UX improvements and roadmap

[mindhells]

There's already many cool enhancements identified . I would like to extend that list with some more. But I think I might be better to share some ideas before creating any issues. I would like to know what do you think about them, if they'd add some value to Gnofract4d. The goal is to know what users of Gnofract4d expect it to be in the future, and what should be the focus of development.

Here is a list of identified "problems" and possible solutions. Some of them are already proposed, others have been discussed in a different context:

• Installation is complicated for regular users, and they also have to struggle with dependencies:

  • Redistributable installer by platform: using flatpak and flink as discussed in #146
  • Embedded python: pyoxidizer
  • Web interface (no installation at all) (WASM)

• calculation speed

  • GPU support: #78
  • XaoS algorithms: #80

• zoom level limitation

  • arbitrary precision: #77
    • raw (calculate every point with AP)
    • using perturbation theory (calculate only a reference point and the rest calculations relative to it)

• user interface feedback

  • on screen help: some kind tutorial (see https://introjs.com/ as a reference)
  • immediate (for example, let the user know he’s just zoomed in or out with some kind of visual effect/animation)
  • zoom level scale (like google maps)

• formula language complexity and lack of documentation

  • formula creation wizard
  • syntax checker
  • autocompletion

• we don't have any usage statistics (see #163 )

  • integrate a 3rd party analytics service

[nayraq80]

Regarding to the installation process, and trying to improve this process and make it possible for any user (with or without technical skills), I would suggest take into consideration the option of building and distributing binary packages for different Linux distros. I think it would be great if an automatic process would be in charge of this process, for example, «Open Build Service».