Hacker News Reader

Comment by joegibbs | Hacker News Reader

“Get me a burrito. Ok cool are there any other places? Ok but how about ones under 50 minutes delivery? Ok now of those which has higher than 4.5 stars? Alright that one looks good, can I get a large lamb burrito with extra salsa and guac? Ok they don’t do lamb that’s fine, beef is good. They don’t do a large size, that’s alright go with a small. Yes confirm payment.”

Text is just not a good format for ordering things. You want to see pictures and just click on the options you want - we’ve had people paid 300k working on the optimal visual interface for this for probably 30 years, just trying to improve the sale rate 2%.

Agents are good for open ended problems which require stringing together lots of different data sources, which couldn’t be further from ordering food.

View on HN

> Text is just not a good format for ordering things. You want to see pictures and just click on the options you want

Do you look at pictures and click on things at a restaurant? I suspect you most often read text and talk to people. I think it works just fine.

I think they meant text as in texting/chat interfaces.

I mean a text interface - it’s more like if you went to a restaurant and there was no menu so you had to ask the waiter everything, but he also spoke absolutely no English so you had type out your questions on your phone for him - the latency while the model collates the data is not good

With a well performing voice model I don't think it's going to be much different than ordering at a restaurant.

To be clear I'm not advocating for this and likely wouldn't use it myself but it seems perfectly viable for a certain kind of customer. Particularly those with decision paralysis that are spoiled for choice.

I’m still waiting for voice commands for controlling your computer to take off.

They’ve been around since at least the 90s, but the keyboard and mouse still are dominant.

Will well performing voice models eliminate the need for touchscreen devices?

Do you really want to vocally explain everything to your computer? Every move of the cursor? Every word typed, backspace, every button click?

That only works in Star Trek because it's a plot device. In real life I think that would be insanely tedious.

Yeah, I definitely want to buy a $1,799 mobile device from OpenAI that has a dynamic, AI-generated user interface I navigate by voice. It will be revolutionary.

this diva reads the menu at the restaurant like the missal in a church pew, not some ipad kid with sticky fingers

> Agents are good for open ended problems which require stringing together lots of different data sources, which couldn’t be further from ordering food.

i feel like you were on the right path but arrived at the wrong conclusion. if you think you're right because the blocking factor is text, you arent paying attention

I don’t think it would be good UX with voice either. An agent could dynamically recreate something resembling the DoorDash UI which probably would work, but would lose out on the familiarity the user has with the actual doordash

Reply on news.ycombinator.com