Listen With Others

Are you sitting comfortably? Then we’ll begin

Archive for Sep, 2009

4050: Recipe, a Setter’s Blog by Parsnip

Posted by Listen With Others on 25 Sep 2009

Having made a first attempt at setting with the Magpie I decided to grit my teeth and make an attempt at scaling the walls of the Listener. As someone new to whole setting game, this being only my second puzzle, I’ve discovered that I thoroughly enjoy the rather different discipline of writing rather than solving clues, and also the noodling about with themes, but I find grids absurdly difficult and headachey. For boring technical reasons I don’t have a proper version of Sympathy, but in any case I suspect my problems are as much to do with not being very good at composing grids and filling them as they are to do with electronic tools. Maybe I’ll improve – but those issues with grid setting and filling caused problem number one in my journey to publication.

Perhaps inevitably, with three small children, when I was vaguely checking things that passed through my mind as potential themes, the ‘what are boys and girls made of’ rhyme stuck and I wondered if you could quite literally use the ingredients of that verse, taking the letters from slugs and snails to make names and so on. Interestingly, I have only ever known the verse as ‘slugs and snails’ and never thought to check for alternative versions – had I done so, I probably would have settled for the ODQ version for safety’s sake – and that was problem number two.

Matching 13 letters in ALL THINGS NICE and PUPPY DOG TAILS meant that I could have those as unclued ingredients in the grid, along with the unclued names, which refined very quickly how the rest of the puzzle would unfold. Choices were rather more limited for slugs and snails than they were for sugar and spice but in both cases I managed to find a group of names that worked relatively quickly with the first set of ingredients I chose (which means that it’s probably not actually as hard as it seems) and I enjoyed the fact that although the theme might drop before the end you needed to solve all the ingredients in order to remove ambiguities. But again, despite having found the names more easily than I had originally expected, I settled for the first group I found. Problem number three.

The title had been obvious from the start, but the idea of setting food-related clues only came to me once I’d already made my way through the list of ingredients, by which time I’d come up with the Aston Villa/VANILLA link which I felt very pleased about and was rather reluctant to rewrite. But I did like the misdirection and thematic constraint of foody clues, so I justified the non-ingredients as separate and rewrote the few other clues I’d already done. With hindsight, I rather wish I’d rewritten the lot, just for neatness’ sake, but it’s not a big deal.

Potential problem number 4 then popped up when I had almost finished setting all the clues – Aedites’ Babes puzzle, based on the same idea. The penny had dropped fairly quickly for me on that puzzle and I was at first completely distraught. As I got through it, I realised my exploration of the verse was very different from Aedites and there was a possibility it would be OK, especially with a decent time lag (as proved to be the case) and in any case, I wasn’t going to make the editors’ decision for them on that front, so I continued with the puzzle, thankfully.

The grid, as it originally appeared was this:

L4050 Recipe by Parsnip

On to test solving. My test solver set to it manfully and I received some very positive feedback, although they did rather struggle with the ingredients – and this was the point at which I discovered that not only was the version of the verse with which I had grown up not the only version, but it seemed from the references to be one of the more unusual versions. My tester went first for ‘snips’, then for ‘frogs’. Unable to face rewriting the whole puzzle, I stuck to my guns and merely changed the preamble to allow for alternative versions of the recipe.

I then sent a tentative submission off to the Listener and to my delight received an incredibly rapid response indicating broad approval of the idea and the general standard of the clues (woohoo!) and some rather worrying sticking points. The Babes puzzle was raised but then dismissed, as I had hoped. Similarly, there was less concern about the alternative versions. But my grid was deemed to be substandard (low average light length and unjustifiable double unching on the two 13 letter entries) and more worryingly still, I had made a stupid mistake with the names. The day that I had been noodling about with the anagrams I had with me the 90s version of Chambers, not 2003 (the 2008 hadn’t come out when I was setting it) and in there DOD exists as a short form for George. It wasn’t a particularly satisfactory name, given that DODMAN was one of the ingredients, but I wasn’t about to complain. Needless to say, Dod no longer appears in the list of shortenings for George in Chambers. A lesson for the future, to use proper references and not rely on old dictionaries (which means I must update my ODQ if I’m going to set anything based on a quotation…).

