Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Google in, Google out

techcrunch.com

Post reply on HN
Print view View original

111–120 of 135 posts

Re: Google in, Google out

#111

Post by TeMPOraL »

jaflo wrote: post #18
paradite wrote: post #14

On the topic of Google Lens, I don't really understand the hype around it. People are talking about how it is able scan a bar code of SSID and password to connect to router. Isn't that ancient technology? And for other uses like overlay maps info, aren't those just running standard CV/ML algorithms on their mega-servers plus simple integration with other apps in their ecosystem?

I think it is that simple integration that makes or breaks the product. Because Google controls the OS, it allows for a more seamless experience: you wouldn't download an app just to scan and connect to Wi-Fi networks, but because it's a feature built into your phone you are much more likely to use it.

If only they did that. One of the things that annoys me the most about modern smartphones is that every single feature has a separated, isolated app for it (usually each from a different vendor), and none of those apps properly integrate with one another. The more they can integrate at OS level, the better for the user experience (though I'm not having high hopes - Google has a long history of dumbing down applications to the point of making them barely useful for anything).

Re: Google in, Google out

#112

Post by ocdtrekkie »

TeMPOraL wrote: post #109
ocdtrekkie wrote: post #99

Earlier quoted context omitted.

Google. "When Voice & Audio Activity is off, voice inputs won't be saved to your Google Account, even if you're signed in. Instead, they may only be saved using anonymous identifiers." https://support.google.com/websearch/answer/6030020?p=accoun... Ergo, Google is going to collect your voice data no matter what you do, the only question is whether or not it's connected to your account, where you can manually go in an…

So the question now is, if you opt in and later delete your data, does the data actually get deleted, or just disconnected from your account and left only with "anonymous identifiers"?

That is the big question.

I actually had this argument out with a former Googler who expressed with incredible strength how importantly Google addressed deletion of data users said to delete. And so his argument was that if you deleted it from your history, it'd be deleted, but if you turned off your history, you'd have no way of telling them to delete it.

Re: Google in, Google out

#113

Post by fixermark »

marssaxman wrote: post #106
euyyn wrote: post #88

Earlier quoted context omitted.

Good luck asking Wikipedia what's the weather today, what's your first meeting in the morning, and which route of your commute has less traffic at the moment.

I don't ask for those things anyway, precisely because I don't want Google to have that much information about me. I'd love to have a local-only, offline map system.

You can pretty much get that right now by buying any one of the fine Garmin navigational products on the market, right?

Re: Google in, Google out

#114

Post by fixermark »

TeMPOraL wrote: post #107
fixermark wrote: post #79

Earlier quoted context omitted.

With the inception of digital cable boxes, many cable providers no longer need the Nielsen data because the cable box tells the provider what channel it's tuned to (a technical necessity if the box is using switched digital video). It's much easier (and safer for the end user) to passively collect usage data via the application you provide than to trust some third-party to (a) collect a statistical sample unbiasedly…

> The rightness question seems to me to look a lot more like "If you can collect data from a billion users without doing any harm to the users, and that data is going to be more useful than a statistical subset of that data, why should you not collect it?" In this particular case (speech recognition), one harm done to the users is tying the product to the Internet, and thus requiring what should be closed-loop tasks…

"Should be closed-loop" is an interesting assumption that I'm not convinced aligns with the reality of the technology. To what extent has speech recognition been improved by being able to feed it through a constantly-updated set of open-loop ML infrastructure?

Re: Google in, Google out

#115

Post by TeMPOraL »

fixermark wrote: post #114
TeMPOraL wrote: post #107

Earlier quoted context omitted.

> The rightness question seems to me to look a lot more like "If you can collect data from a billion users without doing any harm to the users, and that data is going to be more useful than a statistical subset of that data, why should you not collect it?" In this particular case (speech recognition), one harm done to the users is tying the product to the Internet, and thus requiring what should be closed-loop tasks…

"Should be closed-loop" is an interesting assumption that I'm not convinced aligns with the reality of the technology. To what extent has speech recognition been improved by being able to feed it through a constantly-updated set of open-loop ML infrastructure?

