Hacker News Reader

Comment by crowbahr | Hacker News Reader

But none work together so seamlessly.

I'm with that guy: the ease of life is amazing with Google. I'll buy movie tickets online then get an alert about driving conditions and when I need to leave to arrive on time. That's something your open source maps program can't do because it doesn't have access to my email.

And if it did I wouldn't use it because it'd be sitting there combing my email constantly which isn't efficient to do on my phone.

Google's ability to personalize data is without compare honestly. It comes from being huge... And I'm OK with it.

View on HN

Open source could work together flawlessly like that, but it'd need open, standard protocols to do so - and an adherance to them.

A closed system just makes it easier for someone to do - because they don't have to care about any standard protocols and can hack together whatever they want and everyone will just deal with it.

I'm not really arguing against you - but I think it's important to note that what we have is possible, just not _right now_.

Of course, _right now_ is when we all want it.

It has always bugged me how desktop Linux has never developed a "share" feature akin to what Android has - URIs and URLs that get indexed against a database of services to handle them, then presented to the user as consumers of the data.

There have been a lot of attempts at dbus-based services to advertise or invoke other processes for various tasks but it never became as pervasive as share buttons and never had the UX.

Reminds me a lot of some attempts by google to create a JavaScript API for that sort of thing.

[1]: https://www.w3.org/TR/web-intents/ [2]: https://github.com/WICG/web-share

But you need to consider the amount of data Google is amassing from people using solely their services. And then sells that data to advertisers. If this trend continues eventually they'd be able to know more about you than even the closest of your friends.

Is this really such a great price to pay for your privacy? For it to be easier to look up your friend's phone number from your email list?

I'm just curious, what data do you believe google gives to advertisers? I work at google, and I see this trope on HN a lot, and I always like to see what data people think is given away, since a lot of the time they seem to be under the impression that either Google's privacy policy is different than it actually is, or that Google's privacy policy isn't actually something that Google follows, or that it doesn't protect them from something for some reason. So, what data do you think google gives to advertisers/third parties, and what do you think Google does with the data it does have about you? (and by you I mean the generic you if you personally stay away from Google, and by Google I mean Google, but fwiw, you could probably replace Google with Facebook or Microsoft and ask the same exact question)

Alright, let's say that I have a general mistrust of corporations just because from some modest reading of history I know that if someone has power they'll abuse it for their own interest. So my opinions about Google and others are admittedly influenced by that. It has happened before - all it takes is for someone who stands to profit from a law to make it happen, and then suddenly it's the status quo and everyone's fine about it, not even thinking about the alternatives.

But since you said that what applies to Facebook could also apply to Google here are some examples - http://www.salimvirani.com/facebook/.

Finally, I'll admit that I have no concrete evidence that Google mishandles users' privacy. I will research that in the near future and perhaps my opinion will change accordingly. However, just the fact that they have access to all that data is insane. Potentially, Google and other advertising companies can know more about you than your closest friends and family. If that would be true wouldn't that bother you? Even if they don't abuse the data now who says they can't do it in the future if that would make them more profit and they could get away with it?

So, I asked you for examples of Google (or Facebook) sharing your information with third parties. While I agree that Facebook's practice of advertising via "Josh liked XYZ product" is not something I want, it is not an example of facebook sharing your data with advertisers. That happens entirely within the facebook platform.

I think its a misleading way to present the information, but its not sharing any information (your likes) with anyone who didn't already have access to that information (your friends).

Even if it happens within a platform it's still a very concerning problem.

If a company decided to put cameras in people's homes to improve "customer service" that business would be out of business the next day. And yet companies are able to get much more information about us by having access to what we do on the internet and everyone is fine about it.

I understand that this is a result of people not wanting to pay for access to websites and online-services but, in my opinion, that's because they don't understand the ramifications. I don't have a solution for this problem, but I hope we'll fine another way, one that doesn't involve companies knowing so much about you.

>Even if it happens within a platform it's still a very concerning problem.

I don't necessarily disagree, but "Xcorp is presenting personal data that you've released publically in a misleading way" very different than "Xcorp is giving personal data that you assumed was private away to third parties". I think the second is a lot more serious than the first.

>If a company decided to put cameras in people's homes to improve "customer service" that business would be out of business the next day.

I'm not sure I agree, given the popularity of smart TVs and the recently announced Amazon Echo Look. I make no judgement of whether this is good or bad (I personally don't have any always-on recording devices), but I don't think its true.

>I understand that this is a result of people not wanting to pay for access to websites and online-services but, in my opinion, that's because they don't understand the ramifications.

See, and I disagree here. There are certain kinds of services that money cannot provide. Google Now recommends news articles to me, it can give me reminders about upcoming events from my emails and Google gives me some personalized search features (ie. if I type "frequent flyer number" into the default google search bar, my number appears, scraped from an email years ago). Those features have a utility that I couldn't get from an application that didn't track information about me. Potentially I could self host something with that level of utility, but it would take a research lab and millions of dollars to create.

