What I think we're understanding now is that something that fails 1% of the time, might not fail for just 1% of queries, it might fail for 1% of the users and work perfectly for the rest, and that small set can be really unhappy as a result. This is one small step in the direction of fixing that, but it's an issue we're paying close attention to, thanks in part to the feedback we've seen on HN.
We very much want to avoid systematic failures like this, and if technical queries are a blind spot then please send them over whenever you see them. My email is in my profile and I'll make sure they get routed appropriately, and I tend to watch threads about Google trying to get good examples to debug.
https://market.android.com/search?q=yore&so=1&c=apps
I appreciate that Google thinks I'm an illiterate chimp who doesn't know how to spell "your", but really truly, I was looking for a game with "yore" in the title. That's why I typed it. It's even an English word.
Excellent insight. Linux kernel developers have had a similar experience with the use of the likely() and unlikely() hints on conditionals; it took a while to realize that unlikely() has to mean unlikely for any user, not just "unlikely for most users, and likely for the rest". So, for example, it doesn't make sense to write "if (unlikely(feature_disabled))", because anyone with the feature disabled will meet that condition every time.
I've definitely had the experience that Google's automatic correction does the wrong thing for me more often than not; I almost always end up clicking the link to go to the search results I actually asked for.
However, if the computer guesses wrong, that will cause significantly more frustration to the user. This disproportionate amount of annoyance means that the computer must guess correctly far more often than it guesses wrong. Unfortunately, Google's current system frequently guesses wrong, and then compounds those incorrect guesses by using them by default, taunting the user with their original search.