Exactly.
It’s really hard to describe what I’m seeing and convey it in a way everybody could understand. Seems like you’re catching on.
Richard Stallman wanted to build OS and applications that respects users freedom. So he build emacs. Emacs is very powerful application that it can be run as it’s own operating system. Anything inside of emacs can be run applications or play games while retaining the UI as the same. Nothing changes the UI unless the user explicit to do so. This way, I could run games, trade, communicate all in the same UI, instead of using application that deploy different UI. These days, every application has it’s own UI. Like websites… Every website provides quite an annoying experience where we all have to search to find a button to access to another webpage. Contact us could be at left column, or top column, or bottom column. We all have to search for it which makes experience unpleasant, and wasting time. Why not allow user make that decision? The best part is about this, devs wouldn’t have to worry about building UI at all if it uses emacs. This principle applies with music, gaming, videos, and other means. Redox is right up in that ally. It could be like… file system UI(thunar, spacefm, etc) blended with emacs, and web browser into one unified application.
If one access to the file via url, the inner-box viewer should display it. Just like how emacs and browser does it…For an example, If one want to run a game AI but doesn’t want to deploy the game. It could open up blender inside of file system ui browser, and see how the ai functions inside of that box viewer.
As I said multiple times in other threads about the emerge of conversational commerce, we all want UI experience to be the same. We don’'t care about how data is stored as long it is projected the way we wanted to project. If we want particular data to be on left column, we want to do so. We shouldn’t have devs enforce it unless users choose to do so.
Redox has potential to become the ultimate personal os. Data are static. Apps are dynamic.
This is where “blur between the lines” that dirvine was talking about.
It’s all about layering, and setting up ways to deploy scripts at certain areas within the inner-box window. Each inner-box window could be sand boxed. We can layer sandbox of the entire file system browser, or layer it in each sandbox. Not all apps has to be sand boxed, and can tax on resources. Layer does matter. Emacs does so brilliantly. It even has a mode that could look html/css live as you type. Very interactive system for users, and it doesn’t depend so greatly like monolithic kernel, and package system.
Thus, my confusion on URL hashing with safenet dns. I personally would like to have safenet dns to be at app level, not core. This way, I or others would build a pet naming system within the file system browser. Users would have more freedom on how it’s labeled. Apps devs wouldn’t have to worry about it majority of the time. They could even share pet naming system config file to users who would like to use default naming convention. At least give them freedom to change the name without affecting the app behavior. This is why I love about safe and ipfs hash table. It remains the same but naming can always change. With linux fs, if one change the name, the app wouldn’t recognize it. Redox could do the same. Another thought was redox runs on zfs, pool system. Each pool can run on it’s own instance, and expand. We can tweak the usage of the each resource buffer.
Lets switch to dns subject, is there a way to view raw data without dns?
There is a lot of items on our immediate TODO list atm but things are really paying of with rust here as we can make changes very quickly. So this is taking shape. Tests on Friday were showing instability but today with a few more tweaks they were actually very good. Lets see what tomorrow brings. Adam has a really nice PR for many platform deployment (I know one 