On the other hand, reddit, which tracks comparatively little information about me, reliably recommends trending posts that are of zero interest to me.

>Google Now recommends news articles to me, it can give me reminders about upcoming events from my emails and Google gives me some personalized search features (ie. if I type "frequent flyer number" into the default google search bar, my number appears, scraped from an email years ago). Those features have a utility that I couldn't get from an application that didn't track information about me.

And I think that's the main difference between our viewpoints. Personally, I'm not ok with a company having so much information about me just for the sake of convenience. I'd rather have to write my own scripts to scrape through news feeds or just continually check blogs that I like rather than a company having all that data about me.

We are living at a time, imo, when data is the most important asset. And companies like Google are one of the biggest purveyors of it. It's dangerous when a company could have enough information about you and have the tools to know exactly what makes you "tick". They could provide services to make "specially crafted" advertisements that would manipulate you at a scale even more than they do now. "You like tall blondes? Here's a computer generated tall blonde that talks about the new eau d'toilette and how it seduces her! Buy it!"

It's bad enough that companies are using methods to addict their users to make their apps more profitable now, imagine what a world it'll be when they'll have the technology and ability to do it to the point that no one could possibly refuse their advertising.

Now I realise that this is taking it to the extreme, but my point here is precisely that we need to consider these possibilities in case they don't catch us unaware and when it's already too late to do anything.

>And I think that's the main difference between our viewpoints. Personally, I'm not ok with a company having so much information about me just for the sake of convenience. I'd rather have to write my own scripts to scrape through news feeds or just continually check blogs that I like rather than a company having all that data about me.

This is a fair view to hold (I think the rest of your post goes off into slippery slope land, but its based on a realistic fear), but at the same time, I think you need to be aware that you (or I) won't be able to recreate some of the things that these companies can do, no matter how much time or effort we invest.

Its also worth realizing that you're essentially stating that these experiences should be inaccessible to those who are nonexperts. That is, if in your perfect world, MiGooBookZon doesn't exist, and you make due with a set of scripts that you wrote that scrape some RSS feeds you care about and order articles based on some ratings you hardcoded for personal use, where does that leave all of the people who don't have the technical chops to write a set of personal scripts to scrape the news sources that interest them?

Ideally, they should be packaged in an accessible format. Sort of like how in the Linux world Ubuntu is doing pretty well with making Linux accessible even to people who are not technology-inclined.

The problem in "my perfect world" would be not that people with less technical ability won't have access to the tools, but that the tools would be less efficient(but as I've stated in another comment, a program being (F)OSS doesn't necessarily mean it's worse, sometimes quite the opposite), due to not having all that integration.

And I consider it to be a good deal. If some people don't then I'd like for it be an option for both parties to make their decisions based on pros and cons, however, the issue with monopolies/oligopolies is that too often they don't leave you a choice. They'd buy up all the small companies that have good ideas, integrate them into themselves and so on, until there's realistically no other option but to actually start using them.

I agree with you: There is a misunderstanding here.

Google obviously gives no data to advertisers... they sell targeted ads. The difference is vast. As an advertiser I would purchase an ad targeted at 20-something male video game players, for example. I would not get to read their emails to get better insight into their minds or anything insane like that.

And if that started happening at Google they'd die, and they know it.

Aggregated?

They can cross reference one's data across all their services.

So, let's say your Android phone let's Google know where you live, work and hang around by checking maps history and gps records, and they accumulate all the search words and they know who you are, really and check every email for finding who is related and they can really come up with a model of what kind of person you are at any specific moment.

Group together, see what the citizens are like in a certain area, I think that alone is already valuable for targeting.

And they surely randomly get hidden gems where it can't possibly go public but use for their own merit.

But that's not what I asked. Let's assume for a moment that you trust google completely, and believe that they won't do anything untoward with your data on their platform. But we all know that google gives some data to advertisers. I'm asking what you think that data is, hence I asked

>I'm just curious, what data do you believe google gives to advertisers?

It sounds like you're saying that Google has a lot of information on you, which is probably true, but how much of that gets given to third parties? (and if you're claiming that the information you described is given to third parties, I'm confident that you're mistaken).

> a lot of the time they seem to be under the impression that either Google's privacy policy is different than it actually is, or that Google's privacy policy isn't actually something that Google follows, or that it doesn't protect them from something for some reason.

Sorry, last year we've seen how several large car makers have had systematically undermined environmental regulations for years. That does not exactly inspire a lot of faith regarding self-policed, "we'll promise really really hard" privacy policies to me.

So just to clarify, because a different company, in a different industry, did something illegal that had nothing to do with privacy, you don't trust privacy policies?