John helpfully offered the suggestion that the idea would probably find favour and publication with a better grid (even with the Dod problem, as there was at least a reference somewhere for it) and at this point I’m incredibly grateful for his help in suggesting alternatives, getting the puzzle to the shape it’s in now, including an improved list of boy’s names without Dod. Of course, while we had kept the original list of ingredients, everything else had changed so I needed to write an entirely new list of clues, if I could be bothered. I had been pretty pleased with a lot of those clues (maybe if continue setting eventually I can find room for them) but I was keen to break into the ranks of Listener setters and I had been given too much encouragement from very helpful people to back out now. So I set to in my painfully slow way and many weeks, or probably even months later, I had found enough quiet moments to come up with a further bunch of clues and sent it back to the Listener.

At this point, I was properly in the queue and the wait was a few months, but frankly every week that passed took us further from Babes by Aedites which could only be a good thing. A few weeks ago I received the very welcome communication that Recipe had made it through, and shortly after that, the proofs arrived with the very thorough notes on the clueing, together with some deft rewrites very sensitively done. As a newbie, I felt very chuffed to have only a very few total rethinks but even with that, the puzzle certainly improved no end thanks to their work, for which I’m most grateful.

With further hindsight, I’m not happy with the clue for GENA – I really struggled with finding a food related surface that didn’t have a very obvious definition and in the end fixated on the collision of GAUR and the potential of ‘ox cheek’ which really constrained where I could go. I have to admit to not having thought of alternatives since, but it really is very contrived and I can’t think why I felt the need to add the word ‘conventional’ – I must have thought it was necessary for the definition of ‘ur’ but I can’t find that in the 2003 Chambers or anywhere else. There are a couple of others that stretch the constraints, but that’s my least favourite.

Finally, along with thanking the editors I have to express my gratitude first to John for the extraordinary feedback he provides – both in its breadth and its detail – and to all the solvers who take the time to write feedback in such detail themselves. It not only encourages me to set more, and as much as possible, but also to feed back myself more on other puzzles, given the pleasure it brings (and that includes the querying of many clues, which can only make me determined to write more watertight wordplay in the future).

Posted in Setting Blogs | 1 Comment »

Recipe by Parsnip (All things nice, indeed!)

Posted by shirleycurran on 25 Sep 2009

We can understand Listenercoholism; the sheer ecstasy of having an almost complete grid on Friday evening. Words went in at a steady pace and we were hooked. The junior 8X8 team can cope with straight-forward clues like Parsnip’s, that led to answers that fitted together so neatly in a helpful grid. (Yes, I can anticipate those experts on the message board muttering that they had a long empty weekend looming ahead, but spare a thought for the neophytes!)

Filling the grid did give us some problems. DARWIN was the obvious answer for 15d, ‘Scientist’s book with one recipe initially contained’, especially since we already had ORIGIN at 30d (Oliver’s first set up at home: Fifteen’s work securing this start) but the wordplay tested our skills. We finally understood that DAN was the book, (W)ith one (I ) and (R)ecipe going in initially (thus RWI) – but we do, so often, have to work backwards from putative solution to wordplay!

Grannie must have been a bit of a sexist, as she used to quote that rhyme at her grandchildren; ‘What are little girls made of? Of sugar and spice and all things nice, that’s what little girls are made of.  What are little boys made of? Of slugs and snails and puppydog tails, that’s what little boys are made of’ (Not very kind to my poor little brother but we girls were very smug about our ingredients! And it was ‘slugs’ in those days, none of the ‘snips’ or ‘frogs’ that Wikipedia produced.)  Some of those ingredients (rather ambiguously ‘the last two’ in the preamble) appeared on the top and bottom lines of the grid, fairly early on, so we were obviously meant to put little girls and boys into the remaining unclued lights (‘little’? We expected diminutives until Reginald and Melissa banished that delusion).

