2d:null 2e:["$","title",null,{"children":"Collocations ƒ | John Cayley | Observable"}] 33:T46d,The content of Collocations, with associated exposition and interpretation, is discussed in the chapter corresponding to this live-code notebook. The code examined and/or reengineered here is chiefly concerned with the processes driving what I called, at the time, ‘collocational hologography.’ As outlined elsewhere, the creation and application of these processes was originally intended to render the texts that were generated by purely acrostic or mesostic transformations a little more ‘natural’ – when possible – by preferring acrostic/mesostic rule-satisfying two-word sequences that actually occurred in the supply texts, while, that is, these supply texts were being subjected to letteral transformations. The Collocations HyperCard stack offered, amongst other items, a collocationally constrained version of ‘Under It All’ and also two purely collocational ‘hologograms,’ ‘Critical Theory’ and ‘Indra’s PoemPoem.’ This notebook will expose and demonstrate their collocational process in their own terms, more or less, and based on their original HyperTalk scripting.34:T7a6,At the time I made *Collocations*, 1993, I had never heard of Markov models or Markov chains and, as I say, my motivations was to render other letteral deformations more linguistically ‘natural’ – with respect to their texts-to-be-read – than they would otherwise be. The implicit data structure that I came up with is, basically, the same as the abstracted and then generalized version which is set out, for example, in current research by my long-term collaborator and friend, Daniel C. Howe. Howe is, as of 2024, exploring the relationship between Markov models and large language models (LLMs).[[1]](#notes) A Markov model is, after all, a single-dimension vector space, graphing – for corpuses of natural language – the collocational frequencies of textually defined token sequences:– of words and/or punctuation in the initial examples that Howe is exploring; of white-space separated words in the case of my early 1990s HyperCard pieces. Howe outlines model-building algorithms that produce data structures whose entries have `keys` consisting of arbitrary numbers of one or more (*n* `- 1`) tokens, and `values` which are a `collection` of next (`+ 1 =` *n*) tokens. The total number of tokens in each collocated `gram` that can be generated from such a structure is the *n* of Markov models when they are referred to as ‘*n*-grams.’ Howe is currently investigating the theory and practicality of Markov models where *n* `= infinity` with the idea that this may shed some heuristic light on the large language models and their vector spaces which, by contrast, have incomprehensibly high training-generated dimensionalities. In the 1990s I appreciated that collocations taking account of even one additional preceding token – a 3-gram model – would better anticipate the next – third – tokens, but I did not have the coding or (I thought) the computing power in HyperCard to build such a ‘collocational’ model. ---35:T91d,>*there is a temptation to make the critique itself an object like the work considered an object of attention presented without further comment neither declaiming nor explaining it presents still the crowd cries out for reassurance values a ground to stand on wants to know what to think
these works have an essential seductive quality no matter how troubling their form and content no matter how painful the impressions created by materials and imagery these stark configurations and massive wall bound constructions of sombre colour found objects rusting scrap obscure photographs torn printed extracts all juxtaposed with dim minimal abstraction they welcome the eye a tableau allowing it rest enjoyment pleasure they still the heart lead inwards
this is work founded on an idealism which demands that in the midst of the green world there must be some reminder of the grey the contrary consolation is of little interest
everyday inanimate objects are simplified abstracted made more substantial and weighty turned to stone and metal conjoined or constructed in closed and enclosing configurations of intensely wall related installations chairs rooms tables tableaux
something has been added to transform many of these simple objects all too easy to behold into instruments of torture the table burns its electricity silently without a flame
the materials and their configurations shabby equipment always deteriorating pretend an inadequacy an inability to present the articulate tableau but insist on a venture a raid a fight to recover what is lost neither gain nor loss restlessness
representations of living things parts of the body and its clothing gloves coats torsos arms are presented materially transformed by the media used lead wax rubber plaster media which both solidify and give dead nature
