Well thanks a buncharoonie. Today is a public holiday here in Finland. I was supposed to go hunting for the week-end, but that got cancelled because my car won’t start (again…) So instead I figured I’d go out into the forest around my place for a long walk. But instead, it looks like I’m gonna be sitting home and pour over this Qubes thing.
Damn you man…
I am sure you are gonna love it 
I’m not sure what OS you use right now, but I know that you are quite security-conscious, and I think it would probably fit you quite well
Im assuming they think you have it sandboxed 
Didn’t Flash end EOL a few days ago? How would that work now?
Wrong thinking: this is a page to convince a security-conscious invididual to switch to Qubes (who, presumably doesn’t run Qubes yet). And how to they propose to convince you they’re serious? By feeding you some Flash shit.
Hmm, you know what? I’m not terribly convinced all of the sudden.
But I’m generous: I assume the Qubes folks outsourced their webpage design to some web outfit. Those aren’t the brighest bulbs in the lighting department.
Still, Flash… Not only is it officially dead, it’s been banned from the computer of anybody with any concept of security for years. First looks matter, and this don’t look good.
Anyhow, I took a look, and this is exactly what I thought it was: a sandboxing environment with some glue around it to make it work day to day by your average Joe Blow. Nice, but like I told Amal, this is not economical in RAM, disk or CPU. Not something that I could run on my measly home laptop.
Still, nice. I’ll have to give it a whirl on one of the work machines. Although I already run most of my stuff in Virtualbox VMs there, and that already has a whole lot of integration features built in - the main difference being essentially the granularity of the compartmentalization.
Ok @anon3825968, how do you do this? How are you so damn security conscious?
I try to be, but it’s like I get security fatigue after a while. I set everything up, and then I barely use it. I set up Duckduckgo and keep using !g because search results of scientific articles are just lightyears ahead.
The only reason I only have Facebook is because my study colleagues set up groups for sharing of exam questions etc. up there, and are too stubborn to change. even though I set up a custom website for that. Same thing with other social media/messaging apps
How do I unchain myself from this?
It is one of the many cogs on the process.
Same thing to decrypt a password for a WAP2 protected network. Many small issues which allow for a complex attack method.
… then write your own. 
you are definitely right with your options there though!! 
+1 to that! XD
Although…
Considering how secure is Linux…
If you do a good install of a barebones distro, even like Arch, and then do all the config yourself, without taking shortcuts, then it will be the most secure option you have.
But who would afford the time to do that? 
I got to agree here…
It actually gets me thinking that the marketing team is fully isolated from the dev team.
Which leads me to assume they have multiple isolated teams… Which is a formula for security issues…
I can only trust security from a company as much as I trust it’s most incompetent team member: Upper management.
And even if they outsource webdev… Upper management should have approved that… which means they have 0 clue about security.
Ever since I saw Docker containers being accessed by applications on other containers, I got skeptical of that as well… I believe they already fixed the latest of that. it’s a fairly old hack, but happened quite a few times in the past, therefore bound to happening again… 
I think this is the 5th derail track already!! 
Nothing against that, though!
That is the big problem with security!!
A system is only as secure as the use you make out of it.
Just keep in mind that someone will always be able to find a way into anywhere.
And the only way to keep your computer 100% secure is by keeping it shut down 100% of the time.
Just like with house security.
The goal is not to make it impossible for anyone to get in, but to make it so hard that there will be other better targets for invaders.
Unless you are a high profile target… dodging wide hit tactics and automated hacks should be enough.
You can’t. Not fully. Google is so good and so pervasive it makes it exceedingly hard to wean yourself off it.
You can mitigate the problem at the cost of make your life a miserable hell online. That’s what I do. For instance:
I use Noscript and uBlock origin in paranoia mode by default. Whenever I hit an unknown page, it usually takes 5 or 6 reloads before I get it working with the exact number of scripts / features it needs and none of the security / privacy invasion shit. That’s for EVERY SINGLE PAGE on the internet that I don’t visit regularly. And on pages I do visit regularly (this forum for example), I enable and disable things as needed. Discourse looks rather broken for me.
I never watch a Youtube video in Youtube. I redirect all video URLs to whichever invidious instance happens to be working at the moment. I only use Youtube to search a video or browse some guy’s youtube channel, because inviduous sucks for that.
I browser through TOR whenever possible, and I have to switch circuits regularly, as Google / Akamai / CloudFlare decides my exit node isn’t to their liking and blocks me.
I use OSM instead of Google Map whenever possible.
My browser is set to serve up bullshit meta info on a per-page basis if I can get away with it (rarely), or change the browser profile whenever I restart the browser. Similarly, all my cookies are cleared every hour or upon exit, unless it gets too painful (here for instance). That means I have to log back into all the sites on which I have an acocunt regularly.
I use DDG in html mode. That means no back button when I click on a search result - meaning I have to open it in another tab. When DDG doesn’t cut it, I try Bing, then StartPage. Only then do I hit Google Search - but I have to reenable a bunch of scripts first and reload, because Google Search plain doesn’t work without any Javascript (there’s a meta refresh bomb in their HTML-only search result page to make you believe it doesn’t work, if you’re wondering).
I translate things using Microsoft Translator whenever possible. When that doesn’t work, I hit Google Translate - and reenable even more script.
I use a bunch of Tampermonkey scripts to decrapify Google services, and more importantly, make them usable without the stupid consent cookies (which I clear regularly - see above), and without clicking on “No I don’t want a Youtube account” every 10 minutes. I write GM scripts for other websites I patronize that do things I don’t like, that Noscript or uBo can’t take care off in a satisfying way.
I host jQuery on my local webserver and my host file resolves ajax.google.com to my local address.
I have several browsers in several user accounts to to different things.
