Aside; I hope some day/time in the future that AI can recover, from Wikipedia article history, in some form, all the useful info that has been removed from Wikipedia by deletionists, or whatever they're called more recently, or even just through bad copyediting/merges.
Great concept, but the obscurity of the terminology at times will make this....trying. Plus, it does not appear to validate words nor automatically fill out variants. For me, this was fun until I got past guess 200 or so.
There's a setting (annoyingly non-persistent) to make it try out plurals and some variants, but it's buggy.
This could be so much better with a few fixes and tweaks, but the author appears to have abandoned it immediately after launch. Such a shame, it's really quite a challenge.
Have been playing this for a week or two now, landing a few perfect scores, and it turns out that the guesses that will get you far are words like "defined", "known" etc., and some words you'd expect to be unredacted from the get go, like "can", "be", "they". Another thing is structure of the articles, mostly parentheses. Two-character words? Date and place of birth (and maybe death). Colon after first word? The name of another language (mostly). Try "named" and you might have another hint about the etymology. Look for subscripts (superscripts are out with the rest of TeX)
...given, there was a bad day to not know the word "injunction", so there's that.
12 comments
[ 5.1 ms ] story [ 37.0 ms ] threadThis could be so much better with a few fixes and tweaks, but the author appears to have abandoned it immediately after launch. Such a shame, it's really quite a challenge.
...given, there was a bad day to not know the word "injunction", so there's that.