Puzzles/GooglePoetry
Problem
Provide a query which causes Google to return a J Dictionary page in the top 3 search results
Exposition
But not just any query. If you type: [ "finite state machine" "emit vector" ] (Search) you might reasonably expect to get the Vocabulary entry for ;:. In fact, you do, and that's all you get.
But you probably wouldn't expect: [ The strange, powerless police mapped Greece ] (Search) to do the same. And yet it does.
So the true challenge is to craft an amusing query which a person who is familiar with J would not obviously expect to return a Dictionary page.
Now, that last query was a bit of a cheat, because it relys solely upon Roger's (Ken's?) penchant for strange words and obscure quotes. Which brings us to the:
Constraints
1.#0 Query must be to the Google search engine at [1]. Use of other search engines is barred, to keep the playing field level. But if you find a good query on another engine, certainly submit it for an honorable mention. 1. A J Dictionary page must appear in the top 3 search results. 1. Proof is required. With your result, provide the __exact url query string__ that resulted in a J Dictionary page in the first 3 results. If you need help, see the Notes section. 1. Fewer words is better (because it's harder). 1. More total results is better (because you're fishing in a bigger pond). Total results are reported in the top right corner of the results page. as "Results A-B of C for QUERY". Maximize C. 1. Use of double quotes is discouraged (too specific). Unless the query is really good. 1. Use of other special syntax (like | or &) is probably also discouraged. 1. Pages falling under Ndx are discouraged (too many possibilities). Unless the query is really good. 1. Don't rely solely upon a strange word or turn of phrase not directly applicable to J. But using them as spice is permitted; perhaps encouraged. 1. The higher the Dictionary page in the result list, the better. Position 0 is better than position 1 is better than position 2. 1. Queries that result in "new hits" (pages in the Dictionary not raised by previous queries) are better. 1. A single query that results in more than one Dictionary page in the top 3 results wins you the "Walking god" award. 1. If you happen to be the author of the Dictionary of J, changing it to win the contest is discouraged.
These constraints are subject to change, addition, and tightening, because I can't possibly think of all the fun-ruining cheats in advance.
Remember, this puzzle is much "softer" than the others -- the goal is to have a collection of amusing queries, not for someone to take the prize by adhering strictly to the constraints. There is no right answer; only good answers.
Notes
The best and easiest way to provide your proof link (the link to Google showing a Dictionary page in the top 3 results) is to follow these steps:
1. Type your query into Google 1. Copy the resulting URL from your address bar 1. Optional: pare it down as much as possible, so it's easier to read and less apt to break. As a bonus, this tends to increase the result count. To reduce it, just remove as many parameters as possible (those &name=value bits after the search? part) without changing the results. Usually, all you need is search?q=.... For example, searching for
[foo bar baz] (Search) will result in the minimal URL http://www.google.com/search?q=foo+bar+baz
1. Now create a Wiki link. This is usually of the form Text typed in to Google. Note the space between the URL and the link text (the words you typed into Google).
If this is too hard or confusing, just use the other solutions as models or templates.
Spoiler Alert!
Solutions
| No. | Query & |
Dic. page(s) |
Words | Total results | Notes | Author/Sig |
| 0 | Fret! Worlds ends! Maximal atom yield size! Shards, shards! | Cut (;.) | 14 | 530 | Yes, you could elide or rearrange some words, but that would make it less cute. I can't quantify "cute" but it's more important than "short". | -- Dan Bron <<DateTime(2006-05-18T22:15:35Z)>> |
| 1 | eleemosynary index | Index of last (i:) | 2 | 49100 | top hit as of today | -- Roger Hui <<DateTime(2006-05-18T23:55:02Z)>> |
| 2 | Power to the gerund! | Power (^:) | 4 | 126000 | -- Roger Hui <<DateTime(2006-05-19T00:03:47Z)>> | |
| 3 | Mnemonics copula | Intro 1. (Mnemonics) | 2 | 754 | Ok, so it evades the spirit... | -- Dan Bron <<DateTime(2006-05-19T01:24:37Z)>> |
| 4 | An inflected name denotes membership | Section I. (Alphabet and Words) | 5 | 9090 | All 3 top results are J-related. But 1 and 3 are identical and 2 is not in the Dictionary, so I'm no walking god. | -- Dan Bron <<DateTime(2006-05-19T02:11:29Z)>> |
