Wednesday, March 11, 2015

Quadruple Pangram in THC

The Hindu Crossword 11335 by AfterDark featured a technical marvel – a quadruple pangram in a 15x15 British-style grid.

Quadruple Pangram, The Hindu Crossword  

Why is it a technical marvel?

Fitting the entire English alphabet four times in a 15x15 blocked grid requires the setter to predetermine 104 letters in an about 160-letter grid. Doing this with meaningful, clueable words, while also meeting the requirements of symmetry and checking, is an extraordinary achievement - not just rare, but possibly the first-ever in a British-style 15x15 grid. [If you have come across any other, please let me know.]

First-ever quadruple pangram in a 15x15?

Quadruple pangrams have appeared before in American-style grids. The first quadruple pangram in the New York Times (NYT), created by Peter Wentz, was published in Nov 2010. To accommodate the quadruple pangram, a non-standard 16x15 grid was used for this puzzle. In Dec 2012, American setter Matt Gaffney took up the challenge of fitting a quintuple pangram into a 16x15 grid. He didn't quite achieve it, ending with a grid six letters short of the goal.

In March 2013, Raymond C Young's crossword created a quadruple pangram in an American-style 15x15.

British-style grids have fewer lights in comparison with American-style grids, which makes the quest for a quadruple pangram even harder.

How did AfterDark do it?

In the setter's own words:

The first step was to choose the right grid. I had to look for a 15x15 that had a high letter count, yet avoided 3-letter solutions (in THC, 3-letter solutions are rare). Also, I had to find a grid that didn't have long words (they pose a big challenge in fitting in rare letters) and could accommodate more words. Fortunately, Crossword Compiler software helps in giving out such stats. I zeroed in on a grid with a mean word length of around 6, and an option of 38 words.

The second step was to deal with the difficult letters like Z,Q,J. In general, I chose slots where these letters could avoid a crossing. I then started from the NW quarter, trying to optimise the usage of difficult letters. And moved anti-clockwise, quarter by quarter, constantly checking for the number of times I had used the letters.

That was only the beginning. Each quarter ended up changing many times.

Strangely, after trying too hard to fit in the rare letters, in the end I was stuck with K, H & G. This happened because, every time a word choice was given, I opted to ignore the easy letters as I believed I could fit them in any time. When it came to the NE quarter at last, I was left with just 3 words but 7 letters to be included. This, again, led to a total alteration of two quarters.

There were words I deliberately avoided for one of these reasons: they were obscure (the barometer for that essentially was whether I had heard of the words somewhere or not), they didn't lend themselves to good definition, or they were "unclueable". Despite that, a few obscure words crept in, but I ensured that the wordplay and crossings would help the solver.

The last word that fell in place was – ironically - 1A, which throughout the process had remained intact (it was the first to be filled). To make place for a G in the end, DAZZLE became DANZIG (I had a luxury of an excess Z). Then what was taking EONS (3D) came out GUNS blazing.

AfterDark (Shrikanth Thirumalaiswami as he's known outside of crosswords) is no stranger to crossword grid artistry. Some of his past specials have been: every answer containing double letters, a pangram of end-letters of answers, 3 lipograms (A, E and T) with the letter excluded from the grid as well as the clues, a triple pangram.

Pangram of End Letters 

Do you like pangrammatic puzzles?

I landed on this very interesting article which reasons that pangrams don't matter in puzzle-making and, with their Scrabble-y words, tend to annoy the solver.

Compassionate solvers, not wanting to hurt a puzzle's feelings, might soft-pedal a bad solving experience by saying: "This puzzle is lame, but at least it's a pangram."  My friends, it's probably lame because it's a pangram.

Another puts forth a defence against that argument.

Like all technical accomplishments, it has nothing to do with whether a theme is clever or the fill is fresh. It’s completely artificial.

By the way, so is the syllable count in haiku or the meter in a sonnet.

Solvers and setters, which side of the debate are you on?

