3 ms·

I don't know how perfectly you could do this. If there are going to be arbitrary mouseover effects then JS has to be able to see your mouse. If you want Jupyter

by iguy 8y ago

I don't know how perfectly you could do this. If there are going to be arbitrary mouseover effects then JS has to be able to see your mouse. If you want Jupyter notebooks to indent text & do tab completion then JS has to see text you haven't submitted. That said, I'd like my browser to be much more on my side -- by default disclose very little about my setup, by default partition cookies according to the headline site I visited, and delete them soon. I'd like JS completely off on most sites I visit, and more convenient ways to turn on as much of it as needed on some sites.

asark 8y ago
> If there are going to be arbitrary mouseover effects then JS has to be able to see your mouse. If you want Jupyter notebooks to indent text & do tab completion then JS has to see text you haven't submitted. Just don't allow those things. Those aren't hypertext browsing, they're app-things. Wikipedia or Timmy's Ninja Turtle Fan Club or even banking and ordinary ecommerce sites don't really need those things, and having those things on the same platform as all these "apps" get them all mixed up in a spying and vulnerability nightmare for no good reason. I think hypertext browsing is distinct enough from what the current web is, and that it's still a useful enough idea, that it'd be nice to split that off from... whatever this has become.
- iguy 8y ago
  Indeed. I guess you're picturing two apps, where I'm picturing one app with much easier per-site settings, with much the same goal. My current plan is to make lots of Fluid-app browsers, for certain sites, and then leave JS off most of the time. But there are enough sites which don't work at all, yet I want to visit, to make this a bit of a pain.
  - asark 8y ago
    One major benefit of a separate program and network (network of hypertext, that is, not IP network or whatever) is that if you can overcome an initial usage hurdle (big if, I know) it'd provide an audience incentive to get current websites that could work just fine on it (which is lots of them) to provide a second experience on it. The way they do now with web+mobile app, but it'd obviously be cheaper to provide than either of those, by a long shot. Gaining the ability to access, banking, ecommerce, and government resources in particular through a hypertext network that's not also a spyware and application delivery platform would be huge for security.
    - iguy 8y ago
      I wonder whether pushing accessibility is another way forwards. Maybe the simple-browser should present itself as a screen reader for the blind, and automatically send threats of ADA lawsuits if a site isn't usable. I know for instance that UK govt. websites are designed for such use first, which just means sticking close to the original spirit of hypertext. Bank websites I have little hope of... and anyway I trust my bank quite a lot! But commerce would be great.

by iguy 8y ago

asark 8y ago

> If there are going to be arbitrary mouseover effects then JS has to be able to see your mouse. If you want Jupyter notebooks to indent text & do tab completion then JS has to see text you haven't submitted. Just don't allow those things. Those aren't hypertext browsing, they're app-things. Wikipedia or Timmy's Ninja Turtle Fan Club or even banking and ordinary ecommerce sites don't really need those things, and having those things on the same platform as all these "apps" get them all mixed up in a spying and vulnerability nightmare for no good reason. I think hypertext browsing is distinct enough from what the current web is, and that it's still a useful enough idea, that it'd be nice to split that off from... whatever this has become.

iguy 8y ago
Indeed. I guess you're picturing two apps, where I'm picturing one app with much easier per-site settings, with much the same goal. My current plan is to make lots of Fluid-app browsers, for certain sites, and then leave JS off most of the time. But there are enough sites which don't work at all, yet I want to visit, to make this a bit of a pain.
- asark 8y ago
  One major benefit of a separate program and network (network of hypertext, that is, not IP network or whatever) is that if you can overcome an initial usage hurdle (big if, I know) it'd provide an audience incentive to get current websites that could work just fine on it (which is lots of them) to provide a second experience on it. The way they do now with web+mobile app, but it'd obviously be cheaper to provide than either of those, by a long shot. Gaining the ability to access, banking, ecommerce, and government resources in particular through a hypertext network that's not also a spyware and application delivery platform would be huge for security.
  - iguy 8y ago
    I wonder whether pushing accessibility is another way forwards. Maybe the simple-browser should present itself as a screen reader for the blind, and automatically send threats of ADA lawsuits if a site isn't usable. I know for instance that UK govt. websites are designed for such use first, which just means sticking close to the original spirit of hypertext. Bank websites I have little hope of... and anyway I trust my bank quite a lot! But commerce would be great.