Keith [10.31.06 10:21 AM]

How about a choice of # of questions? 50 seems like a lot for just messing around at work.

  Ryan Grimm [10.31.06 12:20 PM]

Keith -- I considered adding the ability to select the length of a quiz but that raises a few questions about how these games should be treated in the "Top Scores" section of the site. Feel free to add this idea to the Labs Wiki and we can discuss a solution.



In the mean time, if 50 questions is too many for one sitting you can answer a few questions and resume playing the same game at a later date.

  Chris R [11.01.06 03:47 AM]

I started it, but also found 50 questions too much.

Some nitpicks: I had one Java sample come up and 3 Java books suggested out of 4. Have Java and J2EE been considered different languages?

Also, I had PL/SQL come up a few times. I'm good at recognising this (I was a reviewer for an O'Reilly PL/SQL book!) but PL/SQL doesn't appear in the droplist. Was I supposed to choose SQL?

  Danny Dawson [11.01.06 11:00 AM]

This is an absolutely genious way of engaging the user in an effort to collect metadata. I am especially impressed and inspired by the way that metadata collection is only requested from those users who have already demonstrated knowledge of the topic at hand.

In my niche industry, I have to deal with maintaining a database of tens of thousands of products, and one of our many challenges has always been finding a way to obtain accurate, usable metadata about our products. We've tried soliciting metadata directly from the users, but we've always ended up with low-quality results. We've tried hiring experts in our field, but the cost and time required becomes prohibitive. This is an excellent way to quickly identify the topic-knowledgeable and encourage them to submit accurate data.

Nod.