Related Posts:

If you wish to keep track of further articles on Crossword Unclued, you can subscribe to it in a reader via RSS Feed. You can also subscribe by email and have articles delivered to your inbox, or follow me on twitter to get notified of new links.

12 comments

Chaturvasi said...

While I do agree that a single brilliant clue in a standard normal grid may light up the whole crossword, I would say technical feats such as these should not be underestimated.

No setter gets any suggestion from the paper as to what he or she should do apart from keeping up the required supply as per the initial agreement.

Thus it is the setter who, prepared to do some extra work, sets a goal for himself or herself and tries to achieve that. Of course there have been many previous instances of such out-of-the-ordinary work and some of them have been noticed on this blog.

But AD has been consistently excelling himself in doing something different. He certainly deserves the congrats that his solvers shower on him. So do others who have been using novel ideas now and then.

This way THC has come a long way from its initial years. No doubt technology development and computer software have lent a notable hand in the implementation of setters' ideas.

The Colonel's blog has provided a platform for solvers to air their views.

Shrikanth T said...

Thank you very much CV Sir. This is what is 'Vashishtar Vaayal Brahmarishi' :)

anax said...

It is indeed a technical marvel, although AfterDark has used a perfect grid. It’s not the sort of challenge I’d like to take on but, if I did, I’d see each corner as a potential pangram with the middle region serving as a ‘spare’ for any missing letters.
Using a maximum word length of 7 letters is almost certainly an advantage since no answer commits too many letters to other lights. The setter has, if you like, 38 ‘goes’ at using the right letter combinations (a standard 15x15 has 28 answers, assuming they appear exclusively in odd rows/columns and that there are 4 long answers).
Funnily enough, I find the last letters pangram even more impressive – that really is a case of watching every step of the grid fill.
Of course, the most important thing is the clueing. The majority of solvers – if crossword blogs are anything to go by – easily miss in-grid trickery. Even if it is spotted, it will gain few positive comments if the clues are duff. Faced with that quadruple pangram grid, my immediate fear would be creating 38 clues of sufficient variety. For example, apart from ‘question’ the letter Q offers almost nothing, so in the process of wordplay breakdown I’d be conscious of avoiding not just the repetition but the situation in which Q is a left-over fragment to be dealt with.

ranger said...

Most of the setters usually give more importance to clue writing where as the gridfill doesn't get the same amount of attention. When I come across a pangram or a Nina, I appreciate the efforts put in by the setter,and I understand how much time would have gone into it.
But to come up with a quadruple pangram is no mean feat. Fantastic effort Shrikanth!

Shrikanth said...

Wow, Gridman and Arden praising. What more a setter needs in life :) Thanks Arden Sir :D

Ajeesh VM said...

As a solver I love awesome wordplay , surface reading et than pangram, clue acrostic etc which not always give you satisfaction as a solver. But very nice to see these innovations in grid fills. Great AD.

Ajeesh VM said...

+1 for anax comments about clue variety

Raghunath said...

I'm with Ajeesh's comment made at 1200. Kudos for the great effort AD.

Gautam said...

Is it possible to get a copy of the clues? Unfortunately this gem is not available on the THC archives.

Shuchi said...

Hi Gautam, for some unfathomable reason The Hindu randomly removes crosswords a few days after posting them on their site! I had a working link to The Hindu's page in the main post, but because of this change, had to delete it.

You can see the clues with answers on THCC blog.

Lakshmi Vaidyanathan (LV) said...

Hi Shuchi,
Here are the links to the clues & grid for this beautiful CW , posted by Col !
Link to the clues . Have to look for March -6 th post :

http://www.mayyam.com/talk/showthread.php?10502-THC-crossword-thread-6-5-ans-PP-till-4-pm-IST/page288

Link to the old grid which AD used for this CW

http://www.thehindu.com/todays-paper/tp-miscellaneous/the-hindu-crossword-10928/article5341764.ece

Lakshmi Vaidyanathan (LV) said...

It is the link from The Hub , where Col posts clues & the link to the grid daily :)