I don't know how this can be fixed. But I'm going to tell you what happened anyway.

Here is my Chrome history. I did not want to visit Jim Crow laws. I did not want to visit Special pages. But because my taps were interpreted after a different page loaded, they fell upon those buttons.

My goal was to look up Martin Luther King in Google and then from Google visit the Wikipedia page about him, and then from that, visit the Chinese Wikipedia page about him.

But as you see to visit those two Wikipedia pages I ended up visiting two extra pages. And then there's repeats, where I had to go back to get back to the page I wanted.

Perhaps due to various slownesses, when I pressed something, several seconds later the page finished loading, and then ... only then my tap was interpreted. But it was interpreted upon a entirely different page. So landed on a different hyperlink.

Yes, you might say, "Mr Jacobson, just wait for every page to fully load before you tap on anything."

Well that's not that easy for users too be fully aware of in their mobile browsers. Also that would not be the way young people would use their cell phones. They wouldn't just sit there, and after everything stops moving wait an extra 30 seconds to be sure. (Rendering might still be going on below the viewport.)

I sort of think this problem is particular to Wikipedia because Wikipedia seems to be a rather slow website. Or maybe it's a Chrome problem.

But you might also need a slow network and slow cell phone to reproduce it.

Or maybe a fast cell phone, because other sites load more quickly and when you tap a link you're guaranteed to trigger the link you thought you tapped. So people get used to that until they run into Wikipedia and it's slowness. But slowness is not a crime. What is a crime is users tap things and then the tap gets interpreted on not what they saw, but the same position (screen coordinates) on what loads later. Upon which it is evaluated.

Another solution, painful as it might be, is to cover the screen with gray, until it is fully loaded thus preventing users from thinking that they're can start tapping links.

Anyway sorry for not using the official bug report or feature request format. But this is a rather long story.

In summary I'm saying there's various stages, where the user sees a hyperlink and taps it, only to discover they end up somewhere entirely different.

Because their tap, say 33 mm from the left, and 43 mm from the top or whatever of their screen, ended up being interpreted by the system, to be the same position, but on a different page .... unfortunately often landing just by accident upon a entirely different hyperlink.

That's because they perhaps tapped too many times, because something was unresponsive etc.

Anyway, think of an ATM machine. Let's say it's asking you questions, press one for yes, press two for no. But because of some slowness in the system you end up giving an answer to a previous page. Thus there's a 50% chance you'll give the wrong answer. You get the idea. Thanks.