the forms of living things in the work have shapes which are wasted worn distorted they are tortuously arranged or suspended within the formal configurations of simplified everyday objects or even the abstract configurations of no things
there is a crowd present in the work photographic images from newspapers magazines film television all but hidden by its closures staring shouting surging out from within but its concerns are great set in judgement beyond and outside the wall bound frame*36:T524,In the HyperCard background of the card that was holding our supply text, a HyperTalk script – `defOccs` for ‘define occurrences’ – was written to preprocess the text and create other cards in the stack, one for each corresponding whitespace-separated word in the text. ``` on defOccs lock screen put item 1 of fld label into textName set numberFormat to "0000" put the number of words in fld text into totWords repeat with i = 1 to totWords put word i of fld text into newWord push cd go cd textName & ",occs," & newWord if the result is not empty then go cd "blank,def" doMenu "New Card" set the cantDelete of this cd to false put textName & ",occs," & newWord into fld label end if put i & " " after fld defNos pop cd put i & "/" & totWords end repeat unlock screen end defOccs ``` The script created the card, if it did not already exist, and added an occurrence number for this word to a field on the card. If a card for the word already existed, the script simply added its additional occurrence number to the card’s field. This, effectively, created what JavaScript would call a `Map` of serially numbered occurrences indexed by each unique word in the supply. We do the same thing by writing a function in JavaScript like this:37:T836,collocationalPlayer = async (supplyText, supplyTextMap, displayId) => { let words = supplyText.split(/\s/); // put the supply texts words into an array let displayIndex = randomInteger(words.length); // pick the index of a random word to start let currentWord = words[displayIndex]; // and make it the current word let display = document.getElementById(displayId); // this is our display area display.innerHTML = ""; for (let i = 0; i < words.length; i++) { // this loops for the number of words in the supply text // and then allows the function to return (as the original did) let screenWords = []; for (let j = 0; j < 16; j++) { // an arbitrary number, based on the original and Macintosh SE display size let displayWord = currentWord; if ( // insert blank line if currentWord is a paragraph-ending word [ "think", "inwards", "interest", "tableaux", "flame", "restlessness", "nature", "things", "frame" ].includes(currentWord) ) { displayWord += "<br><br>"; // blank line via 2 br tags } // add current word to the display screenWords.push(displayWord); let possibleIndexes = supplyTextMap.get(currentWord); // get current word’s occurrences displayIndex = (possibleIndexes[randomInteger(possibleIndexes.length)] + 1) % words.length; // pick an occurrence currentWord = words[displayIndex]; i = ++i % words.length; } fadeInByWord(screenWords, display, 800); // fade in 16 words, one by one // time is needed to complete this animation await Promises.delay(16 * 800 + 800); // gracefully fade out the 16 words display.style.color = "ivory"; // time is needed for the color transition await Promises.delay(3000); // clear and reset the text color of the display area display.innerHTML = ""; display.style.color = "black"; } return "The collocational player has finished playing. Refresh page to restart."; }38:T4a2,>*indra’s poempoem three dimensions in literature
the visual hologram represents three dimensions in two and the literary hologogram performs a similar feat letters are the graphic atoms of literary representation imagine the words of a text set as usual in two dimensions on a plane a piece of transparent paper tilt the plane slightly and suspend it in space now imagine invisible lines forming an irregular grid in the third dimension and emanating from each of the letters of the text both above and below the words lying on the plane along each of these lines in what will be an acrostic dimension arrange all the other words of the text which share the letter from which a particular line emanates what you see is an irregular solid a constellation of words in three dimensional space associated and structured by two very different criteria through their membership of the set of words which composes a particular text and because they share a particular literary graphic element
