startling

It's a matter of degree. Do the positives of publishing outweigh the negatives? I think so.

Why does Ward need to do that before putting a tarball up somewhere?

I've seen so many sites cease to exist after undertaking ambitious overhauls like this.

So, where's the content? How many backups are there? How many backups are offsite? Are any publicly-available?

The people you're afraid of ("if anything, this list provides a potential target list for other hackers to try and compromise those stores even further") already have access to the data gwillem used as a source.

Yep. Someday.

I'd love to go to Antarctica.

Maybe!

But for example Piwik's main competitor is Google Analytics. It's reasonable to not trust Google for privacy reasons, but Google employs one the largest security teams in the world and produces very well-regarded research. I trust Google with securing Google Analytics more than I trust Piwik's developers with securing Piwik.

I was interested in Piwik until I noticed the many, many open security issues:

https://github.com/piwik/piwik/issues?q=is%3Aopen+is%3Aissue...

Includes an issue from January about insecure random-number-generation: https://github.com/piwik/piwik/issues/9473

and one from March claiming that it doesn't reliably use TLS to update plugins: https://github.com/piwik/piwik/issues/7582

It bothers me a lot when people advocate self-hosted services with really bad security posture to increase privacy (or "control", as the author says).

Sometimes you can do this with things like smtp servers as well.

It was fixed, but not (publicly) recognized as a security issue, so likely not backported + updated by distros in "stable" channels.

I wrote a bit about it: https://blog.lizzie.io/notes-about-cve-2016-7117.html

tl;dr looks like it'd function well a local privilege exploit. not an expert, but as a remote exploit think it would need 1) a service running recvmmsg (which is kind of unlikely) 2) a way to cause recvmsg to error, and 3) a way to cause the service to close its socket in the middle of the recvmmsg call.

Password managers don't have a single point of failure: an attacker needs both your password and the state of the password manager.

I couldn't find anything, so I wrote some https://blog.lizzie.io/notes-about-cve-2016-7117.html

> It is believed that the hack compromised personal data from the accounts including names, email addresses, telephone numbers, dates of birth, hashed passwords (the majority with bcrypt) and, in some cases, encrypted or unencrypted security questions and answers.

https://en.wikipedia.org/wiki/Yahoo!_data_breach

i.e. messages weren't stolen, account information was.

The primes are hardcoded into the source the same way that the primes are hardcoded into the sequence of natural numbers. They're there, but the important bit is finding them.

That's not similar, at all. This is actually removing multiples, as in the sieve of erastosthenes.

And the BIOS?

AIUI, no: https://www.youtube.com/watch?v=JABJlvrZWbY

This is a talk from 2012, so things may have changed.

Once you reprogram the monitor, you can store or exfiltrate all of the data the user sees, and do clever things like erase and redraw the mouse pointer, or draw new prompts, to induce the user to click on the things they wouldn't have otherwise.

I do mean display monitor, not temperature monitor.

Since the monitor is external, I wonder if they've considered monitors as attack surface: https://github.com/RedBalloonShenanigans/MonitorDarkly

Does that work? Can't the evil maid install a malicious hypervisor to dump interesting pieces of memory every few minutes?

if someone has physical access to your computer with secure documents present, and then you use it again, it's game over.

1) dl.sta.li is not on my site, and probably not on yours.

2) That's still suboptimal. You're giving each host on your LAN the ability to interfere with any newly-provisioned server. This is a perfect and hard-to-detect way for an attacker to pivot from "RCE on some random webapp" to "advanced persistent threat".

The stali source includes, among other things:

* a full checkout of libressl: http://git.sta.li/src/tree/lib/libressl

* a full checkout of expat: http://git.sta.li/src/tree/lib/expat

* two full /bins: http://git.sta.li/rootfs-x86_64/tree/bin http://git.sta.li/rootfs-pi/tree/bin

> Installation uses HTTP to fetch a boot volume image.

This is a problem with all of the suckless software and it, well, sucks.

No, I mean that even if you add additional layers of sandboxing, the chrome process can still access the X socket. A process with the ability to read and write to the X socket can record keypresses, issue mouse and key events, etc etc.

(There's not a well-supported way to avoid this, but there are some tools that use Xpra for it.)

My landlord already had it installed when I moved in, but yes, I can talk to the person at the door and press 9 to buzz them in.

All of that is only marginally helpful without X sandboxing.

Do you install firebug?