Hacker News Reader

I opened the document expecting to sympathize with Lin. In my imagination this was some FBI surveillance state overreach, or "cyberharassment" thing getting overblown.

Instead, reading through the allegations, Lin came off as abominable. Contrary to your conclusion that this shows the FBI takes cyber harassment seriously, it seems like law enforcement generally allowed Lin to publicly subject this poor woman to psychological torture for a couple years before doing anything about it.

Provided the allegations are true, whatever sentence he gets will not be enough...

Why would you expect to sympathize with him? Is it usual (i.e more likely than not) that cases of cyberharassment are overblown, or based on faulty knowledge, or outright false? It's strange to me that someone would immediately be so skeptical, unless the rate of false accusations is higher than the rate of true accusations (at least insofar as determined by the imperfect legal system). Is that the case? I'd like to be corrected if it is.

The title changed! I hate when Hacker News does this. Before, it was:

> FBI arrests PureVPN user with log data that was said to not exist (justice.gov)

Nothing about cyberstalking. I too figured it was going to be something stupid like piracy.

I don't know what the rate of false or overblown allegations is. I wrote above about my intuition and not my judgment after a considered study of the issues and evidence. It's just what I expected.

As for why my intuition went that way, I suspect it's because, from the title, I felt like Lin was wronged by the VPN company which misled him, I generally distrust the FBI, deanonymizing VPN traffic seems troubling to me, "cyberharassment" sounds more like being mean to people online than the things Lin did and I don't think sending rude words to people should be criminal, and I suppose that I personally am more likely to be (falsely) accused of cyberharassment than I am to be victimized by it which probably results in implicitly identifying more with the attacker than the victim.

I don't think "cyberharassment" sounds like being mean online at all, I think it's what most would consider harassment but conducted through the Internet to the extent that it is possible to do so, and it may even be more pernicious, since it is very easy to stalk people and submit anonymous comments via the Internet. In the same way, I'd also assume you personally more likely to be falsely accused of rape than you are to be victimized by it - simply because you're probably not a rapist. It does not have much to do with the fact that most rape allegations are true, at least to the extent we determine in legal courts.

Maybe we should think about how we act online if we feel that there is a significant risk of being accused of cyberharassment, even if such conduct should not be illegal (and I suspect we also disagree on where the line ought to be drawn here).

"Harassment" can refer to a variety of behaviors. I feel "harassed" by telemarketers insisting I apply for a loan or buy whatever they're selling. If someone were to repeatedly bother me at the mall trying to sell me something, I'd think of that as them "harassing" me to buy something etc. Harassing seems to me like repeatedly being a nuisance.

In the context of criminal behavior I'd expect harassment to be a campaign of intentionally bothering someone and invading their space. I'd assume someone like Lin might be guilty of criminal harassment if he waited outside this girl's house and lewdly propositioned her every day, bothered her at work, etc.

I think we should have a higher bar for what constitutes cyberharassment because it's so much less invasive and threatening than physical harassment and so much easier to ignore. It's also possible that it's easier to inadvertently participate in cyberharassment when you can't see how the other person is reacting to you or feel how inappropriate the behavior is.

What Lin did in this case is far beyond any of what I described above. I think of his behavior as transcending what I'd describe as cyberharassment - in my earlier comment I called it psychological torture and I think that's far more apt than cyberharassment which seems too milquetoast a phrase for what happened here.

> Provided the allegations are true, whatever sentence he gets will not be enough...

He has already been sentenced to 17.5 years in federal prison (and federal time has no parole).

https://www.justice.gov/opa/pr/massachusetts-man-sentenced-m...

> has no parole

Also, this is not going to be some "white collar resort" prison!

Anyway, it's worth noting how Lin's behavior was not just criminal to the point of outright sociopathy, but also - like most criminals' - incredibly lazy and complacent. That anyone knowledgeable in a CS-related field could target a vulnerable individual in such an extreme, sustained and long-lasting cyber-abuse campaign and expect to get away with it, is rather surprising to say the least.

>Provided the allegations are true, whatever sentence he gets will not be enough...

While Lin is undoubtedly a terrible person, I still think 17.5 years in federal prison is rather excessive unless you've actually killed someone.

I mean, the special agent even calls it a fact:

"the fact that VPN's track activity with logs."

>As an aside, it's good to see another example that the FBI does actually investigate cases of cyberharrasment and takes doxing seriously, contrary to popular opinion.

Seriously, all this over a roommate found on Craigslist. I did like the part where his victim repeatedly smashed his computer monitor with a hammer, though.

Again, its easier to buy a cheap VPS in a country that is at odds with the one you're in. Then, any intelligence the other country gets will likely not be sent to the country of residence.

Ally countries usually have extradition treaties, and have a greater chance of sharing intel.