a free abecedarian or internal acrostic indra’s net is simply a set of readings along the lines of the acrostic dimension of this virtual three dimensional constellation of related words*39:Ted4,### Stack Script ``` -- the 'playing cards' (accessed by 'score' button) are: -- acrostic1, gloss1, gloss2, gloss3, gloss4 -- playText is the main script on playText global ¬ textName,¬ wherePlay,¬ showBase,¬ numOWords,¬ newWord, ¬ i, ¬ trans, ¬ lastTrans, ¬ collocRef, ¬ lastCollocRef, ¬ playing, ¬ interrupt,veFactor,wCount put 0 into wCount put true into playing set cursor to none hide menuBar hide msg set the numberFormat to "0000" put the number of words in fld text of cd textName & ",text" ¬ into numOWords put empty into cd fld baseFld of cd wherePlay hide cd fld baseFld of cd wherePlay put empty into cd fld playFld of cd wherePlay if wherePlay is "collocExp" then open file "collocExp" if wherePlay is "acrostExp" then open file "acrostExp" write "" to file "acrostExp" end if go cd wherePlay if (wherePlay is "colloc1") or (wherePlay is "collocExp") then put random(numOWords) into i put word i of fld "text" of cd textName & ",text" into lastTrans end if put false into interrupt -- if wherePlay is collocExp then put 9000 into wLimit else put 8250 into wLimit repeat until (i is numOWords) -- or (wCount > wLimit) if wherePlay is not "colloc1" ¬ and wherePlay is not "collocExp" then ¬ add 1 to i if the mouse is down and the shiftKey is down then exit repeat put word i of fld text of cd textName & ",text" ¬ into newWord if showBase then put newWord into cd fld baseFld end if send transWord to cd if interrupt then exit repeat end if if showBase then lock screen hide cd fld baseFld do "unlock screen with dissolve" && veFactor end if if i is numOWords then put 0 into i end repeat show cd fld baseFld of cd wherePlay if wherePlay is "acrostExp" then close file "acrostExp" if wherePlay is "collocExp" then close file "collocExp" end playText on seekWord xWord global ¬ textname,¬ wherePlay,¬ numOWords,¬ newWord, ¬ i, ¬ showBase, ¬ trans, ¬ lastTrans, ¬ collocRef, ¬ lastCollocRef, ¬ playing, ¬ interrupt,defName put xWord into trans if there is a cd defName & ",Occs," & trans then put (pickOcc(trans)) into i put word i of fld text of cd textName & ",text" into newWord put newWord into cd fld baseFld subtract 1 from i else put false into wordFound repeat until wordFound show cd fld baseFld put word i of fld text of cd textName & ",text" into newWord put newWord into cd fld baseFld if newWord is trans then put true into wordFound subtract 1 from i else add 1 to i if i > numOWords then put 1 into i end if end repeat end if put false into interrupt playText if i >= numOWords then visual effect dissolve slow go cd "Contents" end if end seekWord function pickOcc xWord global ¬ defName return word random(the number of words in fld defNos of cd defName & ¬ ",Occs," & xWord) of fld defNos of cd defName & ¬ ",Occs," & xWord end pickOcc -- -- ****Contents field scripts -- on startPlay global explName,textName,wherePlay,showBase,¬ numOWords,newWord,i,interrupt,veFactor do "visual effect dissolve" && veFactor put 0 into i -- -- the main routine in the stack script is called from here: -- playText -- if not interrupt then do "visual effect dissolve" && veFactor go cd "Contents" end if end startPlay on loadVars global sFactor,veFactor,chFactor,wdFactor,phFactor,stFactor,cdFactor,ve put cd fld "sFactor" into sFactor put cd fld "veSpeed" into veFactor if veFactor is "normal" then put empty into veFactor put cd fld "ve" into ve end loadVars ```3a:T621,### Script of card *colloc1* ``` -- this card plays collocation transformations on transWord global ¬ textname,¬ wherePlay,¬ numOWords,¬ newWord, ¬ i, ¬ showBase, ¬ trans, ¬ lastTrans, ¬ collocRef, ¬ lastCollocRef, ¬ interrupt, ¬ pickedColloc,veFactor,withColloc,defName,sFactor repeat 16 -- if i is 1 then put "winds" into lastTrans put (pickOcc(lastTrans)+1) into collocRef if collocRef > numOWords then put 1 into collocRef put collocRef into i put word collocRef of fld "text" of cd textName & ",text" ¬ into trans -- now show the transformation if the mouse is down then put lastTrans into trans put trans into cd fld baseFld put true into interrupt exit repeat end if showTrans wait sFactor put trans into lastTrans end repeat if not interrupt then if i = numOWords then wait 5 seconds else wait 90 lock screen put empty into cd fld baseFld put empty into cd fld playFld do "unlock screen with dissolve" && veFactor end if end transWord on showTrans global showBase,trans,pickedColloc,veFactor -- showing is also taylored for each playing card lock screen -- if pickedColloc then put "*" after cd fld playFld put trans & " " after cd fld playFld if "," & trans & "," is in ",think,inwards,interest,tableaux,flame,restlessness,nature,things,frame," then put return & return after cd fld playFld end if if showBase then show cd fld baseFld do "unlock screen with dissolve" && veFactor end showTrans ```3b:[] 31:{"id":"aa5c6a6af393ecda","slug":"inapm_collocations_f","title":"Collocations ƒ","thumbnail":"c536c0bca29e09664f8ca918e175286a78abf4e1c329b5b19a8d9cc1de13b557","access_level":"public","create_time":"$D2024-05-21T16:03:20.033Z","update_time":"$D2026-09-12T00:04:29.333Z","view_time":null,"first_public_time":"$D2024-05-21T16:03:20.033Z","has_unpublished_edits":null,"fork":true,"stars":0,"starred":false,"owner":{"type":"team","id":"8cad4906c41da0ef","login":"shadoof","name":"John Cayley","avatar_url":"https://avatars-next.observableusercontent.com/d4a045c1060f722297a6047ffa90d6cd33806f54fcaa444ac3f1882c0fd4ee4f","tier":"pro_2024","bio":"Observable aspects of my digital language art. I like to ask people to read nllf.net, some of which is driven by notebooks made here."