5 ms·

It is obvious what people want. A single app that does it all. Seriously. We need a general purpose application that can be used for 80% of all use cases. When

It is obvious what people want. A single app that does it all. Seriously. We need a general purpose application that can be used for 80% of all use cases. When you think about it, all apps pretty much re-introduce the exact same features. Preferences, contacts, notifications, authentication, sharing, etc. I want to use the exact same interface and language to order a package from China to my house, or to hail a cab to move me from work to home. I want the estimated ETA and cost prediction to be displayed the same. Tracking a package or the real-time location of a cab should be done through the same interface. I want to be notified the same way, whether it's my package or my cab that's late and/or has arrived. I want to pay for the package and cab ride the exact same way. Reviewing the received product and reviewing the cab driver, should be identical processes. I think the same should apply to pretty much any interaction, be it to ask my coffee machine to brew a cup, to be notified that my colleague will be late to a meeting, to pay for a meal, to review a movie, to schedule an appointment, to locate a friend. My thesis is that all communication problems (which are what apps solve) are difference instances of the same one. An app is not any tool. It's a language. We need to implement a general purpose language, and then all use it to interact with different agents/services.

draugadrotten 11y ago
> It is obvious what people want. A single app that does it all. Seriously. This is the vision of SIRI. Now you know the goal. Please deliver.
- miguelrochefort 11y ago
  Siri's mistake (which also applies to Cortana and Google Now) is to bet on natural language. I seriously believe natural language to be a mediocre interface, and we have the opportunity to do much better. We need to create a completely new language.
  - seagreen 11y ago
    This is actually a really interesting point. So we have English (optimized for person->person), standard programming languages (optimized for text->computer), and now you're saying we should make Language X, which is optimized for speech->computer. I disagree with it, but it's still interesting :)
    - miguelrochefort 11y ago
      We need a language optimized for human->computer->human. We already have a few, they're called GUIs. But there's little to no standard surrounding them. Sure, some patterns like "x" to dismiss something or pinch to zoom are widespread, but they don't constitute a complete language. The binary equivalent is Akinator [1]. It's big on Tinder too. Basically, you want a language that lets you point at things you don't like. This gets converted to a need, which is then communicated to an array of agents whose job is to come up with solutions. Then enters the flow where you pick a solution (most of which come with conditions, such as completion time or cost), which then becomes a contract that feeds the timeline/calendar/task management system that drives your day to day actions. There's not a lot more to it. [1] http://en.akinator.com/ http://en.akinator.com/
      - walterbell 11y ago
        Akinator UX is like a marketing survey, a bit tiresome after the first dozen questions with no end in sight.
  - Kluny 11y ago
    Yes - natural language is dependent on facial expressions, tone of voice, shared mutual history, shared culture, and so many other things that you don't have, or even want, when interacting with a computer. People can learn a set of imperative commands for interfacing with an AI. People aren't dumb. But the execution has to be good enough to make it worthwhile.
  - s73v3r 11y ago
    Problem being that you're now relying on the willingness of people to learn your new language, which creates a friction point, like someone here was talking about.
bsharitt 11y ago
> It is obvious what people want. A single app that does it all. Seriously. We need emacs mobile!
- miguelrochefort 11y ago
  But text is an evil medium.
morgante 11y ago
Check out WeChat. They have a lot of work to do, but they're basically embracing this idea of a universal language for all computing tasks. In their case, it's centered around messaging but can cover a wide range of uses.
- miguelrochefort 11y ago
  I heard about WeChat a long time ago, but I've done some research about them just the other day and I was amazed by the array of services and use cases they manage to cover. I'm not a big fan of the way it's implemented (it's like nested web-based mini-apps), but it's much better than anything we have here in the West.
  - walterbell 11y ago
    How would you implement the meta-app differently with the constraint of extensibility for third-party services? E.g. should the app be updated for each new service or use a central broker? What's the separation/security model for user data that is private to the device, visible to central broker, visible to all 3rd-party services, or visible to one service?
    - miguelrochefort 11y ago
      You start with a semantic knowledge graph. Then you program agents that manipulate that graph. Think semantic web. Nothing is centralized. The challenge is the UI. We need some kind of AI that understands the context and user's preference in order to dynamically generate the best interface, taking into account both data consumption and input.
      - walterbell 11y ago
        If the graph is not centralized, does that imply it is public? Or could segments of the graph be private to individuals, groups, companies, nations, etc? E.g. the industry currently has FreeBase/Wikidata (public) and Google/Facebook/Microsoft semantic graphs (private). Would client endpoints arbitrate between these non-intersecting graphs?
nickpsecurity 11y ago
That's basically Facebook's model: barely good enough at so many things in one place with your friends already there. Works great for them.
- miguelrochefort 11y ago
  - I can't buy stuff on Facebook. - I can't sell stuff on Facebook. - I can't hail a cab on Facebook. - I can't unlock my doors on Facebook. - I can't turn the lights on on Facebook. - I can't play a song on Facebook. - I can't pay for my parking spot on Facebook. - I can't reserve a table at a restaurant on Facebook. - I can't manage tasks on Facebook. - I can't review products on Facebook. - I can't track my sleep on Facebook. Unless you consider Status/Messages/Comments as an interface to all these things. In which case we could all use IRC and live happily ever after. Which doesn't work because IRC is text, and text is an poor medium.
  - nickpsecurity 11y ago
    Oh, I see what you mean. Something doesn't do a lot in one app unless it runs all software on Earth in one app (or some subset thereof). I think you meant operating system or runtime platform instead of app. I recommend Android or iOS for your needs. Back to apps and my claim, Facebook directly provides email, messaging, phone calls, group activities, meetup, games, and so on. Hence, the claim their model is to do much of what people need in one app good enough for convenience to ensure a win. No single app is going to do all on your list because users would reject it outside maybe South Korea where such a mixed bag is common. People always look for a better app with better interface, integration, price, whatever. No incentive whatsoever to build an app with your exact combination of features if 99+% of people prefer to have a set of apps best for that and aren't vendor loyal at all. Not to mention, what startup is going to build that app for $1 or so in an app store? The functionality you describe could cost hundreds of thousands to millions to develop. With no expectation of a return. Not happening any time soon.
    - miguelrochefort 11y ago
      There seems to be some misunderstanding. I'm not talking about one app with all these functionalities as distinct features. I'm talking about one app that enables the use of a modern computer-assisted general-purpose language with which people can achieve most of these tasks. The app won't know anything about cabs, sleep, restaurant tables, doors, lights, parking spots. These things are part of a lexicon that exists outside the app, as part of the graph that powers it. Think of this graph as a knowledge base which smart (automated or not) agents can interact with. Once you nail the UX algorithm, you don't need to manually craft unique user interfaces for all possible use cases. It is all done automatically. Also, I don't understand where this need to discuss cost or revenue comes from. Projects that have the largest impact are not motivated by self-interest or money of any kind. What I'm describing will be entirely free, open source, decentralized, etc. It's closer to "English" or "The Internet" than it is to "Instagram" or "Amazon".