The ingredients with only wordplay seemed to have no function at this stage. We had seen ANISE (One found in parish minister’s house with no roof – I in (M)ANSE), SUCROSE, NERITE and VANILLA. To our simple minds, these were just confirmation of the ingredients. Ah, the team was in for a tumble!

REGINALD had already appeared, with SINE, VERA, UNA and a probable CLARE , PETE, MELISSA, MARTIN and DON (we eliminated RON as he wasn’t in the Chambers appendix as a short form of Ronald). This was fabulous – after about three hours work we had only two cells to fill. We slept on it.

Morning light revealed the complexity of the endgame. Were we going to put MAT, MAX or MAY? Should it be CORA, DORA or NORA? How could we ever have imagined that a Listener Crossword was going to be problem-free? But was it possible that we had a 51-letter anagram to sort out in those ingredients, in order to find which two letters went into our last two cells? Oh dear, yes!

We needed one more sugar and guessed at DEMERARA (from DEME +? – Denis, we need your help again!) and we needed a couple of  slugs and a snail. LIMAX would give us the X of MAX (IMA ino LX – but how are we ever going to learn all that ‘wordplay speak’?), so we had 12 of the thirteen letters of our last two ingredients and one of them had to begin with a T.

Bradford’s Crossword Solver’s Dictionary reminded us of the existence of DODMAN, so we triumphantly worked our final letters (PTEGNRA) into TREPANG and gave a hearty sigh of relief. DORA was the culprit.

All in all, this was a fine test for us. The initial phase was rewarding and the final stage challenging – an ideal recipe. Thank you, Parsnip!

Posted in Solving Blogs | Tagged: | 2 Comments »

4049 The Domino Effect by Googly: Double trouble

Posted by shirleycurran on 18 Sep 2009

Six A4 sheets of calculation and 14 post-its, crammed with figures on both sides – and only half of the 8X8 team was in the running for this one (with the usual Domino Effectwise help). It took us three and a half days to get those cut out dominoes into place.

There were red herrings stacked up in smelly boxes. The first was a conviction that the dominoes were going to respect  clue division lines – but we are learning that it is silly to assume anything that isn’t said in a Listener crossword.

It was clear that our answers had to be the product of those letters and not just a concatenation of the numbers they represented. However, we were soon coming up against impossibilities. Trying to fit a domino into 2d led us into a set of dominoes with leading zeros down in the south-east corner.

EGG at 3ac produced a very large red herring. Towards Sunday midnight, Mr Math forgot that he was multiplying by the 2 of E that had already been established. The error went unnoticed until the very end when identical values emerged for one number. What a nightmare!

The struggle continued until, at last, we were able to address those numbers that appeared only once in the clues, X, I, Q and T. By now we were able to overlay dominoes to suggest the missing numbers and, at last, we were able to find a value for 9d and produce three numbers for the phrase at 8, 31, 18. It was fairly obvious that that had to be factorised and that something about dominoes would appear. Remembering the pentominoes, we looked for extra Cs (C = 1) and managed to find yet another red herring (ONCE at 18ac). I expected ‘All fall down’ or some such domino effect, or, at least ‘Paint all the doubles in zebra stripes’ but it was not to be. SHADE DOUBLE ONE emerged. So, I was given the opportunity to colour at least that little chap in the south-east corner.

Well, thank you Googly. What a challenge! I, for one, am glad there’s a three month gap before the next numerical one.