},"version":710,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"5c3cbd6193f712f7","login":"shadoof","name":"John Cayley","avatar_url":"https://avatars-next.observableusercontent.com/d4a045c1060f722297a6047ffa90d6cd33806f54fcaa444ac3f1882c0fd4ee4f","tier":null,"bio":"Observable aspects of my digital language art. I like to ask people to read nllf.net, some of which is driven by notebooks made here."}],"members":[{"type":"individual","id":"5c3cbd6193f712f7","login":"shadoof","name":"John Cayley","avatar_url":"https://avatars-next.observableusercontent.com/d4a045c1060f722297a6047ffa90d6cd33806f54fcaa444ac3f1882c0fd4ee4f","tier":null,"role":"editor"},{"type":"team","id":"8cad4906c41da0ef","login":"shadoof","name":"John Cayley","avatar_url":"https://avatars-next.observableusercontent.com/d4a045c1060f722297a6047ffa90d6cd33806f54fcaa444ac3f1882c0fd4ee4f","tier":"pro_2024","role":"viewer"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":{"id":"9d74d3a23ee71cf6","slug":"body","title":"Body","owner":{"id":"8cad4906c41da0ef","login":"shadoof","name":"John Cayley"}},"type":"notebook","body":{"titling":"auto","title":"Collocations ƒ","cells":[{"id":4,"value":"Working Draft\n# Collocations ƒ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":15,"value":"$33","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"p1","database":"$undefined"},{"id":68,"value":"$34","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"p2","database":"$undefined"},{"id":66,"value":"Selected HyperTalk scripts from the stack are quoted below. In the following cells, we refer to these scripts but extract textual data and translate the code into corresponding JavaScript with commentary.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":76,"value":"We work, initially, with the supply text of ‘Critical Theory.’ First, for easy reading, we put it into a Markdown element that renders it in italics.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":353,"value":"$35","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"supplyMarkdown","database":"$undefined"},{"id":373,"value":"Then we put this element’s `innerText` into a JavaScript `string literal`. In HyperCard, this text was formatted for human reading over a number of cards corresponding to its paragraphs. It was also added, in its entirety, to the field of a card with a background named `texts.` It retained line breaks that are also respected here, in the JavaScript `literal`:","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":79,"value":"criticalTheorySupply = supplyMarkdown.innerText","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":85,"value":"$36","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":92,"value":"occurrencesMap = (supplyText) => {\n let occurrences = new Map(); // create a Map\n let words = supplyText.split(/\\s/); // split string literal supply text into words\n for (let i = 0; i < words.length; i++) {\n // iterate through the words\n let word = words[i]; // get the current word\n if (occurrences.has(word)) {\n // if we already have an entry for the word\n occurrences.get(word).push(i); // add its occurrence number to the array for that word\n } else {\n // we don't yet have an entry for the word, so\n occurrences.set(word, [i]); // create an entry with an array containing the first occurrence number\n }\n }\n return occurrences;\n}","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":104,"value":"And then, when we evoke it, it generates a `Map` for the ‘Critical Theory’ supply text:","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":106,"value":"criticalTheoryMap = ƒcriticalTheoryMap()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":538,"value":"ƒcriticalTheoryMap = () => occurrencesMap(criticalTheorySupply);","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":109,"value":"This `Map` is, effectively, our 2-gram (bigram) Markov model. In the HyperCard version, a map or model is built into the structure of the stack itself: as a digital representation of a collection of index cards. (See the chapter on [HyperCard](inapm2_hypercard).) Similarly, but in a generalized manner, this notebook’s Reactive JavaScript version now maintains an analogous structure, which you, reader, can ‘see’ and explore above.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":145,"value":"Now, all we need to do is script or code a ‘player’ that visualizes, dynamically in this case, an elaboration of the collocational hologogram that is implicit in the map structure. In the HyperCard version, after clicking the relevant item of the stack’s table of contents card, you were sent to a player card that was appropriate for your choice. Then a text was generated and displayed by a HyperTalk script that performed in the same manner as the following live-code notebook version.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":150,"value":"We define a textual display area in html – one having, in order to be faithful to our original limitations, the aspect of a Macintosh SE’s screen.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":152,"value":"html`

