4 ms·

The sneakiest thing just happened to me on that Washington Post site. There was an overlay ad, no big deal, with an X to close it in the upper right of the scr

The sneakiest thing just happened to me on that Washington Post site. There was an overlay ad, no big deal, with an X to close it in the upper right of the screen. I moved my mouse up there to click the X and the site shifted slightly so what I actually clicked was "Subscribe"!

ezekg 9y ago
Relevant commit strip: http://www.commitstrip.com/en/2017/02/16/the-dark-side-of-coding-the-cross http://www.commitstrip.com/en/2017/02/16/the-dark-side-of-co...
glitcher 9y ago
In chrome dev tools, delete the first 2 div's under the <body> tag, then highlight <body> and in the css pane disable the element style for overflow-y: hidden; YMMV depending on platform :)
edejong 9y ago
During careful A/B testing, this site-modification has proven to increase the 'subscribe'-click ratio from 0.002% to 80% Which self-conscious mid-manager would say no to that?
zkms 9y ago
This is why I disable javascript on my laptop and block ads on my iphone. I shouldn't have to chase around advert-closing buttons that's jinking like a high-performance fighter aircraft.
- milcron 9y ago
  If you run Firefox on Android, you can use the NoScript extension there as well. Sadly, Firefox for iOS is just a wrapper around WebKit, so Firefox extensions do not work there.
  - pythonistic 9y ago
    Firefox Reader view has been my savior both on laptops and on my mobile phone. It doesn't work on every page, but where it does, it removes the styling and JavaScript and I get to concentrate on the content.
greenshackle2 9y ago
Couldn't reproduce. Different A/B test group? Or could just be that the site is hot garbage, every time I reload the banner/overlay ad and top bar behave slightly differently, I can see how that could happen but I couldn't get it to do quite what you've seen after like 20 reloads.
0xcde4c3db 9y ago
I see that on sites all the time. I assume it's just the layout shifting because the page retrieves stuff whose dimensions it doesn't know in advance. It used to happen with images all the time; now it happens with XHR.
ams6110 9y ago
In many sites, hitting the ESC key will dismiss those overlays.

ezekg 9y ago

Relevant commit strip: http://www.commitstrip.com/en/2017/02/16/the-dark-side-of-coding-the-cross http://www.commitstrip.com/en/2017/02/16/the-dark-side-of-co...

glitcher 9y ago

In chrome dev tools, delete the first 2 div's under the <body> tag, then highlight <body> and in the css pane disable the element style for overflow-y: hidden; YMMV depending on platform :)

edejong 9y ago

During careful A/B testing, this site-modification has proven to increase the 'subscribe'-click ratio from 0.002% to 80% Which self-conscious mid-manager would say no to that?

zkms 9y ago

This is why I disable javascript on my laptop and block ads on my iphone. I shouldn't have to chase around advert-closing buttons that's jinking like a high-performance fighter aircraft.

milcron 9y ago
If you run Firefox on Android, you can use the NoScript extension there as well. Sadly, Firefox for iOS is just a wrapper around WebKit, so Firefox extensions do not work there.
- pythonistic 9y ago
  Firefox Reader view has been my savior both on laptops and on my mobile phone. It doesn't work on every page, but where it does, it removes the styling and JavaScript and I get to concentrate on the content.

greenshackle2 9y ago

Couldn't reproduce. Different A/B test group? Or could just be that the site is hot garbage, every time I reload the banner/overlay ad and top bar behave slightly differently, I can see how that could happen but I couldn't get it to do quite what you've seen after like 20 reloads.

0xcde4c3db 9y ago

I see that on sites all the time. I assume it's just the layout shifting because the page retrieves stuff whose dimensions it doesn't know in advance. It used to happen with images all the time; now it happens with XHR.

ams6110 9y ago

In many sites, hitting the ESC key will dismiss those overlays.