I run my own mail server.
Of course, no Teams, Zoom, Skype… I tell people to call me on the phone.
Etc etc etc. The list is endless. Life online for someone who truly wants to mitigate personal data leakage and tracking is very complicated, and even I can’t fully escape Google and all the other big data fuckers. But I voluntary submit to that regimen. If nothing else, it reminds me what the web is really like when you remove the blinders and the sugarcoating: a nasty work of tracking, advertising and massive software stack inefficiency at the best of times. But I’ve been doing it for many, many years, so I don’t mind too much.
It gets heavier and heavier as years go by though, to the point that I see myself dropping off the internet within 5 years. Possibly earlier if Google manages to push webassembly.
I wish I had the patience to follow your guidelines… but they would make my online life hell…
What I do:
Then the most important, but I must say I would neeeeeever do that…
I must say I never did that. I mean… I literally must say I never did that.
But if Amal were to check my IP logs, I bet it would scramble between London, Leeds, Manchester, Liverpool…
Unfortunately if I go too far from UK the lag becomes an issue…
Problem with TOR is that it’s only as private as the number of users in your area.
If a million users in your city, with your same ISP connect into TOR… you are truly anonymous.
If you are the only user in your city with your ISP connecting to TOR at this time… I can figure out what is your current exit node, therefore what is your browsing.
But again, unless you are a high profile target, no one will spend that many resources to track you like that.
Hmm no. Read up on TOR, that’s not how it works.
The two problems with TOR are these:
If one bad actor runs more than 50% of the nodes (it’s been surmised that the NSA does, which is entirely possible) then it can link back to your original IP by virtue of owning both the entry and exit node and matching the metadata of what goes in vs what comes back out. At least if that’s the case, I’ll be tracked by one or two bad actors, and probably not by Google or CloudFlare, which at this point are more of a threat than the NSA, because Google or CloudFlare go for a lower-hanging fruit which is…
It’s far easier to deanonymize someone on TOR using Javascript
I did read on TOR.
Thing is… if there are few enough users connected to the entry node, and few enough requests coming out of the exit node…
Given that the delay between entry-to-exit and exit-to-entry node is, within large numbers of requests, quite a constant there (tied to packet size, of course)…
Then I just need to use a few of the many available “clues” to make a match.
On the simplest scenario:
You are the only user connecting to entry node A.
You send a packet with X size in…
if a packet with that same size and same metadata comes out from exit node B exactly .3ms later, and I know from my tests that a packet with that size takes .3ms to reach exit node B coming from entry node A…
I can make a match.
Of course that would require me to have enough requests being sent from my own data farms from many entry nodes, and another data farm to be connected to monytor networks around many exit nodes…
And your request must be the only one within a margin of error from those alleged .3ms at that exact time.
It is doable, but as I said… you need to be a person of interest for someone to pull this one off.
Again, only if you can monitor the entry node and the exit node.
Out here in the sticks, there must be 3 or 4 users of my local GSM tower. I’m probably the only one using TOR. Say you monitor the traffic at the base of the tower somehow (good luck with that, it’s 0 degree here today - 260K for those who insist on metric )
I have a circuit open with an entry point in Prague, 4 transit nodes in Paris, Brisbane, New-York and Seoul (I usually use 6 node circuits) and one exit point in Toronto.
You, at the base of the tower, can only tell I’m sending TOR traffic to Prague. That’s it. You’d have to be monitoring the exit node in Toronto too - or possibly monitor the entry point in Prague also if you don’t want to freeze your balls off here - to match the traffic out the exit node in Toronto and the traffic into the entry node in Prague.
And to answer that boilerplate, which I keep hearing all too often:
The NSA monitors persons of interest and leaves the rest of us poor schmuks alone, because they’re interested in criminal behaviors. Although that’s getting less and less true as the cost of monitoring plummets.
Google, Amazon, Akamai, Facebook, CloudFlare, Microsoft mass-monitor everybody, because they’re interested in making money out of it. The upside of this is, they don’t monitor someone if the cost of doing so is higher than what they can make out of selling the resulting data. So make it hard enough and you should be okay: they have plenty of other schmuks to exploit.
As to the why, there’s a good line from The Girl in the movie Anon that I always quote: it’s not that I have something to hide, it’s just that I have nothing that I want you to see.
Metric system countries use Celsius as a day to day temperature unit.
Yes. but you don’t need to actually own the nodes.
You can also achieve the same results if you can monitor the network around the exit node (which is a static)
and then to monitor the network around the user sending the request into an entry node.
But this only works if you are a person of interest (to justify the size of the operation needed) and few people are using TOR around you.
You still need to be a rather all-reaching actor to pull it off.
Exit nodes aren’t static: they rotate every 10 minutes.
I meant the network around each exit node being a static network. Not the exit being static for you.
Hence why I mention “person of interest”.
Exactly why I say most of us don’t need to worry
Nice quote!
Right. So my strict regimen of (trying) to control who knows what of me removes me from the list of persons of interest for Big Data - and probably puts me squarely on the NSA’s, since my internet usage patterns are unusual
LOL.
Oh, the Irony…
Not that I’m too far from that.
Actually, most of us should worry, because it’s a lot harder to make it cost-ineffective to gather data on you than it is for big data to gather the data.
I guarantee you most internet users don’t do enough to protect their personal data. Most people use Noscript and Ghostery and think that’s good enough. How do I know it’s not enough (and why do I do more)? Because if it was, Google and all the others would wage an all-out war on those tools. And they don’t. Which means they’re perfectly happy to let people run them, to lull them into a false sense of security.
When you strip away the layers of data-gathering cruft like I do, the internet become essentially unusable, which is precisely what the motherfuckers want. And it works: you yourself say you experience security fatigue.
