Hacker News Reader

Comment by throwaway122378 | Hacker News Reader

They also do open most links in their own google.com wrapper. Time to break them up

Which is extremely annoying when it’s a pdf or other document that’s set to immediately download; it’s rather tricky to get the actual URL if you want to share it as it won’t ever show up in the address bar.

It's getting harder and harder in general to get the raw source document for everything. Look at image search. You search for an image, and the search results are [not the original] thumbnails. Then you click a thumbnail and it takes you to a larger scaled down [not the original] image. Then when you click that it takes you to some image sharing site slathered with ads and chrome all over the page and another scaled down [not the original] image. If you're lucky, you can use some incantation of right-click and/or View Page Source to get the actual high-resolution original image that you're looking for. Total madness!

When I'm looking for an image, I expect my search engine to return me a URL with .png or .jpg at the end of it.

I use the extension "Don't track me Google" to circumvent this. Converts all links to the direct URL on Google search pages.

https://addons.mozilla.org/en-US/firefox/addon/dont-track-me...

You can usually right-click the file in the downloads panel of the browser and copy its url.

Thanks, somehow hadn't discovered that!

Pet peeve of mine, see comment above.

I just tried it on my phone and for some reason, links are wrapped except when I select "desktop version".

Depending on some conditions that aren't always you get the true link or you get the wrapped link.

My guess is that they have an array of techniques for tracking your clicks (js trickery, "ping" attribute,...) and resort to wrapping only when the other ones are less likely to work.

I think the reason why they did that originally was so HTTP websites would get referrer information from Google search results (HTTPS). Referrer data is dropped when going from HTTPS to HTTP. When Google first started using HTTPS, most of the web was unencrypted, and they wouldn't be able to track the referrer in Google Analytics unless they added a redirector to search results.

Is this still true? The html spec now supports a 'ping' attribute in addition to 'href' for pinging tracking URLs. I'm seeing that all results have the actual URL in 'href'.

Example (Chrome):

  <a href="https://developer.mozilla.org/en/docs/Web/HTML/Element/a" ping="/url?sa=t&amp;source=web&amp;rct=j&amp;url=https://developer.mozilla.org/en/docs/Web/HTML/Element/a&amp;ved=2ahUKEwjmxMHk5MvlAhWVvp4KHS6lCqIQFjAAegQIABAB">

on my firefox, on click, it switches to the tracking url (there's an onmousedown event). It does use the ping attribute on chrome.

They've done this for well over a decade, I remember the transition and there was little outrage.

Before, all the results generated were direct links without any wrappers. Then Google decided they needed to know which of the links people were clicking so they injected a wrapper around every link so they'd get a request from every click.

Sure, it can be used to increase accuracy of search results. It can also be used to taylor data and profiles around individual users at finer and finer granularity.

Does anybody here know, what the currently recommended Firefox Addon for removing these redirects is?

The first few results from the searches all have recent "Doesn't work anymore"-ish comments.

Nokinside linked a couple upthread: https://news.ycombinator.com/item?id=21427367

The first one works for me.

Reply on news.ycombinator.com