HNHacker News
TopNewBestAskShowJobs

xnyhps

1,013 karma · joined September 5, 2012

[ my public key: https://keybase.io/xnyhps; my proof: https://keybase.io/xnyhps/sigs/m-nsALHMR_0WYxhEtG6-ngXnDELyL0_4cxRt_ufsGkM ]
submissionscomments
xnyhps··on Wikimedia Maps Beta
Looks like a boundary between tiles there:

https://maps.wikimedia.org/osm-intl/4/2/6.png

https://maps.wikimedia.org/osm-intl/4/3/6.png

Placing labels on a map is quite a hard problem, especially when certain tiles are refreshed.

xnyhps··on H4sh – Exposing the Haskell List library as a set of shell utilities
I was going to post that I was happy to see a new project using hs-plugins, but then I noticed:

> h4sh was written during a cold, but sunny weekend in August 2005.

hs-plugins looks so cool, but I'm afraid it's bit-rotten and abandoned. :(

xnyhps··on Fighting spam with Haskell
I've had some mild success with https://hackage.haskell.org/package/dynamic-loader (which is basically just a Haskell wrapper around the C-functions in the RTS), but compared to writing normal Haskell, this is playing with fire as you loose all type-checking guarantees.
xnyhps··on Pidgin working on implementation of Facebook Messenger protocol
I've used a transport for a while, but I've never successfully done a file transfer using one. Voice/video over transports is going to be even harder to do. I think that makes it a dead-end for most people.

Trust is also a difficult problem with transports, as they often need to store your passwords in plain. A single server storing your Facebook, Google, Twitter and Microsoft passwords just does not sound like a great idea to me.

xnyhps··on Pidgin working on implementation of Facebook Messenger protocol
I think there are multiple reasons why multi-protocol mobile clients haven't taken off:

1. More complicated protocols. iMessage and Skype are many times more complicated than AIM/ICQ, MSN or YIM, with lots of layers of custom encryption and authentication. Reverse engineering of these protocols has never been close to done. 2. Battery usage. Each new account you add is (likely) another connection that you have to keep open. With one connection you can apply optimizations like the server bundling packets, but you can not do that when you need to keep connections to many different services.

xnyhps··on Pidgin working on implementation of Facebook Messenger protocol
They have already removed the xmpp_login OAuth permission, they could shut down the XMPP server any day. I am wondering why they would remove the permission so long before shutting down the server, maybe to not break every client at the same time?
xnyhps··on Hashpipe – Pipe iff the hash matches
There is another assumption:

4. You download the script successfully.

It is trivially easy for a MitM to interrupt the download of the script between two TLS packets, without any CA or server compromises. When you do:

    curl https://project.com/script.sh | sh
Then sh happily executes instructions as they come in. It may be that the script starts by moving important directories aside or by creating large temporary files, so if the script is incomplete the user may end up with a broken system. Or maybe if you're really unlucky an attacker might manage to truncate "rm -Rf /..." to "rm -Rf /".

With hashpipe, you are at least guaranteed to have the complete script before you run it. I still don't like the practice, but it is better.

xnyhps··on Peerwiki: all of Wikipedia on BitTorrent
It looks like it uses a single Wikipedia dump from over a year ago. Would be cool if it supported deltas somehow, so the network doesn't split when a new dump is used.
xnyhps··on No, it’s not the end of XMPP for Google Talk
That's the client to server encryption, which Google Talk has. Server to server encryption is missing.
xnyhps··on Raspberry Pi Z-Wave kit exposes private key
That doesn't protect you from truncation attacks: almost no server out there sends a close_notify before closing the TLS connection, so few tools will throw an error when you receive an incomplete file.

It may be hard to exploit, but executing a script before you have completely downloaded it is simply a really bad idea.

xnyhps··on Extracting the SuperFish certificate
Looks like the software they were using indeed requires a password on the private key:

http://www.komodia.com/wiki/index.php/Komodia%27s_Redirector...

(Down near the bottom, "sslpsd - To the password of the root CA (it's mandatory).")

xnyhps··on Extracting the SuperFish certificate
Assuming, from the password, that it uses "SSL Digestor" by Komodia, then yes, it should generate an invalid cert:

http://www.komodia.com/wiki/index.php?title=SSL_Digestor#Cer...

xnyhps··on Extracting the SuperFish certificate
They added a password to the private key which they didn't have to, so it's not true that they didn't attempt to obfuscate it at all. In practice that only bought them a couple of hours at most, but why add a password unless you're trying to stop people from using it?
xnyhps··on WhatsSpy Public
This has nothing to do with e2e encryption, this is only the metadata of your profile, status and when you are online. I thought it was pretty well known by now that WA shares that info of you with everyone who has your phone number. The content may be encrypted, but it is anything but private.
xnyhps··on Signing in to websites with SSH
So you have a situation where the client has already authenticated the server using TLS, and now the server wants to authenticate the client too. This solution wants to do that by setting up a new channel, authenticating the server /again/ and then also the client.

Yeah, I don't think this is a good idea.

xnyhps··on WhatsApp and Snapchat could be banned under new surveillance plans
The static AES key is for local log storage. This is about encrypted communications. They are working together with Open WhisperSystems to bring the e2e-encrypted TextSecure protocol to Whatsapp: https://whispersystems.org/blog/whatsapp/
xnyhps··on $300k for Cracking Telegram Encryption
It's even counter-productive:

If I find a few small flaws that are problematic, but not enough to claim the prize, I'll definitely keep them to myself until the contest ends. Instead of incentivizing people to share vulnerabilities with them, they incentivize hoarding bugs.

xnyhps··on Shout – A web IRC client
Looks like it isn't: https://github.com/erming/shout/issues/52
xnyhps··on [dead]
Let me shamelessly post this again: https://blog.thijsalkema.de/blog/2014/04/02/breaking-half-of...

