HNHacker News
TopNewBestAskShowJobs

arnarbi

1,550 karma · joined April 30, 2012

Icelandic. SWE @ Google. Security, authentication, authorization. Work on WebAuthn, FIDO & passkeys.

[ my public key: https://keybase.io/arnar; my proof: https://keybase.io/arnar/sigs/lf4Zbpd04PZPXu_eZyGhPVoef2BuJ0p5VUuzV8ZXD4I ]

submissionscomments
arnarbi··on Macaroons Escalated Quickly
In one of our iterations, we did indeed have a mandatory rule that each set of caveats added by a holder was terminated by a “sign off” caveat. But in the end we concluded that if that mattered in the actual enforcement, then one wasn’t really doing things “the capability way” and then that principal probably needed some stronger authentication than merely having held the unattenuated macaroon.

But that’s theoretical thinking and the real world is indeed more shades of gray.

See you in Madrid!

arnarbi··on Macaroons Escalated Quickly
Macaroon is just the English word for the French word macaron.

ducks for cover

arnarbi··on Macaroons Escalated Quickly
> None of these changes are necessarily authenticated - I don't know that it specifically was the user agent that shared the token with the service or added the read-only restriction, at least with the core tech.

Macaroons are very much grounded in capability thinking, where it is a “feature” that principals such as the user agent in your case don’t need to have representable identities. So there’s nobody to authenticate, except the fact that it was a holder of the previous macaroons tag, which means they had the authority it represented.

In other words, not needing something like a PKI is somewhat the point of it all.

If you have the need to authenticate the intermediate principals, eg because they have pre-existing identities already, then the capability model in general may just be a distraction. They could just sign their desired attentuations with their own identity key.

arnarbi··on Biscuit authorization
Not really, no, if you refer to them being used for chaining. In FIDO it’s just used as a plain MAC in the expected places.
arnarbi··on Biscuit authorization
Very much looking forward to the blog post! At G we’ve relied on some existing key replication APIs and the caveats have just been the simplest possible and rigid protobufs for each application.

(And thanks! I’m still citing your Velvet Underground comparison.)

arnarbi··on Biscuit authorization
(author of the paper in your second link, which the Biscuit spec also references)

Macaroons are an abstract thing. They don't define an encoding format, nor any semantics or language for the caveats. So a lot of detail is left up to the implementer, which is why they're fairly hard to use. Their basic form is also built on symemtric keys and hmacs, so only the issuer can verify, which is a significant limitation in the real world.

AFAICT Biscuits define those (protobuf for encoding, a logic language for the caveats) and use public keys instead to allow other parties than the issuer (=holder of the root key) to verify them.

arnarbi··on Photographer showcases the power of iPhone 15 pro max
Light painting photography has been around for quite a long time, I wouldn’t call it an imitation of digital art, nor non-traditional.

https://lightpaintingphotography.com/light-painting-history/

arnarbi··on Why even let users set their own passwords?
> how can you use public and private keys without encryption

Signing is different from encrypting.

https://en.wikipedia.org/wiki/Digital_Signature_Algorithm

And yes, 100% agree we can do better in terms of explaining passkeys. The hardest part is that different audiences require such different approaches.

arnarbi··on Why even let users set their own passwords?
There isn’t encryption nor a generated password involved with passkeys. Rather, your device generates a private key and stores it in your password manager - and only the public key is sent to the server.

When you log in later, the server uses an API to give your client a random unique challenge (just some bytes), you unlock and approve and the pwm signs some stuff including the challenge and sends the signature back. The server verifies with the public key, and is satisfied because only your private key could’ve generated that signature.

But you are exactly right - this is detail and the UX is essentially the same as having the pwm generate random passwords that you never get to see or copy.

arnarbi··on Passkeys now support external providers
> Will both be able to log me in to a Google Account? Or do I need an Android phone for that?

Yes and no, respectively.

arnarbi··on Passkeys now support external providers
You can still use your password today on a Google account with passkeys. And account recovery via other means (depending on a lot of things) is still available.

It's too early to completely replace all methods with passkeys, but the hope is that as they gain better support and understanding websites will be able to make other methods rare/exceptional. For exceptional cases such as account recovery, as opposed to day-to-day account sign-in, there is room to apply a lot of other abuse signals and other methods to make it harder for attackers.

More here: https://security.googleblog.com/2023/05/so-long-passwords-th....

arnarbi··on Google Introduces Passkey Authentication
There’s more info on this here: https://research.nccgroup.com/wp-content/uploads/2022/04/NCC...

In short we can only verify PINs, and only a very limited number of guesses, enforced by hardware.

arnarbi··on Passkeys: The beginning of the end of the password
> How can my smartphone know whether I'm interacting with the correct website on my laptop?

