In order to implement the terraAI system we are going to need many good quality open-source code libraries.
If you have experience with any of the following, by all means please share your insights with us in the comments section below!
This is a live document, meaning it will be updated and republished from time to time as needed.
General requirements
- Open-source
- For user interface:
- jQuery-based, if possible
- Responsive, so that it will work well on all devices (in particular the mobile devices)
- Reasonably efficient. For example, we may reject it if its animation is jerky on low-end mobile devices.
- Good visual design. In other words, it should look good and is easy to use.
- Work well across all supported devices. Following is the proposed supported devices:
- All versions of Android
- All versions of Chrome OS
- iOS (which versions?)
- Windows XP and up
- What else?
- For machine learning
- Python-based on the server side.
- Javascript-based for certain machine learning performed on the client-side. This helps to alleviate workload on the TAI server.
- Application server
- 1.
Candidate components
- Web framework. Candidates:
- Ruby on rails
- NodeJS: is NodeJS ready to be used as a web framework? Perhaps not yet but worth watching (article Should I Use Ruby on Rails or Node.js For My Next Project/Startup)?
- What else?
- Database. Candidates:
- Javascript framework. Candidates:
- Machine learning. These are useful as the code base for supporting the learning capability. See here for further information.
- User interface. Candidates:
- TouchSwipe, for seamless support of simple touch gestures such as swipe, tap, hold, etc., across all devices.
- Visualization libraries. Candidates:
- D3.js: does it work well on mobile devices?
- ....
- ...
Open questions
- TO BE FILLED