meta DNF
hello and welcome to episode #685 of matt gaffney’s weekly crossword contest, “Get Out Your Wite-Out”. after a guest puzzle by malaika handa last week, matt is back in the saddle with this week 3 puzzle, for which the instructions tell us that we are looking for a 6-letter word. what are the theme answers?
well, as has been the case distressingly often in recent weeks, i’m once again blogging a puzzle that i haven’t figured out. but i do know the six obvious theme answers in this 15×21 grid, which are all ten-letter across answers that are goofy made-up phrases:
- {Loaves that Native Americans only eat at celebrations? (4)} PARTY PONES.
- {Where a kangaroo keeps her five dozen babies? (5)} SIXTY POUCH.
- {Warsaw residents upset at a soccer loss? (7)} SALTY POLES.
- {Feel bad for someone after their surgery? (7)} PITY POST-OP.
- {Like some large leads in NBA games? (5)} FORTY-POINT.
- {What the CEO of Hormel wields? (7)} MEATY POWER.
i had two immediate observations: first, these goofy phrases all contain TYPO as a substring across a word break. in the context of the title, that suggests removal of those letters. second, the parenthetical numbers in the clues are 4, 5, 7, 7, 5, 7. that’s not an ordering, so i’m guessing those are lengths of other words (perhaps other entries in the grid, or perhaps words that aren’t in the grid at all) that are related somehow to the result of doing something about these TYPOs.
less directly related to those six theme answers, i couldn’t help but notice (and i apologize that i can’t help but bring up) that the fill in this puzzle is very creaky. matt has often skillfully woven six medium-length theme answers into a 15×15 grid with nary a hiccup, but here he’s given himself six extra rows in the grid to smooth things out, but there is still a giant amount of fill that made my nose crinkle. i won’t go through it all, but just for starters, there are no fewer than ten (!) partial phrases in the fill; one or two is normally about as much as i would tolerate without raising an eyebrow. even the corners that seem to be unconstrained by the six theme answers are very clunky, so there is definitely some extra theme material hidden in the grid. in fact, i find it relatively likely that there are more than six extra theme answers, because just six doesn’t seem like it would be enough to explain why the grid is so bad.
okay, so about the wite-out and the TYPOs. there are two obvious things we could do: wite out just the TYPOs themselves, or wite out all the instances of those four letters in the grid (which is a lot!). going with the first idea first, we would get PARNES (4), SIXUCH (5), SALLES (7), PISTOP (7), FORINT (5), and MEAWER (7). some of those look like they could be something, or be turned into something with one more small change (e.g. PISTOP -> PISTOL or MEAWER -> MEOWER), but then you also have SIXUCH, which, uh, not so much.
what about other entries in the grid? three of the parentheticals are (7), so it makes sense to look at the seven-letter entries, of which there are eight in the grid:
- {Plane’s place} AIRPORT.
- {Witty barb} RIPOSTE.
- {“Give that to me right now!”} I NEED IT.
- {Dark beers} PORTERS.
- {Far-flung base} OUTPOST.
- {Screen-to-printer sameness, in an acronym} WYSIWYG. “what you see is what you get”. those of you who don’t remember word processing in the 1990s, this used to be a revolutionary idea. now it’s de rigueur.
- {Busy periods} UPTIMES.
- {It’s worth about $55 per share} GM STOCK. what a weird and arbitrary entry! i couldn’t help but notice that this could easily have been the familiar phrase IN STOCK or (with a slight reworking of the surrounding fill) the common word RESTOCK. so this does seem like a deliberate choice.
it definitely caught my attention that several of these words have many of the letters of TYPO: AIRPORT, RIPOSTE, and PORTERS have three each, and OUTPOST has five total (three unique). i still don’t really see how to connect any of them to the theme answers, though.
oh hey, here’s an idea: all of the parenthetical numbers are between 4 and 7, which matches up with the positions where the letters of TYPO are in their respective ten-letter theme answers. what if we look at the down answer that crosses at the indicated place?
- {Far-flung base} OU(T)POST.
