Until someone suggests to me otherwise, I'm going to keep saying the following:
By default, third-party password managers are stupid. Before - two people had the secret, now three of them do, one of whom is a juicy central target that has the secrets.
I'm aware that this is an extreme oversimplification of e.g. hashes and whatnot, but it seems true enough for laypeople such that it's better advice than most, especially for individual and personal accounts.
You don't have to manually sync, the phone app has a Dropbox/SyncThing/whatever integration so everything's synced automatically. On desktop, it's a file that you sync via Dropbox/SyncThing/whatever. Harder than a third party app, but if you're comfortable syncing files, then you can very comfortably use an offline password manager (I have for 10 years on my phone and the UX is basically the exact same as using my work appointed password manager).
I used to use a similar approach, in my case: KeepassXC + Google Drive. My problem though is that I got tired of the issues KeepassXC had when saving the database into the mounted Google Drive filesystem provided by Gnome (GVFS, from the top of my head). Basically, sometimes it failed, other times the database would be saved with a different name which was annoying, because then, when using Keepass on my phone or my other computer, I had to be searching for the most recent database file. After months of internal debate and research, I ended up migrating everything into Bitwarden. Despite the paranoid in me has days of "what the hell did you do!", everything has been great so far and it has simplified my workflow nicely.
Yup, and ALSO makes it easy to SCREW UP. That's the point. This paternalistic garbage helps no one.
Right now, the smarter thing is to be like: "Look, passwords are hard. Security is hard, but individuals have to take responsibility" because entirely too many companies are bad at this.
There's also another simple solution, but no company is yet good enough to do this. Provide indemnity.
Practical security is the balance of security against convenience.
Being ignorant of convenience required and capability available of the target audience isn't going to help anyone.
Don't let perfect be the enemy of good.
A 3rd party password wallet from one of the majors is vastly more secure than someone's nan using the same password for everything. Which is the alternative in practice.
You seem to think that normal people are going to learn IT security to a reasonable degree, and be personally responsible.
Many people literally aren't capable of being what you want them to be.
Yours is a beautiful dream that clashes strongly with observable reality.
Observable reality today, right now, is trash cybersecurity everywhere.
If people don't learn -- and maybe they won't, then, the only real solution is actual liability. These companies promise safety and don't deliver; time to sue or regulate or both.
But we need to stop pretending that "third-party," in all its present crappiness, is an acceptable option, even if sometimes it's better than what people do on their own.
You can use an offline password manager (keepass) that's saved to a file and put that file in Dropbox. There's a keepass-compatible app with a Dropbox/etc integrations so your private password store is synced between all your phones.
I have a system known only to me for assigning passwords based on things like the color of the company's logo and the number of vowels in their name. I have a complex and unique password for every site, I don't have to "remember" them because I can recreate them on the fly. This works really well imo, the main hiccup is companies that force you to change your password every x days and don't let you reuse old passwords.
This technique sounds great but has some challenges.
Firstly, companies change logos, change names, force password resets or your password will leak in a breach. How does this system accommodate those scenarios?
Also, once a few of your passwords are in data breaches, anyone who wants to target you can derive your admittedly less complex formula.
It is not out of the question that there already is or will be technology to compare multiple passwords from the same person from different leaks and derive the formula.
> once a few of your passwords are in data breaches, anyone who wants to target you can derive your admittedly less complex formula.
It's true that's a hypothetical attack vector.
In practice, so few people use this approach that you're not going to see a general-purpose deployable attack suite based on this principle. It'd be useful for a targeted attack, but if you're under a persistent, targeted threat there are better tools (spear-phishing, social engineering, physical compromise, etc.).
Or one is cracked and the others are easy to figure out. Because these types of algorithms tend to feel complicated, but are not as you still have to remember it.
You can keep a physical password book. For someone to get it, they need to physically steal it. If you keep it at home in a fairly safe location, it seems fine to me.
If someone breaks into my home, they’ve got it. But that’s not a threat model that scales, so it’s not a major concern to me.
>You can keep a physical password book. [...] If you keep it at home in a fairly safe location,
But people are often away from home and want to log into websites and apps from their smartphone. A physical password book sitting at home is useless for that very common scenario. That "login from anywhere" is the typical motivation of gp's comment: "How can I keep passwords on 2 computers, an iPhone and iPad all in sync?"
The amount of times that lastpass is hacked staggering (this article is about the breach of 2022), but I don’t understand why people still use it (except maybe because of some enterprise contracts)
It would be cool if there was a way to corrupt a vault if it leaks using the master password.
