[Karunamon]

You describe a tech problem, and not a really intractable one. It already has "real time communications (as, yknow, the main purpose of the service)", it can already do GPS coords, and expect payment integration soon.

I'd love to see Twitter become a "command line of the web". Add voice integration on top of it via siri/google/etc, and it come become the plaintext glue between a lot of services.

What we have here is a pub/sub message transmission layer that just happens to be used by a lot of people for personal communication. Having machines use that for other things is something that's already been experimented with - using it as a de-facto API layer is a logical step.

[CPLX]

But this is nonsensical. As far as I can see there are two parties to this hypothetical transaction, me the smartphone owner, and Uber the company. I open the Uber app and use it.

Why do I want to insert Twitter in the middle of this transaction? What problem is it that you're trying to solve?

[Karunamon]

The problem that you need a unique app to interact with every single service out there. How do you feel about random websites pushing you onto their application and then neglecting the web experience? It's fragmentation city.

This is the same problem, but for services (in the people doing things for other people sense) rather than hypertext.

Think of it as a glue layer that gets you around the need to have 50 different apps installed for 50 different services.

[CPLX]

> The problem that you need a unique app to interact with every single service out there.

This is not an actual problem. And to the extent that it is 140 characters of text doesn't come close to fixing it. And in the unlikely event that it did why wouldn't SMS do the same thing without having to bring some random third party company into the mix and have both parties at their mercy for zero value added?

[Karunamon]

This is not an actual problem.

We disagree there. It's annoyance waiting to be solved. Here's a hypothetical, realistic interaction, and as a bonus, one that isn't any less useful due to the length limit:

    >>@friend: Want to catch lunch today?
    @friend: Sure, I'll set that up. Cya there!
    @restaurant Reserve a table for 2 at noon for Karunamon
    >>@restaurant: Confirmed, we'll see you at noon!
    @traffic: How's the traffic between here and @restaurant?
    >>@traffic: ETA 35 minutes, heavy traffic and construction along the optimal route.
    @restaurant Change that last reservation to 12:30
    >>@restaurant: Confirmed, your noon reservation has been moved to 12:30
    @uber Pickup my location to the closest @restaurant
    >>@uber: Driver is on its way. ETA: 5 minutes. Look for a grey Subaru, plates 1234-56.
    @friend: Catching a ride there, see you in half an hour!

And this is pure user interaction, which is the least efficient way to do things possible. If you can't think up a usage for a widespread pub/sub notification system, kindly get your imagination serviced.

As far as the limit, it's up for revision by the sound of things anyways - and keep in mind the original purpose of that limitation was so a tweet would fit into an SMS. If the limit is what you think breaks this idea, you've disqualified SMS by the same rubric.

[setpatchaddress]

You've just described the future of Siri, not the future of Twitter.

[Karunamon]

I don't know about you, but I don't want a separate walled garden of services based on the particular voice recognition app my phone uses (whether that be Siri or Now or Cortana or...). Using some other service that provides a sane communication layer that everyone already uses means that doesn't have to happen.