[caribousoup]

I think this is the wrong solution.

We need strong whitelist controls in all of the browsers. Because when you connect to a website, you have no idea if it's going to start downloading 30 MB worth of images and 40 Javascript requests.

It's gone completely out of control. I say this as a web dev on a rural connection. Not only is it slow, I PAY for this, bandwidth is limited. For some lazy dev who doesn't compress their images, and it downloads without me noticing, I pay actual money for that.

If browser preferences has options for auto disabling Javascript on a webpage that initiated X amount of outbound connections, or auto stops when there are 30 img requests on page load, I should be in control of not allowing that. I should be in control of setting that threshold for different request types.

We don't need a second, lightweight web. We need to fix our tooling at the client level.

[qwerty456127]

> If browser preferences has options for auto disabling Javascript on a webpage that initiated X amount of outbound connections, or auto stops when there are 30 img requests on page load, I should be in control of not allowing that. I should be in control of setting that threshold for different request types.

I doubt this is hard to implement. Somebody can build a Firefox/Chromium fork (or even an extension perhaps?) implementing this functionality and they don't have to be a real genius. Once I needed that for some time I just used a local proxy to cache and limit what is downloaded.

The only problem is the actual web sites are not designed with the fact somebody might want to control them this way in mind and can easily happen to be unusable unless you let them download a ton of stuff.

I doubt the problem can be solved without some sort of enforcement by a major power (Google, governments or whatever). At the same time trying to enforce anything on the existing web doesn't seem making much sense to me.

[anthk]

UBlock Origin can do that. It can block media over X MB, and lot of nasty JS things too.

[caribousoup]

I agree, it's not difficult to implement.

For instance, in Apple Mail, there is a setting that is labeled: Prompt me to skip messages over X MB. I bet almost nobody ever sets this up. But I do. I get a prompt asking me if I want to download 20MB of attachments. It also lets me skip downloading remote content in messages, but I can override and allow it with a click of a button.

If browsers had similar advanced options, that users needed to manually enable, then those users would understand what's happening when a website isn't working because of the content blocking.

There are lots of possible UI implementations. A button that says: X amount of things were blocked, would you like to reload the page and disable blocking? Or replace img tags (etc) with placeholder buttons, and click the placeholder to initiate the http request.

Safari has recently done some great UI for controlling which websites have access to autoplay. I would like to see this expanded, where I can choose to either blacklist or whitelist Javascript, or images and videos over X MB.

I don't think it needs enforcement, I mean, you can't control anything on the client end already. We already have APIs to check if cookies etc are enabled. We could expand this to check for other fine grain controls, like are images enabled. Or just use the noscript tag, which devs should be doing anyways but usually don't.

Anyways. For those on slow connections... this problem will only get worse over time. Something will eventually need to be done about it, otherwise like 20% of the connections are going to get left behind unable to even use the web.

[sloum]

The problem is that sites fail without their JS. Oftentimes even if that JS is purely decorative and shouldnt be required for a site to function. I block all known google domains on my laptop and as a result am cut off from a great many sites (blocks ads and captchas). I also use the dillo browser and disable css, js, and images. It is an unusable mess. Gopher and gemini provide a content first approach that will not work for everything and is in no way going to _replace_ the web. But is really great for certain types of content/discourse.