On its surface this approach seems like a good replacement for regular captchas, but once you start peeling back the usability covers things get pretty ugly. Below is their first example:
| Seems easy enough. Perhaps too easy? |
Seems like a no-brainer, at least for English speaking users, requiring at most two drag 'n drop operations; however it doesn't seem like it would be all that difficult to circumvent if you were designing a bot to detect and automatically solves these types of word puzzles.
| Hope you can understand and write English! |
| Pretty! |
| Did I get it right? Nope. |
Apparently I wasn't smart enough to solve the puzzle in the first go. What's worse is that I failed to grasp why at the time. So I randomly dragged the tile around until they looked like this:
| Getting Closer |
| Done |
What would have helped, at least in my case, is if all panels had shared the same background color, as it would have made it easier to to jump to the conclusion that this was a single image and not a composite one.
Their next example fared even worse from a usability perspective:
| Seasons |
| Fail! |
Nope. Not right. It took me a second and then I realized that maybe they wanted the tiles arranged in a particular order. Well of course everyone knows that SPRING goes first, what with it being the season of new beginnings, birth, creation, and that sort of thing right?
| Spring First? |
Yes, it was correct. But it was also very wrong, not just in terms of usability, but from a complete lack of cultural awareness . In South American countries like Argentina, where I spent most of my youth and teenage years, January means mid-summer, not winter!
The more user friendly way to allow this puzzle to be solved is to ignore the start order, and rather just ensure that the user has ordered the seasons in their correct sequence, regardless of which one comes "first" in a given calendar year.
So then, is re-arranging blocks a good or bad alternative to standard captchas? I think the answer is that "it depends". If the puzzles are not well thought out then you will end up with something that is at best user unfriendly (requiring multiple tries to get it right), or at worst, culturally offensive to some. Given those challenges I would say no to these sorts of puzzles as an alternative; however in general the concept of dragging something around that only takes a second and doesn't impose a mental tax on the user would be alright I think, but would require a significant investment into the creation of said puzzles to avoid the problems mentioned earlier. Given the challenges involved, I think I would rather keep searching for a more suitable alternative.
*Changed CAPTCHA to lowercase captcha so that it's less like shouting.