draugadrotten 11y ago

> It is obvious what people want. A single app that does it all. Seriously. This is the vision of SIRI. Now you know the goal. Please deliver.

miguelrochefort 11y ago
Siri's mistake (which also applies to Cortana and Google Now) is to bet on natural language. I seriously believe natural language to be a mediocre interface, and we have the opportunity to do much better. We need to create a completely new language.
- seagreen 11y ago
  This is actually a really interesting point. So we have English (optimized for person->person), standard programming languages (optimized for text->computer), and now you're saying we should make Language X, which is optimized for speech->computer. I disagree with it, but it's still interesting :)
  - miguelrochefort 11y ago
    We need a language optimized for human->computer->human. We already have a few, they're called GUIs. But there's little to no standard surrounding them. Sure, some patterns like "x" to dismiss something or pinch to zoom are widespread, but they don't constitute a complete language. The binary equivalent is Akinator [1]. It's big on Tinder too. Basically, you want a language that lets you point at things you don't like. This gets converted to a need, which is then communicated to an array of agents whose job is to come up with solutions. Then enters the flow where you pick a solution (most of which come with conditions, such as completion time or cost), which then becomes a contract that feeds the timeline/calendar/task management system that drives your day to day actions. There's not a lot more to it. [1] http://en.akinator.com/ http://en.akinator.com/
    - walterbell 11y ago
      Akinator UX is like a marketing survey, a bit tiresome after the first dozen questions with no end in sight.
- Kluny 11y ago
  Yes - natural language is dependent on facial expressions, tone of voice, shared mutual history, shared culture, and so many other things that you don't have, or even want, when interacting with a computer. People can learn a set of imperative commands for interfacing with an AI. People aren't dumb. But the execution has to be good enough to make it worthwhile.
- s73v3r 11y ago
  Problem being that you're now relying on the willingness of people to learn your new language, which creates a friction point, like someone here was talking about.