Posted in Solving Blogs | Tagged: , | 5 Comments »

Listener 4049: The Domino Effect by Googly

Posted by erwinch on 18 Sep 2009

I thought Googly a well established numerical setter so was somewhat surprised to see that this was only his fifth Listener since 1995.  My records show that I had successfully solved the previous four with the most memorable being Verity Hill’s Dice-Box (No.3737) in 2003.  I remember being sidetracked by a poet of that name, which only goes to show the power of the Internet to mislead.
Well, a return to polyominoes after just six months with the domino, the shape that inspired the nomenclature since do- was deemed close enough to the Greek for two.  However, not an inspiring shape so this is going to be a very different puzzle from Pentomino Factory.  The preamble gave nothing away as to the positioning of the dominoes – would they be placed in a symmetrical pattern, would adjoining numbers match, etc?  It is generally fatal to make assumptions with these things and the only thing that we knew for certain was that the double zero would appear in adjacent blank cells – any one of dozens of positions in the grid.
For the opening I considered OO (2dn) and O + OH (1ac).  The latter is a multiple of O and its final digit must be able to match the first of OO: O = 7, 8 or 9.  Only 64 consists of two domino numbers but make no assumptions!
On to DODO (6ac): D = 2, 3 or 4.
Then CO + OP (11ac) and COP (29ac), which gave a number of possibilities for C and P.  From AA – DO (28dn), A (32dn) has a maximum value of 31 meaning that P, in PA (35ac), must have a minimum value of 4.
And then, well, nothing really.  I could get no further and having no firm grid entries after two days was unprecedented for me.  I started to entertain some pretty fanciful ideas: were there certain domino clues that had to be worked in base 7?  The non-domino numbers would fit nicely into a 5×5 square in the centre with the dominoes arranged around the edge.  We could see 6:6, 6:5 and 6:4 arranged vertically in the top two rows from the left – 666 fits for 1ac but could not continue to 3ac and 6:0 could not start 6dn.  All this was utterly futile.  I had been away on holiday the previous week and had barely looked at Poat’s Listener let alone finished it so could be looking at two blanks in a row.
Seasoned number puzzlers are accustomed to ignoring the longer clues until near the end but in desperation I did look at them and then kicked myself for having missed the following:
COAL + CARP – CARD – COP (33ac) + COP (29ac) = COAL + CARP – CARD (23dn)
The entry lengths meant the following:
29ac = ?0
33ac = 9??
23dn = 100?  At last I had my first grid entries.
Looking at COP (29ac) = ?0 gave C = 1 or 2; P = 5 or 10
After that progress was fairly rapid although I did use BBC Basic quite a bit to help with the donkey work, avoid omissions and minimise errors.  The following grid is as far as I could get without considering the dominoes:
4049 Interim Grid Fig 1
Confirmed values were: A=11, B=13, C=1, D=3, E=2, F=9, H=31, L=5, O=7, P=10, R=8, S=43, T=21 and V=41.
Placing the dominoes was fairly straightforward starting in the SE corner.  We couldn’t have two 1:0’s so double zero and double one were positioned and crossed off the list.  Eventually all were placed, the remaining clues solved and the final values assigned:
4049 Dominoes Placed Fig 2
G=17, I=35, K=95, N=34, Q=25, U=29, X=44 and Y=19.
So, no symmetrical pattern and no matching of numbers except by chance such as with the double five.  I have never seen them but Wikipedia tells us that there are commercially available sets up to double 9, 12, 15, 18 and possibly 21.  The higher values become difficult to distinguish with so many pips so perhaps they should use numbers like Googly.  However, might our instruction be to replace the numbers with pips?  The puzzle had no M so domino and number were out but we could see something like dot all tiles.  But enough of the speculation, the first two missing clues were soon unravelled:
8ac = 87978 = 2×3×11×31×43 = EDAHS reversed gives shade although I had heads at first.
31ac = 79170 = 2×3×5×7×13×29 = EDLOBU or double.
Finding the third missing clue and placing the final domino (5:2) was a delightful puzzle in its own right.  We were at the following stage:
4049 Final Three Cells Small
Ignoring the clues, 18ac could be either 426, 456, 476, 486 or 496.
426 = 1×2×3×71 = CEDI or dice (if I=71)  I checked in Chambers to see if dice might also cover dominoes to give shade double dice but it doesn’t.
456 = 3×8×19 = dry
476 = 2×7×34 = EON or one  Giving a clear favourite for the instruction: shade double one.
486 = 1×2×3×81 = CEDI or dice again (if I=81)
496 = 2×8×31 = ERH or her
However, once the clues were considered 18ac could only be 476 and IQ=35×25.  There was one other fit for IQ (=11×75) but 11 had already been assigned to A.
So, shade double one it was to give our final grid:
4049 Solution Fig 3
Someone once wrote that they didn’t like the number puzzles since they couldn’t be multilayered unlike the word ones.  Well Googly has certainly proved that wrong.  We had the concealed entrance, the number fill, the domino fill and a dénouement.  Granted, not as exciting as a felled cherry tree but really, what more could you ask for?  This was simply superb and you have to feel sorry for those unable to appreciate it as such.
Well, only one more number puzzle to come this year, due at the end of November, and we have not heard from Elap so far…

