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

Siri, What Time Is It in London?

daringfireball.net

Post reply on HN
Print view View original

741–750 of 776 posts

Re: Siri, What Time Is It in London?

#741

Post by cezart »

behnamoh wrote: post #683
metrokoi wrote: post #670

Earlier quoted context omitted.

I wonder which will come first: AI that is truly able to teach itself about our world, or that we are able to define algorithms or otherwise figure out software solutions to most of these these context problems. Of course, if an AI is intelligent enough that problem solves itself. To me it seems that developing AI on that level will be here sooner than developing solutions to many context problems, given the difficul…

That's exactly what gave rise to the first "AI Winter". People had to hard-code things. But with NN and esp. Deep Learning, they thought there's no need to write programs for pattern recognition anymore; the "AI" just learns it on its own. That wishful thinking has turned out to be dangerous tho, as we have moved towards an ML-dominated world where we don't even know how the ML algorithms produce specific results. Ad…

> We'd be basically back to square one.

But with better tools! It might not be going from 0 to 1, but going from 0.1 to 0.2 is still progress.

Re: Siri, What Time Is It in London?

#742

Post by nicbou »

Scoundreller wrote: post #365
nicbou wrote: post #174

Earlier quoted context omitted.

This is rather inconvenient when shopping for niche products. American website assumes that everyone on the internet is American, and you have to go through the shopping cart experience to learn that they don't ship to Canada or Europe.

And too many app developers assume that all their users have massive/unlimited data plans. This is usually correct for... most developed and developing countries, but: They forget that in the telecom-backwater of Canada, a 4 or 5gb/month plan is near the high end. I’m looking at you NPR One app for thinking buffering 300-400mb of content is a good idea.

Even fast devices have slow internet in the underground. So many websites load 3mb of data for a few kb of relevant info.

Re: Siri, What Time Is It in London?

#743

Post by cmckn »

rm445 wrote: post #653
kilo_bravo_3 wrote: post #523

Siri doesn't know that my front door is called "FRONT DOOR". I only have one smart lock, which works perfectly, and it is called "FRONT DOOR" in HomeKit. When I ask Siri about my FRONT DOOR she responds that she cannot find it. When I ask Siri about the status of my DOOR, she responds with "The FRONT DOOR is locked/unlocked". I'll then say 'Alright Siri you literally just used the phrase "FRONT DOOR" five seconds ago…

All these speech assistants are almost stateless, aren't they? I mean, they can ask you questions and go into a listen-for-an-answer state, but, at least last time I tried, you can't have a conversation about your conversation with them and improve their comprehension of anything. It's like talking to a command line, or those old Palm Pilots you had to learn a special alphabet to scribble on. It's quite interesting t…

Honestly, I can forgive the lack of context awareness. That's a hard problem. I have issues even getting consistent responses to the same query over time (even back to back in some cases). Sometimes, Siri will misunderstand me and fail to do the thing, but then I look at the text that's transcribed...and it's correct (I.e. the backend was replying to a different transcription than I saw on the frontend).

I've just been trained to not bother. Unless I'm setting a timer, I just don't try anymore.

Re: Siri, What Time Is It in London?

#744

Post by saagarjha »

kempbellt wrote: post #668
saagarjha wrote: post #622

Earlier quoted context omitted.

> If you saw someone write something like "If you had a human assistant and asked them 'What’s the time in London?' and they honestly thought the best way to answer that question was to give you the time for the nearest London, which happened to be in Ontario or Kentucky, you’d fire that assistant." you'd be quick to call them out as an "obvious" bad employer. No, really, this is just a tired nitpick. I hear a lot of…

I think the point translates just fine. What I read was, "Siri doesn't give me exactly what I want by perfectly understanding my own personal bias that London means, London, England, so I am going to post a blog complaining about it, while simultaneously cracking a joke that I would terminate someone's employment if they made a basic mistake." Which I took to be pretty unproductive conversation. Still, point made. OP…

> Siri doesn't give me exactly what I want by perfectly understanding my own personal bias that London means, London, England, so I am going to post a blog complaining about it, while simultaneously cracking a joke that I would terminate someone's employment if they made a basic mistake.

At some point, an assumption that an extremely large majority of people would reasonably make (I am hesitant to claim the number of "nines", but would guess it to be at least three) stops becoming a "personal bias" and a and more of "unless you were raised by wolves or some sort of homeschool for math savants you really ought to know what I mean". If I'm talking about Springfield and you take it to mean I am discussing the Simpsons and not one of the dozen Springfields that exist in the United States, sure, that is an honest, "basic mistake" (although I would prefer if you clarified the moment you guessed that something might be wrong with your interpretation). Thinking that I really meant the tiny town of London, Ohio that I have never mentioned and more than likely probably don't even know exists when I say "London" instead of the massive metropolis that is the capital of a major world country is just being stupid: something is very off with your sense of context. (To finish the argument: if I said "London" and you assumed "London, England" but I actually did mean "London, Ohio", I would be extremely forgiving, to the point where I would lay the fault on myself for not clarifying when I gave such a misleading view.)

> OP would be terrible to work

Because they say they're not interested in employing people who cannot understand more context than a reasonable human being needs to quite literally live and interact with others?

> Siri could potentially use some improvements in accuracy

Siri could absolutely use some improvements in accuracy.

Re: Siri, What Time Is It in London?

#746

Post by Joeri »

rm445 wrote: post #653
kilo_bravo_3 wrote: post #523

Siri doesn't know that my front door is called "FRONT DOOR". I only have one smart lock, which works perfectly, and it is called "FRONT DOOR" in HomeKit. When I ask Siri about my FRONT DOOR she responds that she cannot find it. When I ask Siri about the status of my DOOR, she responds with "The FRONT DOOR is locked/unlocked". I'll then say 'Alright Siri you literally just used the phrase "FRONT DOOR" five seconds ago…

All these speech assistants are almost stateless, aren't they? I mean, they can ask you questions and go into a listen-for-an-answer state, but, at least last time I tried, you can't have a conversation about your conversation with them and improve their comprehension of anything. It's like talking to a command line, or those old Palm Pilots you had to learn a special alphabet to scribble on. It's quite interesting t…

When i made a meeting room reservation chatbot the hardest part was managing a conversational context without the user getting stuck.

In the end I had a “draft booking” and a conversation loop, where the bot would repeatedly ask to fill in missing parts (eg nr of participants) and then give you a summary and opportunity to correct things. It was hard to do, and definitely required a lot of contextual understanding of how people book meeting rooms. That approach doesn’t scale up well.

I think the basic problem is being stuck in a local optimum. The scripted bot approach doesn’t scale to complex conversations, and you need to start from scratch to do better.

Re: Siri, What Time Is It in London?

#748

Post by gambiting »

hanche wrote: post #615
gmac wrote: post #416

Earlier quoted context omitted.

Same in the UK, and similarly they often cease at 11:59PM, which has always struck me as pretty unsatisfactory. What if a plane crashes on my house, between policies, during those two uninsured minutes? Sure, it's unlikely, but so are all the other things that might cause me to claim on my buildings insurance policy.

So make insurance policies overlap by 23 hours and 58 minutes? And leave it to the insurance companies to figure out which one gets to pay out if bad things happen during the overlap.

Most policies I've seen have a clause that says that if you are double insured for the same event then they won't pay out - which in effect could leave you with neither company paying anything. I have no idea how a court would look at such a clause, but it definitely exists.

Re: Siri, What Time Is It in London?

#749

Post by cannam »

smabie wrote: post #647
cannam wrote: post #596

Earlier quoted context omitted.

Reminds me of when I worked at Yahoo, for not very long in the early 2000s. Yahoo had a consumer product called My Yahoo, a customisable home page with various feeds in it, and at the time they also had a My Yahoo Enterprise Edition, which they tried to sell to companies to serve as an intranet home page. I worked (in a junior sort of way) on the European deployments of MYEE. Anyway, one of our customers - representi…

If this had happened at some of the other places I have worked, the CEO would have made all the developers stay in the office until the problem was fixed. Why did no one care?

To say no one cared is going too far, but it is interesting to think about why something like this was able to happen.

There were various technical and structural factors making it difficult to fix. Weather feeds were known to be problematic (still are I guess) and this code would have been surprisingly low-level C/C++ with custom serialisations and limited logging. Structurally there were problems in getting attention from a team in California to support a problem experienced by a different team in London, especially since it affected relatively few users - a tension in supporting paid products in a company that is focused on non-paying users at far greater scale.

(I am assuming this bug would have needed some actual development work - I don't recall, but I think we were familiar enough with common ops problems that it wasn't just a question of kicking one of our own feed servers.)

But I do think there was an issue about lack of concern - at heart we didn't have enough confidence in our own product to motivate the personal pain of working through these problems and getting them solved. I think that, if you had gathered us together and asked our collective opinion, we would have suggested that this customer would be better off not using our product at all - it simply wasn't ever likely to be good enough. Once you reach that way of thinking about your own product, it becomes extremely hard to countenance fixing the most difficult problems with it.

Post reply on HN