- {Screen-to-printer sameness, in an acronym} WYSIW(Y)G.
- {:, in math class} IS T(O).
- aaaaand here’s where this idea comes to a screeching halt, as the location of TYPO within PITY POST-OP is from 3 to 6 instead, so that (7) is in the wrong place. oh well, i wasn’t impressed at where this was going anyway.
so i’m back to nothing. and now the deadline is upon us, so i’m going to throw in the towel. i’ll toss out a hail mary guess of DELETE, which is a six-letter word that may be related to the mechanism, but i’m not optimistic.
let’s hear in the comments what i should have figured out!
Replace the letters TYPO with the four letter entry on the same row, giving:
PARIS LANES (Rues)
SIX AND SUCH (Evens)
SALOON ALES (Porters)
PILOT’S STOP (Airport)
FORMAL AIN’T (Is not)
MEAN ANSWER (Riposte)
The resulting phrases clue another grid entry (given above). Take the first letters to get REPAIR, your answer.
Put the row-mate entries in place of the TYPOs and look for synonyms in the grid.
Thanks Joon! Here it is:
Wite-out the six TYPOs and replace them with the four-letter word on their row. This forms a new crossword clue that answers another entry:
PAR(IS LA)NES = RUES
SIX (AND S)UCH = EVENS
SAL(OON A)LES = PORTERS
PI(LOT’S) STOP = AIRPORT
FOR(MAL A)IN’T = IS NOT
MEA(N ANS)WER = RIPOSTE
First letters of those spell contest answer REPAIR.
Putting the numbers in makes that second step difficult to pick up on. I was looking for words in the grid with that many letters to put in the place of TYPO. It probably would’ve been better without the numbers. The next step is much more inferable. Great meta which explains the grid fill constraints.
+1!
We did not need the numbers to get this, and IMHO they were a terrible distraction.
I agree that they were a distraction.
But without the “7” in 89a, “Mean answer” would have been ambiguous, with SASS being at least as good as RIPOSTE.
I agree that the numbers might have been a distraction if I had noticed them, which I didn’t
249 right answers, i meant to add
MEAN ANSWER isn’t a very good definition for “riposte”, IMHO. I think I’d personally have preferred WIT’S ANSWER as being more accurate, despite the duplication of WIT and WITTY*, although I expect you probably go the other way on that judgment call.
*assuming you’d be able to fill the grid around WITTY at all, which isn’t a given, to be sure.
My sister had the Crissy doll (https://en.m.wikipedia.org/wiki/Crissy), so that’s what I confidently wrote at 74d, and refused to erase for the longest time even though the grid was screaming at me.
I had the Crissy doll when I was five. My sister had the Velvet. Fond memories.
Thought this was tougher than usual for a Week 3, and startled my wife when I literally yelled, “Aha!” when I got it. Also trying for some reverse jinx ju-ju here, as the last couple times I’ve commented that a Week 3 seemed easier than usual I completely whiffed on the subsequent weeks . The fill didn’t strike me as all that clunky, enjoyed the goofy themers and the clever meta.
Tough crowd today, hm
Like Joon, I am getting used to defeats in week 3.
Good meta, just did not see it. While the parentheticals did not help, I don’t think it would have helped if they were not there.
Once again, the title did not help me. I understood the TYPO right away. At first I thought the obvious lack of an ‘H’ in the title (WITE) meant something when I found there was only one ‘H’ in the grid. But when it appeared in a theme answer, that went nowhere.
The long grid, while clearly needed to fit the themers. had me looking for something that was not there also.
Congrats to those who got it — can’t accuse me of a NASTY POST :)
I liked it a lot! I spent an awfully long time trying to slot in random 4-letter words from the grid before I decided to look for one that ended in ANS which would go with the WER I already had sitting there from MEA WER. When I saw it was NANS on the same line, I was off to the races. The hard thing for me was realizing the substitution would make multiple words, I kept thinking that I’d be plugging in something to make a single word, after which I’d identify the letter in the numbered slot to make the meta answer. After I wrote down PARIS LANES the appropriate RUES to match jumped right out at me since I’d already been looking for that sort of thing (spent quite a while trying to figure out how to substitute AMIIBO for the NES, haha.) Oh, and let’s not forget the crucial decision to print out the puzzle so I could scribble on it. I wasn’t getting anywhere at all just staring at the completed grid on the screen.