Probably a lot; open-ended voice recognition with no prior training is hard (though I'm not convinced it can't be made to work off-line now that the models are there). Still, a lot of devices don't need open-ended voice recognition (structured grammars greatly simplify the problem), and if you allow for users having to pre-train their devices for few minutes, off-line processing becomes easier.

My impression is that the main driver behind cloud-first voice processing is business, not technology.

Re: Google in, Google out

#116

Post by JoshMnem »

fixermark wrote: post #66
JoshMnem wrote: post #13

Earlier quoted context omitted.

Consumers tolerate it partially because they don't understand technology. They would not tolerate it if they really knew how it worked and had knowledge on how to fix the problem.

I think you over-estimate the public's sensitivity to the risk model. I know plenty of people (myself included) who know how it works and cheerfully embrace it, because the benefits outweigh the risks.

I know many technical people who understand the technology side of things, but they don't fully understand the political and societal implications.

That may not be true in your case, and you may have some new arguments that I haven't heard, but I've never found a convincing argument for being careless about privacy, and I've been following and debating the issue fairly closely for about 15 years.

It's a very complex and difficult problem. The answer isn't "we absolutely must have 100% privacy" or "they already have all of my data so it doesn't matter." But arguments along the lines of, "I don't mind if they have my data if it helps me find products that I like", tend to arise from "consumer" mentality rather than "architect" mentality. So I would revise my above comment to say, "Consumers tolerate it partially because they don't understand [the full implications]."

Re: Google in, Google out

#117

Post by JoshMnem »

wordupmaking wrote: post #38
JoshMnem wrote: post #13

Earlier quoted context omitted.

Consumers tolerate it partially because they don't understand technology. They would not tolerate it if they really knew how it worked and had knowledge on how to fix the problem.

And then there's people who do understand but have friends who don't. As everybody should. I can self-host my email all I want, if I communicate with anyone in a corporate silo, the silo records it.

Yes, and there are many dark patterns that companies use to get your data from your contacts, even if you don't use those companies' products.

Re: Google in, Google out

#118

Post by euyyn »

TeMPOraL wrote: post #108
euyyn wrote: post #88

Earlier quoted context omitted.

Good luck asking Wikipedia what's the weather today, what's your first meeting in the morning, and which route of your commute has less traffic at the moment.

Asking for weather is an API issue, it totally does not require uploading your voice.

It requires uploading where you are. And it requires uploading the voice for processing if you want it to distinguish "what's the weather today" from "tonight", "tomorrow", "this weekend at ", or "back at home".

Re: Google in, Google out

#119

Post by euyyn »

marssaxman wrote: post #106
euyyn wrote: post #88

Earlier quoted context omitted.

Good luck asking Wikipedia what's the weather today, what's your first meeting in the morning, and which route of your commute has less traffic at the moment.

I don't ask for those things anyway, precisely because I don't want Google to have that much information about me. I'd love to have a local-only, offline map system.

Well, that makes it not a "Star Trek-esque voice-activated computer interaction", which is the point of it.

Re: Google in, Google out

#120

Post by euyyn »

TeMPOraL wrote: post #105
fixermark wrote: post #74

Earlier quoted context omitted.

It is a technological issue. Without the real-world audio samples to use as training data, the system cannot be improved enough to work reliably.

I built an off-line voice recognition system 10 years ago using Microsoft Speech API and I guarantee you it worked quite well. There's nothing in base voice recognition that would require uploading everything to the cloud. You only have to be willing to read out some text for few minutes, once, to train the model.

> You only have to be willing to read out some text for few minutes, once, to train the model.

I guess that's what kills it as practical in the real world.

Post reply on HN