`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":154,"value":"Now, we write a function (below) that will take the supply text’s `criticalTheoryMap` and display a generated text, word-by-word, based on the supply text, and doing so – between page refreshes – for the same number of words as the supply. The JavaScript code here follows the original HyperTalk scripting in terms of configurations that could be easily changed or improved. For example the display is cleared by the original HyperCard stack every sixteen words and it prints a blank line when triggered by any word that ends one of the supply text’s paragraphs. The visual effects are coded entirely differently here but approximate the visual behavior of the HyperCard stack.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":157,"value":"$37","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":286,"value":"And, finally, our player is evoked – whenever the page is refreshed – here:","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":165,"value":"playOnRefresh = ƒplayOnRefresh()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":549,"value":"ƒplayOnRefresh = () =>\n collocationalPlayer(criticalTheorySupply, criticalTheoryMap, \"ctheory\")","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":603,"value":"The `playText` HyperTalk script from the original stack is given below for comparison, although this is somewhat kludgy and cluttered by cosmetic and interface considerations. ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":605,"value":"For the sake of further exemplification and so as to reengineer another piece that was included in the original HyperCard stack, here is the same live-code treatment for the final piece in the stack’s table of contents, ‘Indra’s PoemPoem.’\n\nHere is the supply text.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":631,"value":"$38","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"ippSupply","database":"$undefined"},{"id":638,"value":"Put this into a string variable:","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":640,"value":"ippSupplyText = ippSupply.innerText","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":643,"value":"Make a Map as our 2-gram model, reusing the `occurrencesMap` function:","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":651,"value":"ippModel = ƒippModel()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":648,"value":"ƒippModel = () => occurrencesMap(ippSupplyText)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":653,"value":"Specify another display element for animated generation from the model:","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":655,"value":"html`

This cell is styling CSS for the display elements.

\n","pinned":true,"hidden":false,"mode":"html","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"_localcss","database":"$undefined"},{"id":24,"value":"---\n## Selected Raw HyperTalk Scripts from the *Collocations* Stack\n*Please note: scripting that relates entirely or chiefly to cosmetics or navigation has been edited out.*\n\nThese stacks had, typically, five backgrounds:\n- *PrelimPlus* : Pages or screens readable in a conventional manner by book-like navigation.\n- *cardSet* : A number of cards shared this background with particular named cards used for one another readable animation.\n- *Illustration* : For cards holding actual graphic material for occasional display.\n- *texts* : With versions of the supply texts formatted for preprocessing.\n- *defs* : Basically used as a relatively simple data structure. Cards with this background were created and named dynamically during preprocessing, storing information about words in the texts.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"scripts","database":"$undefined"},{"id":19,"value":"$39","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"stack","database":"$undefined"},{"id":270,"value":"$3a","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"Bkgrnd","database":"$undefined"},{"id":392,"value":"import { css, fadeInByWord, randomInteger } from \"@shadoof/inapm0_shared\"","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":7,"value":"_css = css","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":416,"value":"In italics if shared CSS is working.","pinned":true,"hidden":false,"mode":"html","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"_cssTest","database":"$undefined"}],"files":"$W3b","stdlib":"2"}} 32:"$31:body"