4 ms·

This is why data caps are bad. Left open all month that could be 400+ GB of data used on your plan. Ad networks are being subsidized by data caps and personal/

This is why data caps are bad. Left open all month that could be 400+ GB of data used on your plan. Ad networks are being subsidized by data caps and personal/business broadband costs. Maybe if ad serving went through the main host and their own bandwidth then maybe ad networks would have a reason to control this abuse of transfer.

kr0 10y ago
Thats actually a really smart idea. You get the benefit of ads but also pay for the bandwidth they consume the ad delivery network
sundvor 10y ago
My phone plan is 6GB/month. After that, I PAY. And they wonder why we use adblockers.
cobookman 10y ago
Or on the flip side. Data caps will make people care about how much data their site uses. Cause customers will stop using data heavy sites as they have economic insentive not to. Making the web all around faster as no more 10MiB webpages, and making the industry move back to server side rendering.
- tedunangst 10y ago
  "total data transferred: X MB ($y.zz)" should be part of a browsers standard status bar.
  - vinnymac 10y ago
    I know we can easily get some info from chrome dev tools and chrome://net-internals/#bandwith. It would be convenient for users if someone put together an extension that would display this information by default, or alternatively browsers just had a setting we could turn on.
    - phaed 10y ago
      On it
  - YSFEJ4SWJUVU6 10y ago
    Perhaps not – I don't think it's the browser makers job to figure out what terms the current connection has (and for the vast majority of workstation desktops there are no such terms, I think; you're getting a pretty rough deal if you pay by the byte on a fixed connection, no matter the price).
- morsch 10y ago
  It's not easy to discover how much data usage a specific site triggers. Android reports data usage per app, if you dig into the right preference panel, but that's not going to tell you anything about specific websites. I'd like a notification to pop up when the current usage (within, say, a 3 minute window) exceeds my typical usage. That'd make it easier to notice a website started autoplaying a (muted) video, for instance. It's also something that could be done at the operating system level. The operating system could also conceivably show traffic by domain, but that's not ideal; you want traffic by referrer domain, ie. if the LA Times embeds megabytes of images from some random third party server, those should be aggregated under latimes.com and not under some meaningless CDN domain. But the OS probably doesn't have that information. Of course browser vendors could add more detailed reporting. Say a brief in-browser notification ("Page finished downloading 5 MB.") after initial load is complete and repeating if excessive downloading continues. I guess that could annoying really fast, though. A less annoying alternative would be a special page (e.g. "about:data-usage") that has an overview per page in a user-selectable time range. Maybe link to it from the browser start page.
- Bahamut 10y ago
  Or make use of things like service workers to cache pages locally.
- makecheck 10y ago
  It has been years since data caps though, and years of obnoxious ads, nothing has changed. The only way in my opinion to fix this is to (a) give phones HARD preferences on maximum data that cannot be exceeded and (b) give web browsers even stricter per-site settings with sane defaults. Then, let the ad companies fight within themselves to figure out how to deal with the fact that none of their content gets through anymore. Force them to innovate into smaller form factors.
  - cobookman 10y ago
    It's cause the data caps are fairly high. For example whats 10MiB when you have a 300GiB data cap. But if we change to a utility like model people will start caring about penny pinching their bills
    - craigds 10y ago
      In the US, and on wired connection maybe. In NZ the largest mobile plan I can even find has 5GB. Another one with flat rate for data is $50 for 3GB (so $250 for this 14GB day). If I load a webpage on my mobile data that uses 10MB of data, I want to know about it so I can shut that down (and never come back.)
amluto 10y ago
I'd like to a see, or at least think about, a system in which everyone pays for egress and no one pays for ingress. That way the LA Times is welcome to throw as much crap as they want at you without charging you much. (One issue is that egresss is involved in every request. Some UI work would be needed to make this work well.) Universal egress fees would go a long way to mitigating DDoS, too.
- gruez 10y ago
  >Universal egress fees would go a long way to mitigating DDoS, too. How? By charging the owners of infected machines?
  - amluto 10y ago
    Why not? If you have a water leak, you get a high bill and then you can ask your water agency to refund you if you fix the leak. ISTM being infected and spewing data could be handled similarly. Currently there is effectively no incentive to avoid having a gadget that joins a botnet.
- kalefranz 10y ago
  How does npmjs.org, pypi.python.org, and rubygems.org survive in this world of universal egress fees?
- PhasmaFelis 10y ago
  I'm curious what you're thinking of specifically, because I can't imagine any remotely reasonable way to make egress fees work from a technical standpoint, let alone widespread acceptance and adoption. You browse to a site, you can read all the articles you want, but you can't load any other sites until you pay?
  - amluto 10y ago
    I'm a bit confused by your question. I'm saying it might be reasonable to charge per byte sent, that's all.