I guess the only way that would work is if it was somehow connected to a large concensus network and could only be unlocked if connected requiring some sort of proof that can't be forged which in itself may be an issue.
Oh god I think I may have just given someone another idea for a crypto startup...
I don't see how that would work, even with a blockchain. No matter what you do, there is nothing stopping you from simulating the network yourself using a backup of the old state of the chain.
How can the attackers eventually exchange crypto for fiat currency without revealing their identity? It seems that now their wallet address is known and all transactions will be closely monitored therefore it would seem that the whenever they try to buy real goods it might be possible to identify them. Maybe sell the crypto to criminals for cash? But than now the chain is tainted and the buyer will have the same problem.
The current meta is to use a mixer and then gradually move the funds through successively less sketchy exchanges.
E.g. coinbase don't let you deposit from mixers (at least in large amounts), but they do let you deposit from okx, which in turn accepts deposits from sketchy chinese exchanges which accept deposits from mixers.
Alternatively, you can go via monero but that only replaces the first few steps in the ladder.
Thankfully I don't think any proposed regulation does much to stop this, short of entirely banning crypto, which I don't see happening any time soon.
You can buy legitimate, verified, fully-KYC-ed accounts for major crypto exchanges and banks on the darknet for dirt cheap. Send crypto to exchange, sell or convert to a privacy coin, or transfer fiat to bank. Use virtual bank card to buy gift cards or whatever else.
Or use something like changelly or a number of decentralized exchanges.
You have more complicated ways to laundry crypto by having legit wallets profit from dirty crypto wallet ‘mistakes’. Think of doing swaps that are MEV profitable and you run it by your own node for example with a MEV bot
Sometimes it’s not about hiding an identity but delaying the tracing and identification via any means necessary (moving the crypto through mixers, less traceable coins, shady exchanges, unsuspecting third parties, etc.) until the whole value is extracted through various methods and fake identities.
The identity behind a wallet or account might be worthless. Years ago a common scam in Eastern Europe (maybe everywhere) was to pay old or homeless people a tiny fee to open bank accounts all over the city and then hand over the credentials to the criminals. Then they could operate a network of thousands of such accounts for various purposes, sometimes for reasons as simple as getting a small loan that doesn’t require any collateral and disappearing with the money.
This can work as long as they can move around the crypto in an obscure enough way so that it takes just hours or days longer for the authorities to trace than it takes for the criminals to cash out.
Do it in a country with much laxer KYC laws for a haircut and then use one of the myriad other methods of transferring large amounts of illegitimate wealth across borders that existed prior to cryptocurrency and which continue to be widely used. Frankly at this point tof you're transacting in cryptocurrency most banks and governments are going to assume that you are either an idiot, a conman or a criminal and treat you as such so the entire chain is already "tainted".
For better or worse, this is literally the use case and design for projects like Tornado Cash. I won't say "easily," but they'll probably be able to get away with it.
There's limited concern if you had a good master password right? I've changed everything important long ago but still plenty of small stuff I wouldn't love leaking.
A strong password will help a lot, but the password iteration value may reduce protection for some people.
One of the most unfortunate things with the LastPass breach last year was that lots of accounts had the number of PBKDF2 iterations set to a low value (typically 1000). A low number of iterations makes brute forcing the master password easier.
This value would have been set when the user's account was created, and for older accounts it wouldn't have been increased unless users changed the setting manually.
LastPass is now setting it a lot higher by default, but that doesn't help the vaults leaked in the breach.
Agreed, I certainly don't still use them, But also I haven't changed ALL account info... I reseted on my master password being secure but I suppose as quantum computers ramp up I'm going to need to change everything as those vaults won't disappear... What a colossal fuck up by LP.
My password was made with the best practices, but I see people mention the amount of iterations done. Does that basically mean just Iterations * each attempt to brute force? ie 2 rotations to figure out a 1 digit pin = (1/10)2= 20 max guesses? Thanks a lot for reference though re 50 bits and 3.5. Helpful numbers to keep in mind.
(Edit: looked more in to how to calc password entropy. Luckily I'm good. Thanks a lot for this info tho.)
As much as I like LastPass as an idea, I never trusted it enough to start using it and I'm frustrated my paranoia seems to have been justified.
I use an alogrithm for swizzling a handful of passwords that fit in my brain. It's the best compromise I've found between using too few passwords and letting them ride around in a third party that could get compromised.