| 5 | It is fit to be tolerant of close relations. Global tolerance is arbitrary. | Section II.D (Comparatives) | 13 | 412000 | Section D has a lot of promise. This could probably be improved manyfold. | -- Dan Bron <<DateTime(2006-05-19T02:34:53Z)>> |
| It is fit to be tolerant of close relations. Global tolerance is complex. | 1470000 | See what I mean? | ||||
| 6 | Floating, floor to ceiling. Everyday, at the point of equality, a part falls. | Section II.D (Comparatives) | 16 | 84200 | -- Dan Bron <<DateTime(2006-05-19T03:01:43Z)>> | |
| 7 | The sentence is final. Last words? Assert. Break. Continue. | Control structures | 10 | 1930000 | Many more possibilities here, too. | -- Dan Bron <<DateTime(2006-05-19T03:30:54Z)>> |
| The sentence is final. Last words? Simple: Assert. Break. Continue. | 11 | 2480000 | ||||
| The sentence is execution. Last words? Simple: Assert. Break. Continue. | 735000 | |||||
| 8 | triplets individually boxed | 8!: (Format) | 3 | 86400 | -- Roger Hui <<DateTime(2006-05-19T03:36:10Z)>> | |
| 9 | assigned root companions | :. (Obverse) | 3 | 1910000 | -- Roger Hui <<DateTime(2006-05-19T03:58:44Z)>> | |
| 10 | formal imperative analytic language | Dictionary | 4 | 1450000 | -- Roger Hui <<DateTime(2006-05-19T04:35:15Z)>> | |
| 11 | passive crosses connections | Reflexive (~) | 3 | 7710000 | These kind are easy! Find the __fun__ ones. | -- Dan Bron <<DateTime(2006-05-19T04:36:46Z)>> |
| 12 | state security immediate execution | 9!: (Global Parameters) | 4 | 12900000 | -- Ewart Shaw <<DateTime(2006-05-19T09:38:46Z)>> | |
| 13 | recognizing strange sequential words | Words - Sequential Machine | 4 | 123000 | -- Raul Miller <<DateTime(2006-05-19T19:46:05Z)>> | |
| 14 | eligible proverbs and bold legend of trident | E. Parsing and Execution | 5 | 645 | Since google ignores "of" and "and", I'm ignoring them in my word count (but, of course, they're on the page) | -- Raul Miller <<DateTime(2006-05-19T19:55:20Z)>> |
| 15 | executing explicit rome standard paths | I. Locatives | 5 | 244000 | -- Raul Miller <<DateTime(2006-05-19T20:08:38Z)>> | |
| 16 | trains interrupted, a domain divided: an error in signals, the result of an argument | [| Cap ] | 16 | 205000 | hard to be brief; one wants to use all the candidate words on a page (how can I work "unbroken" or "a wider range" into the poem?) | -- Dan Bron <<DateTime(2007-03-21T20:46:09Z)>> |
| 17 | Examining foreign sentences and phrases, ignoring unfamiliar Latin inflections. | III. Definitions | 10 | 12100 | -- Dan Bron <<DateTime(2007-03-21T21:25:44Z)>> | |
| Examining foreign sentences and phrases, ignoring unfamiliar inflections. | 9 | 17400 | easy to change this to increase page count or decrease word count, but then all the fun words aren't included | |||
| Examining Latin sentences and phrases, ignoring unfamiliar auxiliaries. | 9 | 15200 | ||||
| 18 | raw roll twister without repetition | Roll/Deal | 5 | 21600 | -- Roger Hui <<DateTime(2007-11-07T18:57:13Z)>> | |
| 19 | pi significant subdomain | circle function | 3 | 25400 | Playing around with the first word yields other interesting listings (some with more total results) | -- Dan Bron <<DateTime(2008-03-24T17:14:54Z)>> |
| 20 | space ace determined to decipher latin alphabet | alphabet/ace | 7 | 19100 | -- Dan Bron <<DateTime(2008-10-07T20:55:02Z)>> | |
| 21 | alphabet ace | alphabet/ace | 2 | 1630000 | Wins by the letter, but loses by the spirit. | -- Dan Bron <<DateTime(2008-10-07T21:01:02Z)>> |
Discussion
- See the corresponding thread in the Forums
- I've made a few of these over the years, and some were really good! Now I wish I had saved them. -- Dan Bron <<DateTime(2006-05-18T22:28:25Z)>>
- Hey Ewart, I don't see J coming up in any of the results for this search. What Dictionary page does it bring up for you?
- Nevermind, it's working now. Weird.
- And so far, you have my vote for the win. Not just from the words%results ratio, but because the query is funny and I certainly wouldn't expect it to result in a J page.
- I simplified & reduced all the Google URLs in the table (my URLS were put together by my browser, Opera, resulting in many unneccesary search parameters). I verified that the new URLs pulled up the same results as the old, with the same ballpark total results count.
- I added a column indicating which J page(s) result from the query. I don't know if that ruins the fun or not, but it'll be helpful when the queires stop working (and start resulting in, for example, this page).