Even if you'd consider their protocol secure enough (which I don't), this third-party app was written by someone who thought rand48 was good enough for long-term crypto keys. I'd stay far, far away from this.

xnyhps··on PolarSSL – Does anyone have any experiences with this OpenSSL competitor?
I've been looking into what options exist and what their pros and cons are for switching LuaSec from OpenSSL to something else (mostly in the context of Prosody's TLS support). These are my pros and cons so far, but it's a pretty specific application, so not all points may be important for others.

Pros:

* PolarSSL seems to be very clear code, well modularized and a much more sane API than OpenSSL.

* Supports setting cipher orders by TLS version.

* Cipher support is good (all the new ECC and SHA2 stuff is included), but it doesn't support SSLv2 or EXP ciphers (I'm counting this as a "pro", though).

Cons:

* Not quite as much support for working with X509 certificates. It parses the extension fields it knows about, but if you want to know more than those, you'll have to dig into ASN.1 yourself.

* Related to the previous point: I couldn't find support in converting different ASN.1 string formats. Probably will have to depend on another library to do that.

* I found a buffer overflow bug within a few hours of working with it [1], which is really not a good sign for a TLS library. However, it did get acknowledged within 24 hours (I've got an OpenSSL patch which has been sitting on their bugtracker without any reply for 6 months).

* Setting a different curve for ECC requires a compile time flag to be set, which is disabled by default.

* Using the server's cipher order or the client's is also a compile time flag, can not be set at run time.

[1] = https://github.com/polarssl/polarssl/issues/83

xnyhps··on Tor Instant Messaging Bundle
The assumption that JS is more secure than C.
xnyhps··on The New TextSecure: Privacy Beyond SMS
Getting carriers on board to make a texting replacement? Well, you can dream. ;)
xnyhps··on The New TextSecure: Privacy Beyond SMS
Not if you want it to be “SMS based”, by which I mean: use phone numbers as identifiers. A server can’t easily prove to another that it serves the user with a specific phone number. There’s no cryptographic proof possible, there’s no hostname part like in email. You can verify by sending a text message, but that gets expensive if you need to do it often.

This is trying to combine 3 points on Zooko's Triangle [1]: You want human-meaningful names (which phone numbers are, because they map to existing things), so you have to make a trade-off between decentralization and security. WhisperSystems opted for security for some reduced decentralization. For something that’s aiming to replace text messaging, I can’t really blame them for that choice.

[1] = https://en.wikipedia.org/wiki/Zooko%27s_triangle

xnyhps··on The New TextSecure: Privacy Beyond SMS
When I looked at it (which is already a couple of months ago), federation wasn't “just works™”. A new server needed to be explicitly permitted into the federated network by WhisperSystems. Every server has a full copy of the list of clients and on which server they are and servers trust other severs completely when they claim “we serve the user with phone number 555-123456”.
xnyhps··on WhatsApp issued takedown against alternative clients a week before acquisition
I don't think there ever was a serious risk for that for WhatsApp: multi-device support isn't just missing, they are actively making it a pain. There's no way to obtain your password, the password changes regularily, logging in with a second client kicks the old connection, etc.

So people would be forced to make a choice: use it on your phone, or on your computer. Aside from the group of people who don't have a smartphone, most people would chose phone.

xnyhps··on Off-the-Record Messaging Protocol v3
> OTR uses DH in a ratcheting protocol that requires attackers to continuously break new DH exchanges; it's not like TLS, where one exchange at the beginning of the session gives you the whole session.

Please correct me if I'm wrong, but as far as I know the required effort to break multiple DH exchanges doesn't scale linearly in the number of exchanges. A single successful index-calculus attack on the used group will make breaking additional key exchanges much easier.

xnyhps··on Off-the-Record Messaging Protocol v3
Synchronizing that info accross your devices is exactly something you'd need a protocol for. If you want to preserve the forward-secrecy that OTR offers, then you can't just "encrypt, upload" on one client and "download, decrypt" on another. You somehow need to do a key exchange between your own devices.
xnyhps··on Off-the-Record Messaging Protocol v3
> (btw there's nothing stopping you from adding OTR support to one of the open, reverse engineered, whatsapp clients)

Except, you know, DMCA notices: https://github.com/github/dmca/blob/master/2014-02-12-WhatsA....

Yes, this is complete bullshit. Especially how they try to use a DMCA notice for trademark infringement.

xnyhps··on Off-the-Record Messaging Protocol v3
Adium 1.6 will include libotr 4, it's already implemented in the nightlies.
xnyhps··on IRC Networks Under Systematic Attack From Governments
> Just looked up CRAM-MD5, and the password is used as a key to HMAC-MD5, which means you can at least store MD5'ed versions of the passwords.

In real CRAM-MD5 this is not true. It uses HMAC-MD5 of the key directly. To be able to calculate that, you need to do

    MD5((key XOR opad) || ...)
Which means that you either store "key XOR opad" (not meaningfully different from storing key), or an intermediate result from MD5, which is tricky.

Quakenet's authentication mechanisms, except for LEGACY-MD5, call MD5, SHA1 and SHA256 before using it as the key, so they could store just each of those different hashes (unsalted). The LEGACY-MD5 mechanism does require the plain text password to be known by the server.

← PreviousPage 3 of 4Next →