange or investment accounts. I've also had utility companies ask for it.

These are in no way secret, I have no idea how people are okay with this. You can easily social engineer so many critical services if you know somebody's SSN.

> PNAS October 26, 2021 118 (43) e2108728118; https://doi.org/10.1073/pnas.2108728118

I'm curious how was this published October 26, 2021? (Currently Oct 15)

Apparently 3G is being phased out by AT&T (and I think most other providers have already left 3G) in the US.

Which sucks because I can get a 3G module for my iot projects for 5$ and I think 4G modules start at 50$?

Are we just at the mercy of 4G now for services that need large coverage but low throughput? I know 5G technically has a spec for IOT type stuff but I haven't seen anything about it, and I haven't seen modules for it.

To me it's fair to say it's not a likely scenerio, it's just that they continously say "this would be difficult" at every step. I appreciate the breakdown, but it comes off as trying to convince you it was a near-complete impossibility. I understand it was unlikely, but it was possible, and that makes it severe either way.

We currently self host a version from ~2019, I haven't looked into upgrading the version but just the old version is working great.

I live in the US and have had comcast inject into http requests. I noticed because of a pop-up in csgo's menu (it loads html for their blog)

one instance of doom compiled with wasm on electron, but only if you lower the resolution to get it to run smooth

Yes! However it gets a bit complex, as they're using dns based geofencing so there's some extra steps.

He is talking about government level threats, DO provides no benefit.

I'll add that rolling your own means you're the only one exiting that IP address, so if your threat model involves websites profiling you and/or alternative accounts that won't help.

The user id though will burn you though. People who use something like BetterDiscord will be able to easily spot it (as well as your join date).

I'd recommend just using the browser to make an alt.

Yeah it's a tricky one isn't it? On one hand many of the best security researches are ex-state employees, and many of them go from that into the private sector. On the other hand it makes it sound like they are friendly with potential adversaries.

I don't think crypto would solve this particular issue. Stripe needs the ability to back out of moving money, so there's several settlement periods for different parts of the transaction and ways to appeal transfers retroactively.

I suspect fraudster's are able to wait out this period without detection so they can cash out. If this is the case, then even time locking smart contracts won't help, as the fraudsters just wait out the time period. At that point Stripe would have even less recourse to recover money, as retroactive transfers are not possible at that point.

I could see services such as their debit card offering being abusable too.

They also likely have to worry about things such as predatory recurring payments as those will result in chargebacks which could ultimately fall on Stripe to foot.

I'm not affiliated with this project but I'll plug https://palmdb.net

There's also the subreddit and a discord channel with a very active community of palm developers/collectors

From my understanding it depends on how naughty your current exit node has been. Most tor exits in my experience allow creation and you just need to verify another email (just use a random burner- they only block a few. Supposedly some well behaved exits require just a captcha but I've never seen it.

Because like 90% of the traffic is malicious. VPNs and proxies obviously change your exit IP, and people rotate through these as they do things like spam, make accounts, credential stuff, etc. This means they're burning through the IP ranges and getting them flagged.

If you're on a VPN with a fresh ASN and IP range you won't have any issues (until people start using it for other reasons).

If you wanted to "fix it" the dirty method, there's extensions for chrome/firefox/etc that will automatically submit your captcha to a captcha solving service and it costs some tiny amount.

tldr; Mischievous basically has to go through VPNs, and they rotate through IPs getting them flagged. You can join the dark side with a captcha solving service and extension, and there's some "solutions" like privacy pass you can try.

I've listened to several police radios with my SDR - "old male with [medical condition] at [person's address]" isn't uncommon.

Some research - paramedics are likely covered under HIPAA (if they work for a healthcare entity that provides ambulance services), I don't believe police or the fire department would be covered.

Same here!

> remote: fatal error in commit_refs

Myself and I'm sure many other's can't wait to try it once you have the self hosted version!

Something tells me you might have a difficult time convincing say Qualcomm with just that reasoning.

Ha facebook can't handle people monitoring them and collecting data eh?

Facebook just submitted this ad to me. I don't know why, they "trust" me. Dumb idiots.

Already beat you to it https://github.com/vlang/vinix

> V and it's creator get a lot of flack here on HN and the criticisms are probably warranted.

Hi fellow Penguin :D!

I'm impressed by the creator's dedication here too, but he's clearly biting off more then he can chew - he's been saying that he's making a

* Web framework * GUI framework * SQL ORM * A language that has a lot of promises - seriously checkout their list. * Github alternative * VIM alternative

This stuff is awesome as a project, but each one of these have teams of dedicated engineers sponsored by many companies to hammer out issues and make them as robust as possible. And while he certainly has working demo's of each, they're demos. There's a huge difference between a POC and a robust framework.

A half baked framework is useless sadly, I think he's spreading himself too thin while have a roadmap that looks like a programmer's christmas list.