It's interesting, the article discusses that the request is for a relatively short period of time (35 minutes) - however for a popular website that could still be thousands of viewers.
It suggests that there's something they're specifically looking to match against - but if that was the case, I would think that specific IP addresses could be provided in the request - e.g. Did any one of these 10 IP addresses view the article in the time period. Much more specific and likely easier to justify.
I'd rampantly speculate that perhaps that time matches to the link being posted in a pedophilia related forum (with the forum behind TOR) - and the FBI would like to get a list of who might have followed it there.
I agree with your speculation, this looks like a timing attack regarding a Tor or VPN server.
That said, while catching pedos is a good thing, their methods are still concerning for regular law abiding citizens. The road to hell is paved with good intentions…
A lot of people use “pedophilia” to mean “sexual abuse of children” and “pedophile” or “pedo” to mean “sexual abuser of children”. (E.g., RMS’s infamous defense of “consensual pedophilia” which, merits of the intended sentiment to the side, isn’t even a coherent thing to defend or oppose except under the misuse of “pedophilia” to refer to an act and not an inclination.)
> Did any one of these 10 IP addresses view the article in the time period. Much more specific and likely easier to justify.
It is, but it also leaks information. Now people know you're looking at those IP addresses. If you were going to leak that, a major news outlet is probably not the place you want it to leak to.
You might be right and it might be an easy cover for a fishing expedition, but it doesn't seem inherently malicious on its face.
That's only 4 billion hashes to search the entire IPv4 address space. I don't think the entire search space is big enough for that really to provide much privacy while still being functional. Whatever you put in the way, that's somewhere in the neighbourhood of cracking a 6 character lowercase+number password.
But this is a rare case where you can use a derivation-extension function like PBKDF2 to make the hashing take practically any amount of time you want. You could probably make each run take a full minute on USA Today's hardware and still have your results in a reasonable amount of time.
Tor is vulnerable to statistical analysis with which if you time it perfectly you can link exit nodes with specific user. [1] The fact that FBI is asking for those specific 35 minutes suggests that they are onto something like this.
Entirely likely, though possibly there are potentially flaws that could allow people to identify it through other means.
I've never used TOR so I didn't realise that this would apply (and it makes perfect sense it would work that way).
It may be that whoever requested the data knows as little about it as I do (or more likely, they know a lot more about what they want and my speculation is completely wrong).
TOR is clunky and slow, it’s not outside the realm of possibility that someone would visit a dodgy forum over TOR, and use their regular browser for other web browsing.
Some sites can be difficult to even access over TOR, especially ones that are very JavaScript heavy or sit behind something like Cloudflare.
>I'd rampantly speculate that perhaps that time matches to the link being posted in a pedophilia related forum (with the forum behind TOR) - and the FBI would like to get a list of who might have followed it there.
now cue in all those stories of how people have been getting in trouble only because their ISP was using the same IP to NAT a crowd of customers.
The although both words have the same last two letters - "le", the third last letters are different. This produces a different syllable when pronounced.
A man smuggled half a tonne of $2 coins out of the Royal Australian Mint in his boots and lunch box and kept notes of when he exchanged the money, a court has been told.
This is interesting, particularly in light of skissane comments about melatonin being prescription only.
Here in Australia, you can buy Asthma Inhalers "over the counter" (no prespcription) from pharmacies (Salbutamol). They sometimes do ask if you're seeing a doctor for your asthma, but that's about it. (I take a regular treatment for it)
I think an interesting variation on this is that even if a safety system is not required by law, but is available - then disabling could constitute criminal negligence. Consider what happens if safety equipment on industrial equipment is disabled and injuries result. I'm fairly sure that criminal charges could result for whoever disabled the safety mechanisms (though the there are likely differences between workplace safety criminal law and road safety).
Keep in mind that it was also used as a significant contributing factor to replace a lot of major legacy IT systems (especially accounting systems) at big organisations (a lot of SAP rollouts in the late 90s had Y2K as part of the cost justifications).
The company I worked for ran a Y2K Remediation "Factory" for mainframe software - going through and change to 4 digits, checking for leap year issues, confirming various calculations still worked.
I worked on a full system replacement that was partially justified on the basis of (roughly) we can spend 0.3x and do y2k patches, or spend X and get a new system using more recent technologies and UIs.
There were still problems, but they were generally in less critical systems as likely major systems had been tested, and were remediated or replaced.
Keep in mind that there was often much more processing that occurred on desktop computers (traditional fat client) - so lots of effort was also expended on check desktop date rollover behaviour. Once place I worked at had to manually run test software on every computer they had (10's of thousands) because it needed reboots and remote management was more primitive (and less adopted) at the time.
I also wonder if encouraging this form of meetings could be viewed as discriminatory if you have employees who have physical disabilities that might make walking unpleasant or slow.
My wife has done walking meetings, particularly with very small groups (2 or 3) or phone meetings. But she currently has a knee problem that makes walking unpleasant and slow.
I guess checking whether all participants are fine with a walking meeting is an essential part of the process. CEOs demanding a certain style of favourite meeting does not go well with that.
It's interesting that the notes regarding transmission efficiency are not present in the similar speach given on the 4th of March (as opposed to the article link, which is the 3rd of March).
I'm not sure what to make of it. It might be pure paranoia on my part - but it is an interesting change in the welcome statements that are otherwise generally similar.
This makes me think of an interesting point around the potential virus outbreak in the US.
If there is a non-trivial outbreak, then could it effectively be (for the US), not just a mass virus management exercise - but be followed by a mass-bankrupting due to medical expenses (which are already a big deal just for "normal" life events)
i had a funny thought on reading bloomberg's summary of today: this is going to ruin trump because he'll handle this so poorly and he'll be up against a democractic primary candidate (either sanders or warren) that supports free healthcare
It suggests that there's something they're specifically looking to match against - but if that was the case, I would think that specific IP addresses could be provided in the request - e.g. Did any one of these 10 IP addresses view the article in the time period. Much more specific and likely easier to justify.
I'd rampantly speculate that perhaps that time matches to the link being posted in a pedophilia related forum (with the forum behind TOR) - and the FBI would like to get a list of who might have followed it there.