kr0 10y ago

Thats actually a really smart idea. You get the benefit of ads but also pay for the bandwidth they consume the ad delivery network

sundvor 10y ago

My phone plan is 6GB/month. After that, I PAY. And they wonder why we use adblockers.

cobookman 10y ago

Or on the flip side. Data caps will make people care about how much data their site uses. Cause customers will stop using data heavy sites as they have economic insentive not to. Making the web all around faster as no more 10MiB webpages, and making the industry move back to server side rendering.

tedunangst 10y ago
"total data transferred: X MB ($y.zz)" should be part of a browsers standard status bar.
- vinnymac 10y ago
  I know we can easily get some info from chrome dev tools and chrome://net-internals/#bandwith. It would be convenient for users if someone put together an extension that would display this information by default, or alternatively browsers just had a setting we could turn on.
  - phaed 10y ago
    On it
- YSFEJ4SWJUVU6 10y ago
  Perhaps not – I don't think it's the browser makers job to figure out what terms the current connection has (and for the vast majority of workstation desktops there are no such terms, I think; you're getting a pretty rough deal if you pay by the byte on a fixed connection, no matter the price).
morsch 10y ago
It's not easy to discover how much data usage a specific site triggers. Android reports data usage per app, if you dig into the right preference panel, but that's not going to tell you anything about specific websites. I'd like a notification to pop up when the current usage (within, say, a 3 minute window) exceeds my typical usage. That'd make it easier to notice a website started autoplaying a (muted) video, for instance. It's also something that could be done at the operating system level. The operating system could also conceivably show traffic by domain, but that's not ideal; you want traffic by referrer domain, ie. if the LA Times embeds megabytes of images from some random third party server, those should be aggregated under latimes.com and not under some meaningless CDN domain. But the OS probably doesn't have that information. Of course browser vendors could add more detailed reporting. Say a brief in-browser notification ("Page finished downloading 5 MB.") after initial load is complete and repeating if excessive downloading continues. I guess that could annoying really fast, though. A less annoying alternative would be a special page (e.g. "about:data-usage") that has an overview per page in a user-selectable time range. Maybe link to it from the browser start page.
Bahamut 10y ago
Or make use of things like service workers to cache pages locally.
makecheck 10y ago
It has been years since data caps though, and years of obnoxious ads, nothing has changed. The only way in my opinion to fix this is to (a) give phones HARD preferences on maximum data that cannot be exceeded and (b) give web browsers even stricter per-site settings with sane defaults. Then, let the ad companies fight within themselves to figure out how to deal with the fact that none of their content gets through anymore. Force them to innovate into smaller form factors.
- cobookman 10y ago
  It's cause the data caps are fairly high. For example whats 10MiB when you have a 300GiB data cap. But if we change to a utility like model people will start caring about penny pinching their bills
  - craigds 10y ago
    In the US, and on wired connection maybe. In NZ the largest mobile plan I can even find has 5GB. Another one with flat rate for data is $50 for 3GB (so $250 for this 14GB day). If I load a webpage on my mobile data that uses 10MB of data, I want to know about it so I can shut that down (and never come back.)

amluto 10y ago

I'd like to a see, or at least think about, a system in which everyone pays for egress and no one pays for ingress. That way the LA Times is welcome to throw as much crap as they want at you without charging you much. (One issue is that egresss is involved in every request. Some UI work would be needed to make this work well.) Universal egress fees would go a long way to mitigating DDoS, too.

gruez 10y ago
>Universal egress fees would go a long way to mitigating DDoS, too. How? By charging the owners of infected machines?
- amluto 10y ago
  Why not? If you have a water leak, you get a high bill and then you can ask your water agency to refund you if you fix the leak. ISTM being infected and spewing data could be handled similarly. Currently there is effectively no incentive to avoid having a gadget that joins a botnet.
kalefranz 10y ago
How does npmjs.org, pypi.python.org, and rubygems.org survive in this world of universal egress fees?
PhasmaFelis 10y ago
I'm curious what you're thinking of specifically, because I can't imagine any remotely reasonable way to make egress fees work from a technical standpoint, let alone widespread acceptance and adoption. You browse to a site, you can read all the articles you want, but you can't load any other sites until you pay?
- amluto 10y ago
  I'm a bit confused by your question. I'm saying it might be reasonable to charge per byte sent, that's all.