Correlation is still quite easy in that scenario, unless you're sharing that VPS with a thousand other people.

NSA has your international traffic logged and timestamped. If you're doing nefarious activity from a non-MLAT country's IP, and you're the only one connecting to it from the US, it's a simple matter of parallel construction for domestic law enforcement to pop you.

>it's a simple matter of parallel construction for domestic law enforcement to pop you.

This is more just putting some circumstantial evidence together. Parallel construction refers to going out and getting evidence a second time, because you don’t want to reveal how you got it the first time. Usually because the first time it was collected was without a warrant, or collected from a sources or via a method you want to keep secret. So the evidence is reconstructed in parallel.

> NSA has your international traffic logged and timestamped. If you're doing nefarious activity from a non-MLAT country's IP, and you're the only one connecting to it from the US, it's a simple matter of parallel construction for domestic law enforcement to pop you.

Couldn't that be countered by using two foreign IPs for your VPN, one for your inbound connection and another for your outbound traffic? It would make it more difficult to correlate your US IP with the outbound VPN traffic without extra information.

In Brazil law requires services to keep logs for a year. This made me very wary of services like ProtonVPN that put some servers there. A lot of people trust them, but I for some reason don't...

>A lot of people trust them, but I for some reason don't...

No offense or anything, but it sounds like you actually have very GOOD reasons for not trusting them.

We don't keep logs in Brazil, and as we have no operations in Brazil, so this is not a point that particularly concerns us.

We have something known as Secure Core VPN that offers additional protection if our Brazilian ISP starts monitoring our connection: https://protonvpn.com/support/secure-core-vpn/

One important distinction could be that these were HTTP requests made on his account page or other. If he logs into the account to update billing, etc that would certainly be logged.

All of what is claimed in the statement would be possible even if we assume no logs are stored for the VPN Server they run, which makes no guarantees about access to their HTTP properties.

Bandwidth counting can be accomplished without keeping "logs" per se, and with WireGuard, I think there would be very little reason to attempt to limit connections.

WireGuard is very hard to run without logging. It simply wasn’t designed for that and the maintainer was paid once to write “a rootkit-like” piece of code for a VPN provider which hired him to help them fix that.

It’s still an open question afaik

edit: I've worded this weird. I was typing on my phone at lunch stuff I'd just learned this morning[0] which referenced this[1] article saying running a log-less Wireguard might not be possible.

AirVPN in [0]:

> "Wireguard, in its current state, not only is dangerous because it lacks basic features and is an experimental software, but it also weakens dangerously the anonymity layer."

and Perfect Privacy:

> "WireGuard has no dynamic address management, the client addresses are fixed. That means we would have to register every active device of our customers and assign the static IP addresses on each of our VPN servers. [...]"

Things may have changed, but it appears that running a log-less vpn provider is actually more complicated with Wireguard than at first glance. Namely the issues around DynamicIPs.

[0]: https://restoreprivacy.com/wireguard/ [1]: https://www.perfect-privacy.com/blog/2018/10/10/wireguard-vp...

I run Wireguard on my systems and to my knowledge it does not log anything on my Linux systems (not that I intended specifically to set it up that way, it is just something that I noticed). Can you tell me where I can find these logs which I am seemingly unaware of? I do know that my iOS app logs things but I'm talking about Linux.

Your statement is vacant without an explanation of what kind of logging Wireguard requires. Currently, all it does is attempt to scare the user with the word "rootkit".

FWIW Jason called it A Defensive Rootkit [0].

But the parent post is wrong, the defensive rootkit is not to prevent logging, it's to prevent extracting the configuration from the kernel. It effectively makes the WireGuard configuration write-only from the perspective of userspace. WireGuard does not do any access logging by default as far as I am aware.

[0]: https://lists.zx2c4.com/pipermail/wireguard/2017-November/00...

At least AzireVPN has some claims of not logging Wireguard:

https://www.azirevpn.com/docs/security#blind-operator-mode

>WireGuard is very hard to run without logging. It simply wasn’t designed for that and the maintainer was paid once to write “a rootkit-like” piece of code for a VPN provider which hired him to help them fix that.

This is a really bizarre misunderstanding of the events.

Wireguard does not generate any log entries by default.

zx2c4 wrote a rootkit which makes it more difficult to retrieve connected users IPs from a running wireguard instance.

>As an aside, it's good to see another example that the FBI does actually investigate cases of cyberharrasment and takes doxing seriously, contrary to popular opinion.

Probably cos they investigated certain cases and found nothing, to the great dismay of certain people who wanted to be seen as victims and others who wanted to be seen defending them.

Why not just use Tor?

For some things, Tor works just fine. For some sites, Tor is either blocked, or may send you through endless captchas. At least, that has been my experience.