Posted in Solving Blogs | 5 Comments »

ROC setter’s blog

Posted by poat on 15 Sep 2009

Dave H having gently twisted my arm for a setter’s blog…I suppose this puzzle had a 20-year gestation period. Many moons ago when I first had the notion of submitting crosswords to The Listener (which was still a magazine at the time), one of the ideas I played around with was a 6×6 ‘magic word square’ that I had devised. Actually it wasn’t very satisfactory, because it included a glaring foreign word in row/column 5:







Despite the familiarity of MAISON, and its use from time to time in certain French-influenced fields such as fashion and cuisine, it has still failed to make it to Chambers in its own right. I nevertheless proposed to embed it in the centre of a grid, concealing its identity by means of a number of long jumbles (e.g. ATOMIC could be part of DOMINATRICES, the clue for which might be something like “Erotic maidens lashing with no end of discipline”). Potentially that could mean twelve intersecting answers of twelve letters each, with the remaining grid fill using answers entered normally, and the uncertainty in the centre would in theory be unravelled by solvers on the basis of the word square pattern. I did come up with a couple of grid fills on this basis, but never got round to finishing the puzzle, probably because the ‘jumble’ component was way too high.

I returned to the idea later, wondering whether to combine it with Thomas Mann’s ‘Magic Mountain’ in some way (for which the final word ASCEND or ASCENT would be important). However, internet research threw up a large number of much more interesting word squares, including a number of 8×8 squares that were compliant with Chambers. Here are some nice ones, which I can’t get to format properly:








…and so on, with some real beauties among them (there was even one with SCRABBLE as the first word, which caught my attention as a lover of the game). But the standout pattern for aesthetic interest of content was the BACKACHE square I eventually settled on. These had all been computer-generated by word buff Graham Toal and were presented online in their hundreds, although subsequent research established that my favourite had first been published in Word Ways magazine in the 1990’s. Unfortunately it looks as though the link to Graham’s webpage has now been taken down.

I also explored the squares of different lengths, picking out a few others of interest that I might be able to combine as an unseen square for reconstruction by the solver. These included a variety of words that might be fruitful for puzzling purposes, e.g. WHOOPS (a missing entry?), but in the end the possibilities of SYNTAX, having consulted its definition, seemed too good to bypass. This enabled solvers to construct both squares piecemeal, but in a thematically coherent way, the preamble being worded so as to mirror much of the relevant Chambers definition.

The Listener website revealed that smaller word squares had been used in a batch of puzzles by Tracer in the 1940’s and 1950’s, but nothing since then. It was therefore just a case of constructing a 12×12 grid around the 8×8, which of necessity included some jumbles as clued entries (although they would later resolve to form part of the central square). I noted that some could be backwards entries rather than mere jumbles which I know are frowned on by some, and made the split as orderly as possible with a 20:10:10 count of normal, jumbled and backwards entries – the last being indicated by the extra words, thus constraining me to two-word definitions in each case (and I tried to keep these as straightforward as possible, ‘quick crossword’ style). Then it was merely a case of writing the clues – for me, always an agonisingly long process.

Posted in Setting Blogs | 1 Comment »