zlacker

Something I was wondering about in this case ... They tracked the killer through his use of burner phones 10 years ago.

And I obviously know phones continuously ping cell towers.

So that means somewhere out there, there's a database of all cell phones and their locations for all time, with fine-grained resolution? They don't ever delete it, at least not in the last 10 years?

Or it's 1 database per cell provider ? I guess phones ping towers that are not owned by the company that provided the phone

I'd be curious if anyone has a link to a good summary of how this works, and the location tracking implications. Do they have to do a subpoena, or is it just a big database everyone's doing joins against? What's the resolution of the data?

I knew that people obviously get caught due to cell phone tracking -- it comes up in every one of these cases, like the Idaho killer recently. But I'm slightly surprised they reached back 10 years and did it

>>chubot+(OP)
There’s an LE chart floating around out there, slightly different than this[1] that shows how long carriers keep records.

AT&T in particular keeps records forever. I suspect all the other carriers do as well despite their claims.

1. https://media-cldnry.s-nbcnews.com/image/upload/t_nbcnews-ux...

>>chubot+(OP)
You get some insight into it if you carefully read Mahoun’s blog about looking for Bill Ewasko.

>>bombca+l1
This?

https://www.otherhand.org/home-page/search-and-rescue/search...

Bombcar, my good HN buddy, it'd be great if you included the link rather than a few keywords. I've never heard of any of these people before. Additionally, lots of people won't bother to search and won't ever see the cool, interesting thing. Unless that's your intent.. ;)

<3

>>chubot+(OP)
I've been running WiGLE on my phones since 2013 or so and I have 2.1 million observations recorded, notably I don't run this 24/7. I'm just one guy with a phone and I wager I could sus out some interesting patterns from my neighborhood and city. I fully plan to as well, it's one of those back burner projects I keep plotting on. That said it really opened my eyes after they added Bluetooth support sometime ago. I can tell when particular nextdoor neighbors are home just by seeing their Galaxy Watch or earbuds in my network list. I can't even imagine what clever people could do with a continuous stream of wireless observations from thousands of points all recorded 24/7. It makes me feel naked in a way but I trust no one will bother to look my way and I go on with my life but I am at least aware of it. I wonder if normal people would care that corporations and the government can see their travels, patterns and habits and what can be potentially be surmised by such data.

>>chubot+(OP)
> So that means somewhere out there, there's a database of all cell phones and their locations for all time, with fine-grained resolution? They don't ever delete it, at least not in the last 10 years?

If this surprises you, what do you think the carriers did with all their SMS traffic?

>>milsor+W2
> I can tell when particular nextdoor neighbors are home just by seeing their Galaxy Watch or earbuds in my network list.

I know when the mail arrives even though I can't see the mailbox from where I'm sitting. The number of recognizable bluetooth and wifi beacons is large enough that you can draw all kinds of very privacy invasive conclusions from them. Occupancy is one thing, but also patterns of behavior, and more. Very, very scary stuff and really easy to abuse.

>>runjak+l
Can confirm AT&T retention. In the late 90's they were already approaching trillions with-a-T metadatums.

Federal LE, telecoms, and "the intelligence community" have always worked closely together.

>>gumby+Y2
I worked the SMS department of a major carrier about 15 years ago.

We only kept messages for about 2 weeks.

Long term storage would have been crazy expensive and I'm pretty it's illegal to mine that data.

>>metada+u2
My intent was to breadcrumb enough but I didn't have the availability to search :)

>>metada+u2
Wow, that was fascinating. Reminded me of the search for MH370.

>>nocoin+Kd
Now that I'm at computer I can note that Bill was found: https://youtu.be/2J9wsJb8P1Y

>>bombca+oe
Thanks! Sorry to be that guy, but is there a text write-up of the conclusion you’d recommend? I searched and came across some local news articles and Reddit posts but nothing as detailed as the blog you mentioned (which surprised me that it didn’t have an update including the discovery).

>>nocoin+Xg
The bones were found in a very strange location "the last place you'd look" and it didn't make sense why he'd end up there - the video provides (during a walk to where the remains were found) about "how would an experienced hiker end up like this" and basically lays out how each step of the journey "made sense" and the last attempt was based on a relatively low detail map - he was heading cross-country to try to get to a road in the shortest distance possible.

It also drives home how you should not deviate from planned travel and always try to back track, but I digress.

>>bombca+il
Excellent advice, appreciate the digression.

>>runjak+l
AT&T doesn't merely keep call record data, but does so explicitly to share it with law enforcement. Going back to the 1980s.

<https://www.eff.org/cases/hemisphere>

<https://en.wikipedia.org/wiki/Call_detail_record>

<http://epic.org/privacy/nsa/Section-215-Order-to-Verizon.pdf>

>>themer+C8
> Long term storage would have been crazy expensive

Not really, an SMS is probably only 200-500bytes to store with metadata so let’s say 1kb just for simplicity.

This site estimates 8.4 trillion SMS per year, globally [0].

So that’s only 8.4trillion kb or 8.4 petabytes.

That’s big, but that’s the whole world.

For comparison, google stores about 2,500 petabytes per day. [1]

So I would guess that not only do they store this forever. There’s also lots of copies and that there’s probably LE firms mining texts for all sorts of pattern recognition, AI stuff.

[0] https://www.sellcell.com/blog/how-many-text-messages-are-sen...

[1] https://skill-lync.com/blogs/how-google-handles-over-40000-p...