“Mutman” said
Like Joon, I am getting used to defeats in week 3.
And I agree.
I made this more difficult than it was, and only solved with some help. Swapping the TYPO for four letters right next to the theme answers? Nope, I made new words with the remaining letters, trying to match the numbers in parentheses. If it hadn’t worked for two of the answers I may have given up. Once I realized the mechanism I felt like a fool, but I spent three days on my foolishness.
Like Joon, I’ve had trouble with week 3s. And 2s. My performance has been PRETTYPOOR. Great puzzle, though.
Can someone explain all the 1/1.5-star ratings? I don’t see any comments attached to them so assuming just sour grapes, but genuinely curious why the sour grapes would be so strong this week. Also didn’t receive any comments with submissions or e-mails that mentioned anything. Lowish number of right answers for a Week 3/5 but not that low.
No idea. I, for one, thought the meta was brilliant. Even though I had to be almost dragged to the finish line.
i’m always grateful to you for these puzzles, which are usually magnificent. and i know from personal experience how much it smarts to put your work out there and then have people on the internet crap all over it. and none of those 1-star ratings belong to me, but since you asked, i can tell you the things that were unsatisfying to me about this puzzle after reading about how it worked:
1. why TYPO? replacing four letters from a theme answer with a neighboring four-letter entry is a neat idea for a meta mechanism, but it does not seem to me that TYPO describes it well at all. an extra letter, a missing letter, a wrong letter (especially from an adjacent key), consecutive letters transposed—these are all TYPOs. four wrong letters in a row? that’s just not what a TYPO is. and yet the puzzle clearly wants us to notice TYPO as a jumping-off point. i jumped in the wrong direction and felt like i never had a chance. reading about the correct solution after the fact failed to elicit the desired “i should’ve seen that!”.
2. i mentioned this part in the blog post because it was a very salient feature of my solving experience, but the fill was quite noteworthy for being really bad. i was correct that there were more than six extra themers (there were twelve), but it still felt like the grid was too strained. i know filling in the grid is just the appetizer, so to speak, but this one was unpleasant enough that i was not really in the mood for the main course.
oh well. i’m extremely confident (as someone who has blogged 600+ of these) that the next puzzle will be more to my liking, and i’m at least hopeful (as somebody who has solved 600+ of these) that i can bring my own meta-solving A game next week.
Not a negative rater, and I did manage a breakthrough on the solution last night, so I may not have the right perspective. But I wondered this morning if something like “Autocorrect” in the title would have helped as a nudge towards the mechanism of replacing four letters (representing a TYPO) with four other letters (representing the intended words). That step, and particularly the use of the other entries on the matching rows, is the part that perhaps could have used a hint or thematic underpinning?
I started off thinking about treating the TYPO as a single letter/unit, to either remove or replace somehow, which felt promising based on the Wite-Out hint. For some reason, my brain thinks Auto-Correct is slightly closer to the mechanism of the four-letter swaps, but it’s always hard to predict how this stuff plays out, so that instinct may be totally wrong.
Well I appreciate your thoughts of course, Joon, and thank you for posting them.
On 1), well, artistic meta license covers this to my mind. If the TYPO is four letters long then it makes sense to put four letters back in there, and the most natural four letters would be the ones on the same row as the entry.
On 2), well, lots of theme, so subpar fill like that happens a lot in metas. Of course we try to minimize it.
I don’t get the negativity at all regarding the fill. It’s actually helpful solving the meta when you have funky short fill. Like the NANS/ENER crossing… Any experienced solver would instantly recognize that as a questionable crossing and (should) realize that it has to be theme-related. A W or V in that crossing would be a huge improvement for a standard puzzle, so it’s fairly obvious that you needed the N there for your theme to work.
