Hacker Newsnew | past | comments | ask | show | jobs | submit | lwhi's commentslogin

I was hoping for 'will 2036 be the year of the Linux desktop' ..

Uber really are the piranhas of the corporate world.

I can't imagine any depth they wouldn't dive to, in order to get a morsel to feed on.


The allegation that driver payouts are manipulated to:

1) Hook new drivers with better than average rates before tapering off 2) Take into account the age/model/value of the vehicle and what payments for it would look like in the market and dole out enough to cover costs but not "too much" that they're getting ahead of other drivers

Totally baseless and sourceless hearsay tho. Still, if true, really plays into the image of "there's no depth they won't go".


Add another: the various platforms talk to each other (or analyze driver movement) in order to manipulate order offerings in such a way as to discourage drivers from taking orders from more than one app at once. One app will wait until the other has confirmed an accepted order before deluging you with their own orders, all taking you in the opposite direction (which makes you late for one or more deliveries, giving cause to terminate your contract).

Source?

Pure anecdata. However, the change from the first two days I multi-apped and made almost 3 times my usual hourly rate, to the following weekend, when

>neither app would send me orders for up to half an hour

>as soon as one had assigned me and order, the other would start sending my multiple per minute

>all of these orders were either comically low-compensation (no tip), a 15-minute-plus drive away from the order I'd just accepted (to areas it had never sent me before), or both

was marked.


From the article: ”aggregate users' data without revealing their identities.”

Also from the article:

> Uber Intelligence will let advertisers securely combine their customer data with Uber's to help surface insights about their audiences, based on what they eat and where they travel.

So the companies have the identities. It sounds like they're going to be learning something about their customers, the question is just how much detail they'll get.


Everything? Ok, everything. Names "anonymmized" (but easily trackable) and the list of addresses. Why not sell it to get money? /s

Over/under on when someone is able to de-aggregate an identity down to an individual user?

I’ve got it on less than 6 months.


6 weeks would be optimistic

how do de-aggregation attacks or whatever you'd call this work?

One of the easiest methods is to find a different data source with overlap and use that to map real people to anonymized lists. Big tech companies find this super easy to do because of all the internal data they already have on everyone

More like 15 minutes.

That's how it starts

Why do you think this makes it better?

You think data tied to individual users isn't any worse? That privacy has no value?

I think they're suggesting that anonymized and potentially aggregated data can still have individual data extracted from it: https://www.theguardian.com/technology/2014/jun/27/new-york-...

What the heck? Where did I say either of those things?

Privacy is very important. That's why I think sharing of customer data - individual or aggregate - is bad.


You asked why someone would think aggregating would make it better.

Aggregating protects privacy when done properly.

It seems pretty obvious to me that sharing individual data is orders of magnitude worse than sharing aggregated data.

If you think they're the same, then you don't seem to value the privacy that aggregation provides.

So what am I misrepresenting about what you said?

I'm tired of false equivalences. One thing that's maybe slightly bad, and another thing that's super-super-bad, aren't equally bad.


And I'm tired of people acting like companies putting on a show of protecting our privacy is doing anything actually helpful. But you're right. I'm wrong and clearly don't care about privacy.

As a completely unrelated aside, I wonder how much social progress is hindered by people alienating people on their own side.


Just don't make false equivalences.

It's easy, just admit the two are not the same and move on. You don't need to get defensive about it. "I wonder" how much social progress is hindered by people making wrong statements and then getting defensive about it? Or by making snarky "completely unrelated" asides?


>Aggregating protects privacy when done properly.

When done properly is going a lot of heavy lifting there. Time and time again it's been found most aggregates are not filtered properly and be deanonymized with eaze.

It's not that one is big bad, and one is little bad, it's the little bad can become big bad with a small amount of work by an attacker/company. Then when you add in zero external third party verification of these company claims, you really don't have any reason to believe them.


> When done properly is going a lot of heavy lifting there.

Not really. There are common practices for it. Yes it hits HN when deanonymization can happen at a well-known company, just like it hits HN when there's a security vulnerability that gets patched at a well-known company.

But "it's the little bad can become big bad" is what's doing the heavy lifting in your argument. No, that's not how it works. There's no universe in which aggregate data can be deanonymized to anywhere close to what all of the individual profiles would be. It's a completely false equivalenace, period.


Whether they know how to best exploit my commercial footprint via anonymous data or personal data, doesn't stop them from exploiting my commercial footprint.

Source: "trust us bro"

Why would you report honestly to your employer about metrics and methodologies that could potentially show you're not needed?

They are still needed though - the code isn't just prompted into thin air by nobody.

How many are needed though.

And for how long.


I wonder whether these documents will be retrieved by archaeologists of the future, trying to comprehend how it all began ..

I've just remembered a similar situation in 1993!

A Japanese factory that made epoxy resin for chips was destroyed and the price of SIMM chips skyrocketed (due to lack of availability).

I remember being very upset that I wasn't going to be able to upgrade to 4MB.


There should be a t-shirt for that ;-) I remember paying insane prices that year.


Ignorance is bliss.

If you're smart, you're taught you should expect more. You're also able to think critically and question.

For me contentment came a lot later; maybe the years have affected my brain plasticity and made me happier (dumber).


I think the only thing that would convince me, is the ability to self host or setup using infrastructure I own / have full control over.


It's more challenging to encourage correct implementation of semantics than implemention of visuals; which is a great reason for using the element that was designed for this use case.


I thought 1 is the country prefix for the USA?


Technically it's the country code for the North American Numbering Plan, which is used by several other countries as well as the US.

But in this context it'd be the first digit of the area code, with no country code being used because the call is within the US. There are no area codes in the the North American Numbering Plan that start with a 1.


Have none starting with 1 been assigned, or is there some technical reason there are none?


Per Wikipedia https://en.wikipedia.org/wiki/List_of_North_American_Numberi...

"The syntax rules for area codes do not permit the digits 0 and 1 in the leading position."

My guess would be it's to avoid ambiguity with the fact that 1 is also the country code. If I recall correctly, historically, dialing the 1 was necessary for any long distance call (even if not international).


>If I recall correctly, historically, dialing the 1 was necessary for any long distance call (even if not international).

You recall correctly. I haven't had a landline for a number of years now but I think it was still required latterly when I still had one. Don't think it was ever needed on cell (or maybe even valid) when I first got one at some point in the 1990s.


It used to be the case that the middle digit of an area code had to be a 0 or a 1. All the O.G. "cool" area codes like 212 are in this format, and the less desirable new area codes like 646 are not (yes, this is an accidental Seinfeld reference).


It is, although I'd imagine many Americans aren't aware of this.


+1 is also the prefix for Canada


Yes, there is a such a thing as a bad question after all.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: