No wonder Siri is having trouble connecting to the network now with all those requests
Me: "Talk Dirty to me"
Siri: "I can't. I'm as clean as the driven snow."
Thanks for trying that mike. If someone complains something does not work, that is what we say first, 'Clear your cache'.. I thought Apple may have put in an easter egg for that.
btw, your complaints about this Siri not being that much more advanced than the Siri app is valid. It is better integrated with the phone itself with access to more apps on the phone. With Apple's resources and massive input data they are going to get, it WILL get better.
Your geofence problem, is that a SIRI problem or iOS 5 problem? Did you check if SIRI set the reminder properly? ( if there is a way to check.. I have not tried geofence based reminders in iOS5 on iPhone 4 yet )
needa said:i am kind of disappointed in siri. she tells you you dont have an address for your contact and then will not put the address in. for me, the reminders is going to be the most useful thing about her and she goes about it in a half ass way. apple should have known that people dont put addresses in for contacts. they should have set her up to put the addresses in for you. either through web search or voice.
Keeping it all server side is a great move by apple,
as it means they can:
- Track queries and responses in a database and refine/correct them for everyone.
- Get good (marketing) data on what people are actually using it for (joking around at the moment it seems!)
- Keep responses up to date - and fresh.
Voice control on ios4 was nearly useless as there was no confirm stage:
"Call Home"
screen: calling Jack Stone
..and its already dialling, leading to me turning it off.
Siri does seem like a bit of a fake which is disappointing.
It's like you giving a job to someone who knows nothing about the subject, just looking up questions that are pre-printed on a sheet of paper and blindly reading back the answers.
Actually, I don't think that most people do realise quite how much is going on, i.e. just how difficult it is to do things that seem so trivial to humans because we do it so effortlessly. Having said that, I do worry that once the initial flurry of enthusiasm passes then people will start discovering the limitations and become frustrated. Siri's survival might become an arms race between the developers adding capabilities and the ever increasing user expectations.The type of thing that appears cool for the 1st few days, but when you want to do something you realise what's going on.
The feedback I've seen is that Siri often has difficulties in noisy environments so I think that the answer to your question is probably no. It might be an obvious requirement but that doesn't mean that it's simple to do.... does Siri recognise you the owner of the phone and ignore other voices? If someone else shouts over your shoulder will Siri respond to that, or ignore it, as it should as it recognises you, the phone owners voice?
That's another obvious requirement.
If any of you remember Eliza (example: http://www.manifestation.com/neurotoys/eliza.php3) , the AI program from the 70's then I don't think you'll be terribly impressed with Siri. I played with Siri for a half an hour and there is some depth to it's programming, but fundamentally, if you ask something out of pattern it defaults to searching the web.
I do see it having some utility while wearing a headset and driving, for example to compose a text message but I don't see it as a revolutionary improvement in AI. It's not a very intelligent system unfortunately. I wonder how much Apple paid for this tech?
not sold... if all it does is parse the input for phrases like "what's the weather", "I'm hungry", "call John" etc. and sends the rest to Wolfram Alpha then consider me unimpressed... I am pretty sure the novelty will wear off quite quickly. That being said, it makes for great marketing
4 minutes to respond to me
Thanks for the interesting reply.
My worry is that Apple might be raising people's, who don't understand, expectations that it's something that it's not.
Oh, it's more than having been around as long as many of you. The reason I did a comp sci degree in the first place was to get into AI, inspired as a child by things like HAL in 2001 and computers in Startrek. I got what I wanted and spent about the first 6 years working in AI research, and in particular I worked in the field of natural language processing/understanding. In fact in 1984/1985 I was working on a project that, when considered in terms of the technology that was available at that time, was a very similar project to Siri.If you've been around as long as many of us have, we know they've been working on AI for decades and are still miles away from it in reality.
If you've been around as long as many of us have, we know they've been working on AI for decades and are still miles away from it in reality.
Personally I would not expect Siri on a 800Mhz mobile phone to beat the words best super computers at AI.
Thanks for the link and you're right I did not finish the video before replying... but really, the "proof is in the pudding" and the results from Siri while neat are not revolutionary.
My worry is that Apple might be raising people's, who don't understand, expectations that it's something that it's not.
It was impressive enough to get massive DARPA funding and evolve into the most ambitious artificial intelligence research project ever made. From that sprung Siri inc which Apple bought. The point is not to be a chatbot, others have done that already.
Here is Robert Scoble of Rackspace doing an interview with some people behind Siri, also doing some tests of the capability of it at the end.
http://scobleizer.com/2010/02/08/why-if-you-miss-siri-youll-miss-the-future-of-the-web/