FWIW this was one of my favorites all year. And I think people are overthinking the whole TYPO thing. It was pretty intuitive that TYPO would be replaced by something, especially given the title. I’ll agree with others who say that the parenthetical numbers weren’t completely necessary, but I don’t think it did anything to take away from the puzzle either.
FWIW, I loved it. Very difficult though — that “same row” trick has flummoxed me over and over. I don’t find it at all natural or intuitive, so it always plays more like a week or two harder than you peg it. I’d guess that some of the low ratings reflect the mismatch of week 3/5 expectations vs. reality for those folks who also find same-row secondary themers to be non-obvious.
I typically don’t rate the puzzles, but I gave a sincere 4.5 for the effort. I thought it was very good.
My question is do you have to be a privileged member to see the ratings? I could only rate I did not see any existing ratings.
Put your cursor over the ratings and you will see how everyone voted.
I have been doing these metas as long as Joon and have never rated them below 4, even the one that had the meta in the instructions. Very low ratings are sour grapes in my opinion.
I also thing the quantity of solvers has significant inflation over the past couple of years as the “other site” has lots of group and hints used for solving that weren’t in play beforehand. 250 solves today feels like a 140-160 solve previously.
“Very low ratings are sour grapes in my opinion.”
I wouldn’t jump to that conclusion. It’s always dangerous to draw conclusions about what someone else thinks or feels and I can imagine plenty of alternative explanations for low ratings, especially given the lack of universal rubric for the subjective meanings of the various points on the scale.
I don’t ever add ratings — coming here for the post-game is my way of interacting:)
I was a DNF, and had the “how’d I miss that!” reaction. Usually that’s because I’m impressed even if I missed it — but I’ll admit this time I was a bit annoyed. The logic chain of mechanisms and clues seems off to me.
My solve was extremely similar to Joon’s. Saw the TYPOs, and then considered every way that the parenthetical numbers could be applied to the TYPOs in each entry. Failing to find a 5-letter grid entry that made a new phrase out of SIX()UCH left me assuming I was totally missing something and I never got back on track.
I feel like your use of parenthetical numbers is usually directly related to the clue in question. In this case, the parenthetical is related to the step you take AFTER the first step is done to the TYPO answers. I would have been far more likely to go on a 4-letter-entry hunt to replace the TYPOs had those parentheticals not been there. Since I was looking around for 4,5, and 7 letter entries, I never even noticed that each theme entry was paired with only one 4-letter in the same row. With the parentheticals, I literally never considered a direct 4-letter grid replacement — in fact I tossed the idea right away and never went back.
Had I seen that and found the replacements, the resulting phrases strongly imply the ‘use the new phrase to find another grid answer that fits’ mechanism. I’m not sure I would’ve needed any other clues.
That said, I’d guess a version of this puzzle with no parentheticals plays more like a week 4 — and a version with more direct/obvious hints plays like a week 2! Dialing in the challenge level is among the many things that constantly impress me about your puzzle making.
So that’s my two cents — which is obviously all with the benefit of hindsight:)
Nitpicks or not, the MGWCC is my favorite puzzle experience every week.
Cheers:)
Amen to that Matt.
I’m still hoping that the star ratings will be abandoned for good, but in the meantime, I’ll just continue to remind people that the star ratings are meaningless and should never be used to judge the quality of any puzzle.
I did it on my phone and the clues didn’t have the parenthetical numbers. I didn’t get the meta and I don’t think the numbers would have helped me anyway as I still don’t understand what they mean.
Enjoyed the puzzle even though I spent way too much time trying to figure it out.
The numbers are the length of the grid entries that are the answer to the associated themer which becomes a clue to the answer
I thought this one was incredibly clever, and am constantly amazed at how Matt comes up with new and innovative mechanisms each week.
I saw the “TYPO”s early on, thought for a while that I needed to replace them with a single letter and that the enumeration was pointing me to which letter to take from that new word. When that didn’t work I thought about the 4 letter words. and listed them all to try to see if any worked. There were so many! But had a big smile when I realized that it was using the one in the same row – that made it incredibly elegant and smooth sailing from that point on.
