3 ms·

For one service alone yes, but if well implemented, the real use case would be: “find me a tennis court where I could play tonight after 6pm and before 9pm for

For one service alone yes, but if well implemented, the real use case would be: “find me a tennis court where I could play tonight after 6pm and before 9pm for one hour. Make sure I have an Uber on time at my door. Report when you’ve ordered the Uber and where the tennis court will be” This would take quite some time to organise manually and such a service would make sense

sokoloff 3y ago
That only has value if you can simplify to “I’d like to play tennis between 6 and 9 tonight; please arrange that.” Having to write/speak the assembly language of each component of that every time in chat is not going to win in the market.
- axlee 3y ago
  How could the system even guess that you'd be departing from home? That you'll need transport to and from the court? That you want to play for one hour? I don't see these statements as "assembly language", they are questions a human assistant would ask as well.
  - sokoloff 3y ago
    The same way it would know where “my door” is in your example: From previous context, other calendar integrations, and information/settings you’ve given it (or given it access to). “There’s a court available 150 miles away; the Uber will be at your door at 3PM to get you to the court by 6” is a legal but non-sensical solution in your version of the interaction. You have to have some kind of implicit rules and some kind of explicit configuration whether that’s “when I say my door, I mean X” or “when I play tennis, the default duration is an hour and I prefer club Y over town court T1 over town court T2.” And an implicit “people don’t want to spent more time in transit than at an activity, so first constrain search and issue a notice to the user if a solution chosen violates this”.
yellow_lead 3y ago
It depends on how much 'error' or inconvenience you're willing to accept. Looking at how we might approach your tennis court scenario I see a lot of mistakes that the AI could make. Maybe the court is members only, but it didn't realize the court is inside a gym or location where a human would be sure to check that. Or it finds one an hour away, so your Uber is expensive. I assume if you do this often, you'd know about potential tennis courts in the area. So for me I'd rather do the research up front myself, but this could be personal preference. In the end, maybe we have to see how good these `AI + some code` apps get, but I wouldn't want to use the first versions.

sokoloff 3y ago

That only has value if you can simplify to “I’d like to play tennis between 6 and 9 tonight; please arrange that.” Having to write/speak the assembly language of each component of that every time in chat is not going to win in the market.

axlee 3y ago
How could the system even guess that you'd be departing from home? That you'll need transport to and from the court? That you want to play for one hour? I don't see these statements as "assembly language", they are questions a human assistant would ask as well.
- sokoloff 3y ago
  The same way it would know where “my door” is in your example: From previous context, other calendar integrations, and information/settings you’ve given it (or given it access to). “There’s a court available 150 miles away; the Uber will be at your door at 3PM to get you to the court by 6” is a legal but non-sensical solution in your version of the interaction. You have to have some kind of implicit rules and some kind of explicit configuration whether that’s “when I say my door, I mean X” or “when I play tennis, the default duration is an hour and I prefer club Y over town court T1 over town court T2.” And an implicit “people don’t want to spent more time in transit than at an activity, so first constrain search and issue a notice to the user if a solution chosen violates this”.

yellow_lead 3y ago

It depends on how much 'error' or inconvenience you're willing to accept. Looking at how we might approach your tennis court scenario I see a lot of mistakes that the AI could make. Maybe the court is members only, but it didn't realize the court is inside a gym or location where a human would be sure to check that. Or it finds one an hour away, so your Uber is expensive. I assume if you do this often, you'd know about potential tennis courts in the area. So for me I'd rather do the research up front myself, but this could be personal preference. In the end, maybe we have to see how good these `AI + some code` apps get, but I wouldn't want to use the first versions.