Computer, Respond to This Email • Hacker News Board

Re: Computer, Respond to This Email

#21

Post by drzaiusapelord » Tue, Nov 03, 2015, 6:09 PM UTC

> But replying to email on mobile is a real pain, even for short replies.

Imo, this has already been solved. I just whisper replies into my watch for both email and SMS. Google's voice recognition is finally good enough for this. I can't think of anything more convenient.

I'm not sure if AI-ish replies are something I'm interested in. Throwing complex fuzzy logic at what should be a hardware/interface problem seems shortsighted. I prefer a much higher level of granular control, especially if these messages are work related.

Re: Computer, Respond to This Email

#22

Post by ewzimm » Tue, Nov 03, 2015, 6:10 PM UTC

x5n1 wrote: post #18

alphydan wrote: post #9

> Another bizarre feature of our early prototype was its propensity to respond with “I love you” to seemingly anything Is it trained mostly on personal gmail accounts?

I love you! We should go out some time.

Would like to see the Maeby Funke addon where "Marry Me!" is always the first response.

Re: Computer, Respond to This Email

#23

Post by hellbanner » Tue, Nov 03, 2015, 6:20 PM UTC

icey wrote: post #2

I just started reading Avogadro Corp ( http://www.amazon.com/Avogadro-Corp-Singularity-Closer-Appea... ) this weekend, and this reminds me quite a lot of the emergent AI that figures heavily in the story (ELOPe). A quick synopsis: developers build a system to "improve" responses from emails. The system at some point is given the ability to send emails on its own, and a poorly issued directive. It's been an engaging r…

Is Amazon down for anyone else? Error loading products

Re: Computer, Respond to This Email

#24

Post by sunnyps » Tue, Nov 03, 2015, 6:21 PM UTC

zitterbewegung wrote: post #20

dikaiosune wrote: post #4

While the technology is very cool, this pushes me that much closer to a "dumber" email service. It's always a conflict for me as I love shiny new things, but I'd rather we let AI loose on someone else's email, especially when the AI's revenue stream is advertising (a business predicated on knowing as much as possible about your audience).

You can make gmail "dumber" by using the IMAP api and then encrypt your email? Unless you use google apps for work since you aren't paying for the service you are the product.

Stop with this "you're the product" nonsense. Please.

Re: Computer, Respond to This Email

#25

Post by dikaiosune » Tue, Nov 03, 2015, 6:23 PM UTC

bobmichael wrote: post #14

dikaiosune wrote: post #4

While the technology is very cool, this pushes me that much closer to a "dumber" email service. It's always a conflict for me as I love shiny new things, but I'd rather we let AI loose on someone else's email, especially when the AI's revenue stream is advertising (a business predicated on knowing as much as possible about your audience).

I don't think I understand your reasoning. You're already tolerating an AI processing your email to produce the ads, so assuming that, why does a cool new feature push you closer towards a "dumber" service?

Because I'm lazy and switching would be time consuming. Each new privacy-related "innovation" motivates me a little more. You're right that the practical effect of this new service is negligible on a technical level, but it still reminds me that I should find an alternative I find acceptable.

Re: Computer, Respond to This Email

#26

Post by daveguy » Tue, Nov 03, 2015, 6:24 PM UTC

sunnyps wrote: post #24

zitterbewegung wrote: post #20

Earlier quoted context omitted.

You can make gmail "dumber" by using the IMAP api and then encrypt your email? Unless you use google apps for work since you aren't paying for the service you are the product.

Stop with this "you're the product" nonsense. Please.

Stop making the user the product. Please.

Re: Computer, Respond to This Email

#27

Post by dikaiosune » Tue, Nov 03, 2015, 6:25 PM UTC

zitterbewegung wrote: post #20

dikaiosune wrote: post #4

While the technology is very cool, this pushes me that much closer to a "dumber" email service. It's always a conflict for me as I love shiny new things, but I'd rather we let AI loose on someone else's email, especially when the AI's revenue stream is advertising (a business predicated on knowing as much as possible about your audience).

You can make gmail "dumber" by using the IMAP api and then encrypt your email? Unless you use google apps for work since you aren't paying for the service you are the product.

Using IMAP + encryption obviates all of the features that keep me on gmail, and would also be time consuming (especially with my non-technical contacts). On a practical level, I think your suggestion is equivalent to using a different email service -- high effort of implementation, loss of gmail's "smart" features, etc. Whether I'm using Google's servers or not, hiding my stuff from their systems amounts to the same effects on my life.

Re: Computer, Respond to This Email

#28

Post by willhinsa » Tue, Nov 03, 2015, 6:26 PM UTC

pjmlp wrote: post #17

Waiting for the news about social and legal problems that will happen from typing on the wrong answer.

You can anyways undo the sent message! A great feature of Inbox.

Re: Computer, Respond to This Email

#29

Post by imh » Tue, Nov 03, 2015, 6:31 PM UTC

I appreciate the privacy standards they used (no humans reading your email to develop this), but am concerned that it's not enough. As I understand with language models, overfitting takes the form of returning a sequence of words seen in the training set. If this is overfitting in any part of the response space, this could happen. Out of a million emails, how many suggested responses are going to substantively resemble another response the original author wouldn't want read by others?

Re: Computer, Respond to This Email

#30

Post by douche » Tue, Nov 03, 2015, 6:35 PM UTC

hellbanner wrote: post #23

icey wrote: post #2

I just started reading Avogadro Corp ( http://www.amazon.com/Avogadro-Corp-Singularity-Closer-Appea... ) this weekend, and this reminds me quite a lot of the emergent AI that figures heavily in the story (ELOPe). A quick synopsis: developers build a system to "improve" responses from emails. The system at some point is given the ability to send emails on its own, and a poorly issued directive. It's been an engaging r…

Is Amazon down for anyone else? Error loading products

That link seems to be broken, as well as most of the product links that Amazon surfaces on the error page. Searching for the title seems to work fine.