5 stars from me.
Same solving track here. Also agree with the comment down below about us geezers that remember using Wite-Out (that is the actual brand name) to cover the entire typo. Although I was a Liquid Paper guy.
My submission comment was “home run”.
I got flummoxed by a couple red herrings. The first being the fact that “Warsaw” appeared in two clues (and one of them a themer). The other, most evil, one was the fact that when you remove the TYPOS one of the remaining sets of letters was NES. And one of the clues was “Nintendo operating system.” Anyone else get stuck on that?
Yep, right there with you on that ones. I was convinced that the mechanism was “TYPO is announcing that everything prior was to be stricken, now replace it with the right alternate answer for an existing clue”. PARTYPONES minus the PARTYPO plus S gives SNES, a Nintendo system. PITSTOPS can kind of be made in the same way, though PI was already there and not a typo. Luckily I was talked down from that by a friendly member of the muggles forum and didn’t waste hours trying to force that approach on the other 4.
Yeah, I was totally convinced that I was onto the second step with NES -> Nintendo platform , especially because AMIIBO was so obscure. Spent quite a while on that, but the SIX convinced me that that couldn’t be the path and wound up getting it in the end.
Ah dangit, that was gettable. I’d had PARIS LANES but didn’t realize it was a same-row substitution. I thought the parenthetical numbers might be how many letters replaced TYPO, but didn’t get it to work with the others so I dismissed it as nonsense and gave up :(
Didn’t get this (and I’m kicking myself over it) but I think it’s a very clever construction. I figured something was up with PORTERS since that corner would be much cleaner with MORTARS crossing MTNS and AWRY, but that obviously wasn’t enough to put me over the edge.
Wow, this is such a brilliant meta and I’m soooo bummed I didn’t get it.
En route to the satisfying aha moment in solving this one, I couldn’t help fixating on the grid entries AIRPORT and PORTERS (perhaps due to their symmetric placement) and mentally overlapping them as AIRportERS, with the ends coming together as AIRERS, which sounds so much like ERRORS (aka typos). That may have been the awkward stepping stone for me to focus on only 4-letter replacements for TYPO in each theme entry (given that I had been considering replacements of all sizes up to that point), but finding that little gem made me chuckle and appreciate this puzzle that much more. Thanks again, Matt!
Just popping on to say that I really enjoyed the meta and thought it had a very nice “AHA!” moment. I agreed that some of the fill was a bit clunky…but whenever I see that in Matt’s puzzles, it always indicates a some serious meta constraints. In this puzzle, it wasn’t just the volume of meta-related content, but also the fact that basically there was an entire row of it every 3 rows. Also, some of the more crosswordese answers (OONA, NANS) were actually meta-related. My vote is that the low reviews are sour grapes, as there was nothing “1-star” about this puzzle!
In addition, I thought TYPO was a perfectly reasonable marker for an error, and that the title was excellent (autocorrect or delete wouldn’t have been quite as accurate). Imagine you are typing on a typewriter or writing long hand (like us “olds” used to do”). You make a mistake and reach for your Wite-Out. Even if your error is only one letter, you almost always had to cover up a few letters, so you Wite-Out 4, and replace them with 4 new ones! Maybe there are just too many young folks out there used to using one of those new-fangled machines, where you just hit delete and fix the one letter…
In sum, I thought this was a great concept, well-executed!
Matt I thought it was outstanding and while I don’t often get week 3 this one came naturally. Can’t please everyone as you well know. It was fun and had a nice aha.
I’ll join the “loved it” crowd. Definitely helps that it came to me quickly for a week three, but every step made perfect sense, and I was very impressed at the construction. Great one Matt!
I also DNF and almost threw ERASER or variants of that as a Hail Mary, but decided no answer was more legitimate.
The fill – the TYPOs – hit me early on and I thought it was magnificent (even given others’ comments about the fill), but the numbers blew me away. I was looking for 4, 5, 7 (etc) letter answers in the grid which might take the place of those TYPOs, did not consider the 4 letter words on the same lines.
Well worth the 75 cents of entertainment.