We need an internet archive outside the US! We can't put all our eggs in one basket.

Oh, wait: we *have* an internet archive outside the US. Let's support it.

And now they have an office - a physical presence!

https://flaminghydra.com/freedom-and-sharing-at-the-internet-archive-europe/

https://www.internetarchive.eu/

@mttaggart Do the tend to positively discriminate Italians?

Inky Impression 用に額縁を設計してプリントして収めた。安っぽいけどだいぶフォトフレーム感が出た #shapolab

Workslop: bad ‘study’, but an excellent word

wasting coworkers’ time with AI — as directed

https://www.youtube.com/watch?v=xnI-Yr6plXI&list=UU9rJrMVgcXTfa8xuMnbhAEA - video
https://pivottoai.libsyn.com/20250923-workslop-bad-study-but-an-excellent-word - podcast

time 7 min 10 sec

https://pivot-to-ai.com/2025/09/23/workslop-bad-study-but-an-excellent-word/ - text

So I just reported a very interesting bug in Outlook - took me a while to analyze and understand (part of) the root cause..

The bug allows to read the value of an MAPI property* and use it *directly* as a memory pointer. We know OOB (e.g. caused by integer overflow) bugs are not rare but they're all some sort of "offsets", not someone just read the value and directly use it as a pointer.. So basically you can set the DWORD 0xdddddddd in the email and Outlook would happily access memory at 0xdddddddd. Not sure what severity it could be as it's an OOB-read at first glance, but def. an interesting find for me.:)

Full details will be released after MSRC's review.

* https://learn.microsoft.com/en-us/office/client-developer/outlook/mapi/mapi-property-overview

#VulnerabilityResearch #Outlook #EmailSecurity

System/38 project update. Edith is off the pallet and on her own wheels. WE REPEAT THE EDITH HAS LANDED.

https://www.gofundme.com/f/ibm-system-38-computer-rescue-and-restoration

@robdaemon The AS/400 was a direct drop-in replacement for the S/38, and IBM didn’t want used 38s on the market to torpedo AS/400 sales, so they were *extremely* aggressive in trade-ins and take-backs to keep them off the market.

In our research we’ve found only 10 or so S/38s still in existence world-wide and several of them only exist as piles of parts. We only know of 6 intact, of those, only 2 power up, and NONE IPL all the way to an operating system. CPF as the historical predecessor to OS/400 is pretty much lost.

On IBMi the core OS task is still called SCPF, 48 years after the S/38. :-)

The crates.​io team was notified of two malicious crates (with similar names as legitimate crates) which were actively searching file contents for Etherum private keys, Solana private keys, and arbitrary byte arrays for exfiltration. The malicious crates have been removed.

See the blog post for details: https://blog.rust-lang.org/2025/09/24/crates.io-malicious-crates-fasterlog-and-asyncprintln/

I wrote an article about what I wish juniors in cybersecurity would ask for and contribute when asking for a mentor and career guidance, in light of the terrible tech jobs market.

https://tisiphone.net/2025/09/24/reasonable-expectations-for-cybersecurity-mentees/

#Cybersecurity #Infosec #MentoringMonday #Mentoring #CybersecurityCareers

How a 20 year old bug in GTA San Andreas surfaced in Windows 11 24H2

https://cookieplmonster.github.io/2025/04/23/gta-san-andreas-win11-24h2-bug/

[RSS] Why is Windows still tinkering with critical sections?

https://devblogs.microsoft.com/oldnewthing/20250924-00/?p=111624

@cR0w @talosvulns I don't think so, thanks!

stty

https://wizardzines.com/comics/stty/

(from The Secret Rules of the Terminal, out now! https://wizardzines.com/zines/terminal/)

You just don't understand the fourth industrial revolution

@cR0w @talosvulns OK but why aren't the ones at Talos listed at Nvidia? (I just did a quick Ctrl+F, I may have missed something)

@cR0w Interesting... @talosvulns just released a bunch of Nvidia writeups but CVE's don't seem to match up

@cR0w Also one by Rapid7 on @attackerkb: https://attackerkb.com/topics/LbA9ANjcdz/cve-2025-10035/rapid7-analysis

@linux Are you an LLM?

"CVE-2025-1727 reveals a critical design flaw: the EoT/HoT linking protocol — which sends emergency brake commands [to trains] over a radio channel — has no cryptographic authentication."

https://cervello.security/blog/vulnerabilities-incidents/research-cve-2025-1727/

🚨 SolarWinds, the gift that keeps on giving: a new Web Help Desk patch bypass, CVE-2025-26399, enables unauthenticated RCE via deserialization.

It’s a patch bypass of CVE-2024-28988/CVE-2024-28986 - previously exploited.

Given SolarWinds’ past, in-the-wild exploitation is highly likely. Patch now.

Need help assessing your exposure? https://watchtowr.com/