5 ms·

This makes it sound like there's only a single platform for "augments," and we all need to share it. In reality, I imagine there will be many different platfor

This makes it sound like there's only a single platform for "augments," and we all need to share it. In reality, I imagine there will be many different platforms, some free-for-all and some tightly curated. Users will choose which platform's augments they want shown at any given time. Which doesn't address any of the platform-related free speech concerns, but I think that's a much more general discussion, that has little to do with AR.

doctorhandshake 4y ago
Bummed to hear it came across that way. In the next post I’m going to detail exactly how I envision it working, but my intent with Augmented Realist is to advocate for something like an AR web. The ideal system would allow users to choose their own markup/language, payloads, hosting, search, filtering, ‘browsers’, hardware etc., with scene understanding happening on-device. The secret sauce to this design is to allow users choice in the 'semantic lookup' labeling step between scene understanding/segmentation and the use of language to search the world of available augments. In my opinion, the most important component of a hypothetical AR web absolutely nobody is talking about is how language is used to describe the world, and my proposed solution is focused on assuming that there can be no consensus in that regard. I believe if we can design with that acknowledgement in mind we can unlock a truly modular, private, free, decentralized AR web.
- robertlagrant 4y ago
  > the most important component of a hypothetical AR web absolutely nobody is talking about is how language is used to describe the world, and my proposed solution is focused on assuming that there can be no consensus in that regard. This is the opposite of how the actual web happened, with just HTML, CSS and Javascript used to describe it.
  - doctorhandshake 4y ago
    To clarify, I meant language like English or Spanish. This is a hard point to compress into a comment but - if there is a central authority that labels things definitively (think DNS-> IP), we have a problem where the very first step in attaching an augment to the world requires you to use the language someone else has chosen for you to describe it. I framed this up in the ‘Manifesto’ post on the blog: https://noahnorman.substack.com/p/manifesto https://noahnorman.substack.com/p/manifesto
    - superb-owl 4y ago
      Obviously we should standardize on Esperanto!
      - doctorhandshake 4y ago
        Even beyond the language (eg Esperanto) the words are expressed in, the words themselves are likely to be subject to some serious debate. It is entirely wrong, IMO, to believe that you can acquire consensus on what things are called even in ‘objective’ labels, to say nothing of the clearly subjective language we use to describe things.
- superb-owl 4y ago
  Definitely a noble undertaking! Will be curious to see what you're thinking on a technical level, especially to what degree you want to build on top of existing web standards, web 3 tech, or invent something new.
  - doctorhandshake 4y ago
    It’s quixotic and theoretical for sure. Please watch that substack (or subscribe) - next post I’m going into hard details after a year of talks on the topic.

doctorhandshake 4y ago

Bummed to hear it came across that way. In the next post I’m going to detail exactly how I envision it working, but my intent with Augmented Realist is to advocate for something like an AR web. The ideal system would allow users to choose their own markup/language, payloads, hosting, search, filtering, ‘browsers’, hardware etc., with scene understanding happening on-device. The secret sauce to this design is to allow users choice in the 'semantic lookup' labeling step between scene understanding/segmentation and the use of language to search the world of available augments. In my opinion, the most important component of a hypothetical AR web absolutely nobody is talking about is how language is used to describe the world, and my proposed solution is focused on assuming that there can be no consensus in that regard. I believe if we can design with that acknowledgement in mind we can unlock a truly modular, private, free, decentralized AR web.

robertlagrant 4y ago
> the most important component of a hypothetical AR web absolutely nobody is talking about is how language is used to describe the world, and my proposed solution is focused on assuming that there can be no consensus in that regard. This is the opposite of how the actual web happened, with just HTML, CSS and Javascript used to describe it.
- doctorhandshake 4y ago
  To clarify, I meant language like English or Spanish. This is a hard point to compress into a comment but - if there is a central authority that labels things definitively (think DNS-> IP), we have a problem where the very first step in attaching an augment to the world requires you to use the language someone else has chosen for you to describe it. I framed this up in the ‘Manifesto’ post on the blog: https://noahnorman.substack.com/p/manifesto https://noahnorman.substack.com/p/manifesto
  - superb-owl 4y ago
    Obviously we should standardize on Esperanto!
    - doctorhandshake 4y ago
      Even beyond the language (eg Esperanto) the words are expressed in, the words themselves are likely to be subject to some serious debate. It is entirely wrong, IMO, to believe that you can acquire consensus on what things are called even in ‘objective’ labels, to say nothing of the clearly subjective language we use to describe things.
superb-owl 4y ago
Definitely a noble undertaking! Will be curious to see what you're thinking on a technical level, especially to what degree you want to build on top of existing web standards, web 3 tech, or invent something new.
- doctorhandshake 4y ago
  It’s quixotic and theoretical for sure. Please watch that substack (or subscribe) - next post I’m going into hard details after a year of talks on the topic.