(Stack Exchange’s system tells me that my question appears to be subjective and is likely to be closed. But judge for yourself if this is the case.)
Do you remember when Google used text captchas only? When you were confronted with them whilst browsing with your normal browser (for example the normal Firefox browser), they were usually pretty easy to solve and did not cost much of your time. But once you were using a Proxy server or a Tor exit node, things could have become very complicated. Some text captchas were downright impossible to solve because it wasn’t even possible for a human to decipher the letters reCAPTCHA wanted you to enter. Unfortunately, I cannot find an example on the Internet right now which shows how they looked like but they consisted of unreadable computer-generated gibberish – and the characters were distorted to utter disfigurement.
Today, things tend to be different since Google uses its Street View pictures for captcha generation. But the problem still remains to be (pretty much) the same.
That is, when browsing with your normal browser, solving a captcha like this should succeed most of the time:
If the squares are numbered as
1 2 3
4 5 6
7 8 9
, I pick 1, 4 and 9.
The next step looks like this
and I pick 1.
After clicking on 1, it looks like this:
The captcha has obviously been solved correct and I click on VERIFY. But what does Google give me as an answer?
This happens more often than not when using Tor. Sometimes I have to solve four or five captchas in order to proceed. It rarely happens during normal browsing.
Therefore, I suppose that reCAPTCHA is lying to Tor users and making it artificially and unnecessarily harder to solve captchas. This has nothing to do with “stronger protection mechanisms”.
Do you think that this assessment is correct? And if it is correct: What can we as Tor users do about it?