Because when evil.example requests the passkey it has to do so through browser APIs [1], and it can't lie about its domain name. Your browser is what reaches out to your phone, which is how your phone learns what domain you are actually on.

[1] https://www.w3.org/TR/webauthn-2/

In this case your phone wouldn't even show you your passkeys from good.example.

arnarbi··on Passkeys: The beginning of the end of the password
You can create a passkey for your account on their device, e.g. by selecting "Use another device" when creating it, and scanning the QR with their phone (their phone does not need to be signed in to your account).

And if you both use iOS, you can Airdrop a passkey: https://support.apple.com/guide/iphone/share-passkeys-passwo...

arnarbi··on Lego Googol Machine
At the 350 rpm of the input gear it’d be early by only 1.3x10^80 times the age of the universe.
arnarbi··on High-res image reconstruction with latent diffusion models from human brain
Subject 4 in the first line also looks very different from the ground truth, but clearly an airliner. I'm curious if there is also a closer match to that one in the set.
arnarbi··on Visually symmetric words
“pod” is rotationally symmetric and in the dictionary.
arnarbi··on Ncdu – NCurses Disk Usage
OP’s link has a list at the bottom with alternatives (that in itself is very cool). It lists both gdu and a Rust alternative called dua.
arnarbi··on Ask HN: WebAuthn – Replace Password or Second Factor?
> New developments like Apple's PassKeys

Note that passkeys come out of the collaboration of many industry partners within FIDO and WebAuthn. I.e. it's not "Apple PassKeys" -- just "passkeys".

See for example https://fidoalliance.org/passkeys/

(disclaimer/context: I work on passkeys at Google)

arnarbi··on Bringing passkeys to Android and Chrome
There will be, in the password manager settings on Android (under Settings, "Passwords & Accounts" and "Google"), but it will unfortunately take a couple of more days to roll that out to the beta.
arnarbi··on Two envelopes problem
Ah! Apologies, I did not read that properly.
arnarbi··on Two envelopes problem
Try thinking of it this way: The information is in which door the host does not open.

2/3rds of the time, your initial choice will be a goat. In those cases the host deliberately avoids opening the remaining door with a car - thereby telling you where it is.

If you chose the car door initially, then indeed you get no new information. But that only happens 1/3rd of the time.

I am curious if that changes your opinion.

arnarbi··on Ask HN: Teach me something new
(Apologies if I’m misunderstanding and you already know this.)

Not specifying a register (e.g. "0) just means the commands, including x and p, will use the " register, aka the unnamed register. By itself, p is equivalent to ""p.

:reg shows you all the registers and their current contents.

This seems like a good description: https://www.brianstorti.com/vim-registers/

Edit: Oh I guess you are saying it was unexpected that delete or change commands clobber yanked contents of the unnamed register.

arnarbi··on Ask HN: Teach me something new
xp is a quick way in vi to swap two adjacent characters (a common typo correction)

Ctrl-T does the same in emacs, and also in readline based programs like bash, as well as everywhere on macOS.

arnarbi··on Ask HN: Teach me something new
This also works in any native text input on macOS.
arnarbi··on Zero rupee note (2015)
Er, “employer” should read “employee” in the last paragraph.
arnarbi··on Zero rupee note (2015)
In the short term yes. But that is exactly what the industry and employers rely on.

Long term, tipping just means these employers can compete for labor against other (non-tip) industries forced to pay higher, and without having to reflect that additional cost in their prices to the end customer.

The idea that the employer is the one getting the tip is a myth, and a well bolstered one for obvious reasons.

arnarbi··on Apple unveils contactless payments via Tap to Pay on iPhone
Because it doesn’t actually work that way and it’s in no way an “option”. In reality tipping becomes expected and people withhold it when the service does /not/ meet expectation. That’s awkward and too confrontational for a lot of people so they’ll just tip anyways.

Personally I also find it exhausting to have to judge every single service I get. So I just tip a standard amount, there’s absolutely no feedback involved. So there is no point.

There are lots of downsides with very few upsides.

arnarbi··on Interactive piano reference to major and minor scales
The text at the top specifies it's the natural minor (Aeolian)
arnarbi··on Ultorg: A General-Purpose User Interface for Relational Databases (2020) [video]
The part of the demo where he creates a computed field inside the sub-result (in a hierarchy), and then uses that in an aggregate at the next level above is excellent. And then he can still see the constituent rows contributing to the aggregate and continue adding filters on them.

I am sold, even as someone comfortable with writing complex queries.

← PreviousPage 2 of 16Next →