bsharitt 11y ago

> It is obvious what people want. A single app that does it all. Seriously. We need emacs mobile!

miguelrochefort 11y ago
But text is an evil medium.

morgante 11y ago

Check out WeChat. They have a lot of work to do, but they're basically embracing this idea of a universal language for all computing tasks. In their case, it's centered around messaging but can cover a wide range of uses.

miguelrochefort 11y ago
I heard about WeChat a long time ago, but I've done some research about them just the other day and I was amazed by the array of services and use cases they manage to cover. I'm not a big fan of the way it's implemented (it's like nested web-based mini-apps), but it's much better than anything we have here in the West.
- walterbell 11y ago
  How would you implement the meta-app differently with the constraint of extensibility for third-party services? E.g. should the app be updated for each new service or use a central broker? What's the separation/security model for user data that is private to the device, visible to central broker, visible to all 3rd-party services, or visible to one service?
  - miguelrochefort 11y ago
    You start with a semantic knowledge graph. Then you program agents that manipulate that graph. Think semantic web. Nothing is centralized. The challenge is the UI. We need some kind of AI that understands the context and user's preference in order to dynamically generate the best interface, taking into account both data consumption and input.
    - walterbell 11y ago
      If the graph is not centralized, does that imply it is public? Or could segments of the graph be private to individuals, groups, companies, nations, etc? E.g. the industry currently has FreeBase/Wikidata (public) and Google/Facebook/Microsoft semantic graphs (private). Would client endpoints arbitrate between these non-intersecting graphs?

nickpsecurity 11y ago

That's basically Facebook's model: barely good enough at so many things in one place with your friends already there. Works great for them.

miguelrochefort 11y ago
- I can't buy stuff on Facebook. - I can't sell stuff on Facebook. - I can't hail a cab on Facebook. - I can't unlock my doors on Facebook. - I can't turn the lights on on Facebook. - I can't play a song on Facebook. - I can't pay for my parking spot on Facebook. - I can't reserve a table at a restaurant on Facebook. - I can't manage tasks on Facebook. - I can't review products on Facebook. - I can't track my sleep on Facebook. Unless you consider Status/Messages/Comments as an interface to all these things. In which case we could all use IRC and live happily ever after. Which doesn't work because IRC is text, and text is an poor medium.
- nickpsecurity 11y ago
  Oh, I see what you mean. Something doesn't do a lot in one app unless it runs all software on Earth in one app (or some subset thereof). I think you meant operating system or runtime platform instead of app. I recommend Android or iOS for your needs. Back to apps and my claim, Facebook directly provides email, messaging, phone calls, group activities, meetup, games, and so on. Hence, the claim their model is to do much of what people need in one app good enough for convenience to ensure a win. No single app is going to do all on your list because users would reject it outside maybe South Korea where such a mixed bag is common. People always look for a better app with better interface, integration, price, whatever. No incentive whatsoever to build an app with your exact combination of features if 99+% of people prefer to have a set of apps best for that and aren't vendor loyal at all. Not to mention, what startup is going to build that app for $1 or so in an app store? The functionality you describe could cost hundreds of thousands to millions to develop. With no expectation of a return. Not happening any time soon.
  - miguelrochefort 11y ago
    There seems to be some misunderstanding. I'm not talking about one app with all these functionalities as distinct features. I'm talking about one app that enables the use of a modern computer-assisted general-purpose language with which people can achieve most of these tasks. The app won't know anything about cabs, sleep, restaurant tables, doors, lights, parking spots. These things are part of a lexicon that exists outside the app, as part of the graph that powers it. Think of this graph as a knowledge base which smart (automated or not) agents can interact with. Once you nail the UX algorithm, you don't need to manually craft unique user interfaces for all possible use cases. It is all done automatically. Also, I don't understand where this need to discuss cost or revenue comes from. Projects that have the largest impact are not motivated by self-interest or money of any kind. What I'm describing will be entirely free, open source, decentralized, etc. It's closer to "English" or "The Internet" than it is to "Instagram" or "Amazon".