XWord Info famously displays lots of stats about NYT crosswords. Our analyze your own puzzle feature has long provided the capability for constructors to see that same data on their puzzles in progress. When you upload a .puz file (your construction software lets you save files in that format) you see more or less what your published puzzle would look like on our site. You’ll see how you compare statistically with published NYT puzzles, and you’ll find other crosswords that might be similar. Each of your answer words is linked to a complete list of all the clues the NYT has ever used for it.
As a bonus, after you upload your puzzle, you can generate a private link for your testers who can try solving it on a web page. It’s a great way to get feedback on your work.
We’ve just completed a massive rewrite of this feature.
What’s new?
Anazlye is now easier to use, more secure, and it gives you more control.
- The Analyze page now lists all the crosswords you have uploaded in the past few months. You can view them at a glance, and quickly see which ones have active solver links.
- Solver links can easily be created or disabled.
- You now upload files by dragging and dropping them onto the Analyze page. Way easier! (We no longer support old browsers that can’t handle drag and drop.)
- We now require you to be logged on to your XWord Info account to upload puzzles. Test solvers do not need an account.
How is security improved?
Previously there was no easy way to remove solver links when they were no longer useful. The Internet remembers everything, and Google has nefarious ways of finding such links. Maybe someone posted it on a website somewhere. Google can even scan your gmail, supposedly looking for ways to provide more relevant ads. However they leak out, those links can now easily be turned off.
Additionally, test-solve pages are now explicitly tagged with a directive telling search engines not to index them.
What’s next?
Any change this large might introduce bugs. Please let us know if you find any.
Hey XWI overlords. I noticed on a puzzle I recently “analyzed” that the cheater squares didn’t quite make sense. Some blocks that were necessary to close off themers were marked as cheaters. Have you heard of this before? Am I just not seeing my grid through the same lens as the software? Would love an explanation if you get time to look into it. Thanks!
Wonderful update! Would it be possible to make a feature request? As a constructor, I’ve always been curious to see how original my cluing is. It’d be neat to be able to see the number of times a clue has been used when using Analyze, similar to the Finder. Thank you so much for all you’ve done so far! Cheers!
Interesting idea. I’ll have to think of a good way to do that, but I’ll add it to our (admittedly long) suggestion list. Seems useful.