Yes, pretty much this.

That scandal has shown that self-policing doesn't work if all the incentives point into the opposite direction. I don't see what significant differences at Google would make them immune to the forces that were at work at VW.

I think there's a very simple one: google employees, by and large, use google products. If google is mishandling user data, that means that they are mishandling my data, and my boss's data, and my boss's boss's data.

That is, unlike with VW vehicles, individual employees are directly negatively affected.

Given that Google can revise their privacy policy unilaterally at any time (and, as far as I know, they are under no legal obligation to delete or restrict access to data collected under previous iterations of the policy) I don't see why any of us should feel comforted by the existence of some mutable words on a webpage somewhere.

>as far as I know, they are under no legal obligation to delete or restrict access to data collected under previous iterations of the policy

(obviously IANAL) I'm fairly sure that they cannot without your permission, although they can assume your permission if you continue using the service after they change the policy (and you're informed of the change), and at least with the changes I can find online, Google at least generally seems to give users the chance to opt out of changes.

Edit: a quick search for examples backs up my claim, I can't find any situation where a company could modify its privacy policy in a way that would affect existing user data, and apply that new policy to users who were no longer using the platform. Either its not legal, or companies voluntarily don't do that.

I trust Google for privacy a lot more than I'd trust 10 different little specialized services that are fighting for their lives financially and don't have the decade of experience as the biggest target on the planet with no bad breaches.

And that's a big part of the problem: People don't realize privacy is essentially about fragmenting your information. 10 different little specialized services are far more private, even if they're subject to more breaches.

Thinks like your coworkers not knowing every last detail of your family, nobody you know IRL knowing about your weird online kinks, etc.

Your privacy is largely reliant on your different social circles being separated, your various identities being apart, and Google has made it so they can link all of your data together.

(EDIT @joshuamorton: Security and privacy are actually distinctly different concepts with very different implications. A former Googler told me this is not well understood inside the company.)

Why edit your post instead of responding? That's such an odd way to do things and breaks the order of conversation.

>Security and privacy are actually distinctly different concepts with very different implications.

True, but you cannot have privacy without security. That is to say, if I pieces my personal information to 5 different groups, but all of them are vulnerable, your data, all of it, is vulnerable, and an untrusted Nth party can attain all of it. On the other hand, if you give all of your personal information to a single secure party, you know who has the data.

Obviously, its a bit more complex than this because often times larger groups are single sources of failures, but still.

For a nondigital metaphor, your list of kinks is less likely to get out if you put it in a safe deposit box than if you rip it into 4 pieces and tell your closest friends to put their piece under their mattress, even if no friend as the whole list and you trust each friend more than your bank, because an adversary can probably figure out that you gave it to your friends, and can break into their houses, but won't be able to crack the vault.

I edited the post because I hit a rate limit on HN. I don't like it either, but I have to choose my use of replies carefully.

The problem you miss is that people have different identities. And people want to interact with others, to communicate with people, through those identities. Locking up your kinks in a vault isn't really an option if you want to share them with a community of like-minded individuals. But just because you do, you don't necessarily want your family to be able to look them up on Google.

People have different even potentially public personas, that they still want to keep separate. Under a variety of names, I'm involved in a variety of different communities, some of which I don't really want people I know IRL to know about, some which I don't mind. But all of them are out there on the Internet, because that's how I interact with them.

Actually, apart from the Google-centric notion of it and the lack of multiple identities (exacerbated by Google's "real name policy", Google+ was probably the closest Google got to getting this right: I could put specify which Circles people belonged in, and share information with each independently.

This is by definition security by obscurity, is it not?

You mean Google would give up digging deep into their gold mine data just because they have another steady revenue stream? Hell no. They know about you better than you do. You don't remember what you were interested in on each month, Google does.

They may not sell all of those to third parties but what does it matter if they know all of it.

What is a privacy anyway after you've already revealed who you are constantly on multiple Google products.

Data selling from big tech companies is the least of your worries tbh. It's all presented in aggregate form and protected very well. It isn't even really data selling, although if you want to argue semantics and model it as such, it would be data selling limited in scope in that the only action an advertiser can use the data for is presenting you with an ad.

It's the financial industry that you have to worry about. The finance industry has atrocious security and access to very personal information on you. Finance companies (your bank, credit card, etc.) sell your information to the highest bargainer with much higher privileges (physical mail and junk emails) and don't allow you to opt out.

Edit: And also the health care industry. Super personal information with archaic tech and paper thin security. Data might not be sold but the risk of bad parties gaining access to your information is high.

...until they lock your account without giving you a reason and without recourse.

This is why AmaGooFace is so dangerous, because the way things are going they will have unprecedented insight and control into global populations. Multinational corporations are a much more dangerous threat to humanity's future freedom than governments.

Reply on news.ycombinator.com