Sabitlenmiş Tweet
Will Strafach
6.5K posts
Will Strafach
@chronic
building great things. breaking others. | founder @guardianiosapp (acq @dnsfilter) | @[email protected]
San Francisco, CA Katılım Ekim 2009
5.8K Takip Edilen49.4K Takipçiler
interested in working with very unique and high volume network data to discover emerging threats? come work with me at DNSF!
job-boards.greenhouse.io/dnsfilter/jobs…
English
@JHDN @BrendanEich @powerpig @brave @guardianiosapp @mullvadnet just as a data point - we don’t retain data related to connectivity attribution either, and do not have the ability for any party, raid/subpoena/hackers/scammers/otherwise, to do so. we are a neutral carrier of network traffic and it is not our business who is doing what.
English
@BrendanEich @powerpig @brave @chronic @guardianiosapp @mullvadnet Mullvad got raided and the police came out with nothing
No logs policy
Regular audits
DNS based blocking filters
Open source apps
Fair pricing (no discounts)
No false advertising
No personal information on sign up
Accepts Monero & Cash
A fair amount of first party servers
English
@BrendanEich @powerpig @brave it’s an interesting feature idea for sure, since we intentionally use vanilla WireGuard and IKEv2, the credentials can technically be used manually on a router like GL iNet models, though officially we focus support on endpoints.
English
@mert @usedjpgtrader @BrendanEich @zooko @guardianiosapp @DNSFilter correct! it uses native system (iOS, macOS, Windows, Android) APIs to switch on system-wide.
English
@usedjpgtrader @BrendanEich @zooko @guardianiosapp @DNSFilter but it still tunnels traffic originating from outside the browser?
English
@BrendanEich @Lin1985N @richardkoo @DJRawDeal_MD @brave @lukemulks I would love to know as well. we’re always trying to get better, and dealing with GFW is a big one.
English
@Lin1985N @richardkoo @DJRawDeal_MD @brave @lukemulks What VPNs have you tested? Thanks, and cc'ing @chronic again.
English
@richardkoo @BrendanEich @DJRawDeal_MD @brave @lukemulks further, GFW sometimes allows things only to later disrupt/devgrade the connection.
English
@BrendanEich @DJRawDeal_MD @brave @lukemulks I tried the oft-advertised VPNs last year when I was there. Those VPNs never worked for me. I will be there again and am looking for better VPNs.
English
@richardkoo @BrendanEich @DJRawDeal_MD @brave @lukemulks China is tricky. I have anecdotally heard that our WireGuard support works, but not IKEv2, but allegedly it also may depend on the region and network (I guess hotel networks may be more relaxed). honestly we haven’t done a comprehensive study.
English
@DefuseSec please be careful if anyone reaches out. there is no publicly available method to get a “true” forensics image of the device. it may be wise to do a backup in Apple Configurator ASAP though to preserve any traces of whatever the border people did
English
If anyone is an expert on iOS forensics, specifically for an iPhone 15 Pro, please get in touch with me. This is for paid work so please let me know your background experience and rates.
English
@matthew_d_green @cryptoishard it was some wild west shit. per my recollection, I am fairly confident that they did not even constrain it to only the analytics FQDNs.
I went back-and-forth with senator Blumenthal’s office to try to help craft something that’d elicit more details from FB, but FB blew it off.
Will Strafach@chronic
from an instructional video for Project Atlas users. this is how they guide the user through trusting the Facebook Researxh Root Certificate. does this sound like a “rigorous consent flow” to anyone?
English
This thing Facebook did — running an MITM on Snapchat and other competitors’ TLS connections via their Onavo VPN — is so deeply messed up and evil that it completely changes my perspective on what that company is willing to do to its users.
English
"installd": 30ac18dff606a78723ad4988000a9c698cebcd77d796b66ab37962140b69f57c
120.46.152[.]202
English
NEW: Facebook snooped on Snapchat users' encrypted network traffic to study how they behaved, unsealed court documents reveal.
This was part of a secret program called "Project Ghostbusters," and even inside the company, it was very controversial.
techcrunch.com/2024/03/26/fac…
English
@HowitzJame12303 @brave @BrendanEich @brianbondy happy to try to help out here! a few questions:
1. by Desktop, would that be Windows or macOS?
2. are you using an In-App Purchase, or do you have a Brave account?
3. have you used “Contact Technical Support” in the app to send us device information and debug data?
English
What's a good alternative to @brave VPN? It's totally broken.
Desktop will load sites then gets a DNS issue in <30s.
Mobile version won't even let me log in.
Worked for about a week with little issue, but now useless.
Pretty terrible app :/
cc: @BrendanEich @brianbondy
English
@justinschuh @KimZetter how would the technique apply to an iPhone 5c, though?
English
This keeps coming up in panels/classrooms/conversation - people mischaracterize FBI v Apple case. FBI did *not* ask Apple to unlock SanBernardino phone. They wanted Apple to create a tool that would undermine iPhone security so they could unlock ANY iPhone wired.com/2016/02/apples…
English
@EnglishManDanB @BrendanEich if you’d be willing to indulge on testing to see if using the Brave VPN helps resolve your Brave Search problem, it’d be very interesting to find out (we’d especially want to know if the same issue occurs, as that should never happen). there should be a free 7-day trial, btw.
English
@Brave September stats
Browser (historical method)
MAU: 63.64M (-0.3% in Sep31 days)
DAU: 23.38M (HWM +1.8%; 28 day mean: 22.7M +2.3%)
DAU:MAU: 0.36
Browser (visible funnel method)
MAU: 54.65M (+1.9% Sep31)
DAU: 23.03M (HWM +2.4%; 28 day mean: 22.36M +2.6%)
DAU:MAU: 0.41
1/2
BrendanEich@BrendanEich
Browser (Dark+Visible-Funnel historical method) MAU: 63.83M (+7.5%) DAU: 22.96M (HWM +2.6%; 28-day mean: 22.18M +2.2%) DAU:MAU: 0.35 Browser (Visible-Funnel-only method) MAU: 51.61M (+3.8%) DAU: 22.49M (HWM +2.2%; 28-day mean: 21.79M +2.9%) DAU/MAU: 0.41 Search stats next. 1/3
CY
@EnglishManDanB @BrendanEich this is a hard problem to solve, but one idea I have is to consider looking through the Proton server list and finding a server in your area with the lowest amount of congestion; it may or may not help, but my thinking is that lesser-used servers might not be classified that way.
English
@EnglishManDanB @BrendanEich ah, that output actually helps explain it - it looks like they’ve got their IPs classified as a Network Sharing Device on one or more anti-abuse lists. this can happen when many users are exiting traffic through one single IPv4 address.
(1/2)
English