2b:null 2c:["$","title",null,{"children":"The Two-Way Resume | ento | Observable"}] 31:T4ab,md`# The Two-Way Resume tl;dr: I'm going to try using a [resume format](https://two-way-resume.netlify.app/) that lists **_what I'm looking for_** and my **_weaknesses_** alongside the commonly seen sections. I hope shifting how information gets shared **_to the left_** (i.e. as early as possible in the recruiting process) improves efficiency and effectiveness, leading to a more fluid and adaptive work-personnel allocation across society. ## Situation I have been in the business of looking for a job for the past few months, and truth be told, I've been reluctant to go full-throttle. Why? Well, among other things, the process felt stilted; while the resume continues to be the primary package by which you present yourself when tapping your network or applying for a position, the commonly used formats leave so many things undercommunicated. This didn't seem right, and I wanted to at least attempt to do something about it before giving myself up to the status quo. ### What a resume doesn't tell you Let's take a step back. What's the goal of searching for a job...? — I think it comes down to finding a mutually beneficial relationship that satisfies both parties' needs. `32:T46b,md`

(Those in parentheses are sections that may often be omitted.)

More notes

You might say: but some of these cover both _needs_ and _can-do_'s. You're right. Salary range of a job post, for example, says "We can pay you this amount" and _also_ says "We need someone who's okay with this amount." For this breakdown, I tried to simplify by asking the following questions and designating each section as one of _need_ or _can_:

How customary is it to write [the section] in the language of {need, can}?
What is the common perception of [the section] - demand or supply?

**What the job seeker is looking for** is undercommunicated. The more conscientious employers will ask "What are you looking for in your next role?" while interviewing — but why can't it be communicated upfront? I suspect that a lot of **interview questions can be answered in the resume itself**, including dealbreakers from the point of view of the employer. That is, what kind of **needs the job seeker cannot meet**. `33:T458,{ const rowHeight = 120; const columnWidth = 480; const rowsPerColumn = 20; const numCandidates = 160; const svg = d3 .create('svg') .attr('width', width) .attr('height', 600) .attr('viewBox', `0 0 2000 ${rowsPerColumn * rowHeight}`) .attr('preserveAspectRatio', 'xMinYMin'); const rc = rough.svg(svg.node()); const seekerSymbols = randomSymbols(3, 6); const rowScale = d3 .scaleLinear() .domain([0, 1]) .range([0, rowHeight]); const columnScale = d3 .scaleLinear() .domain([0, 1]) .range([0, columnWidth]); const roundGenerator = d3.randomExponential(1.6); d3.range(0, numCandidates).forEach(candidate => { const numRounds = Math.min(5, Math.ceil(roundGenerator())); const column = Math.floor(candidate / rowsPerColumn); const row = candidate % rowsPerColumn; const g = randomMatch( rc, `Company ${candidate + 1}`, 80, seekerSymbols, numRounds ); g.attr('transform', `translate(${columnScale(column)},${rowScale(row)})`); svg.node().appendChild(g.node()); }); return svg.node(); }34:T4a7,md`## Hypothesis: the two-way resume Based on these hypotheses: > * The more information we share early on... > * The quicker we'll know whether a job is a good match and less time will be spent on each opportunity. > * The more effectively our networks can function as a bridge toward the right opportunity, improving the good-match-to-poor-match ratio of opportunities considered. I've been working on expanding my resume and personal website with new data, which are opportunities for the job seeker to share information as early as possible in the process: ||Where|Time Spent On Opportunity|Number Of Opportunities Considered| |-|-|-|-| | What I need or what I'm looking for | Resume | ▽ | △ Good match | | What I cannot do, aka weaknesses | Resume | ▽ | ▽ Poor match | | Answers to common (behavioral) interview questions | Personal website | ▽ | △ Good match
▽ Poor match | The rest of this notebook will focus on the format of the resume that I arrived at, which I've dubbed the two-way resume. My personal website is still a work in progress and I may do a roundup of how I organized questions in another notebook. My new resume looks roughly like this: `35:Tb09,{ const svg = d3 .create('svg') .attr('height', 330) .attr('width', width) .attr('viewBox', '0 0 300 220'); const canWidth = 85; const needWidth = 25; const section = ({ title, x, y, height }) => { const titleHeight = 20; return { append: 'g', transform: `translate(${x}, ${y})`, children: [ { append: 'text', text: title, x: 0, y: titleHeight - 5, 'font-size': 12, 'font-family': 'Caveat,sans-serif' }, { append: 'rect', x: 0, y: titleHeight, width: canWidth, height: height - titleHeight, fill: breakdownChartConfig.cellColorByGroup("Can"), 'data-fill-style': breakdownChartConfig.cellStyleByGroup("Can") }, { append: 'rect', x: canWidth + 5, y: titleHeight, width: needWidth, height: height - titleHeight, fill: breakdownChartConfig.cellColorByGroup("Need"), 'data-fill-style': breakdownChartConfig.cellStyleByGroup("Need") } ] }; }; const marginLeft = 20; const mainContent = { append: 'g', transform: `translate(${marginLeft},0)`, 'font-family': 'Caveat,sans-serif', children: [ { append: 'text', text: 'RESUME', x: 0, y: 20, 'font-size': 14 }, { append: 'text', text: 'Can', x: 30, y: 40, 'font-size': 12 }, { append: 'path', d: `M0,47 v-3 h${canWidth} v3`, strokeWidth: 1, stroke: '#333' }, { append: 'text', text: 'Need', x: canWidth + 3, y: 40, 'font-size': 12 }, { append: 'path', d: `M${canWidth + 5},47 v-3 h${needWidth} v3`, strokeWidth: 1, stroke: '#333' } ] }; const columnLabelsHeight = 48; // depends on the column labels paths above const mainContentHeight = [ ['Strengths & weaknesses', 40], ['Skills', 30], ['Experience', 35], ['Education', 25] ].reduce((y, [title, height], i) => { mainContent.children.push( section({ title: title, x: 0, y: y, height: height }) ); return y + height; }, columnLabelsHeight); const page = [ { append: 'g', transform: 'translate(0,5)', children: [ { append: 'rect', fill: 'transparent', strokeWidth: 1, stroke: '#999', x: 0, y: 0, width: 110 + marginLeft * 2, height: mainContentHeight + marginLeft }, mainContent ] } ]; render(svg, page); return coarse(svg.node()); }36:T565,md` #### Stating what I cannot do, aka weaknesses The resume starts off with a high-level overview section that summarizes my strengths and weaknesses, framed in these two questions: * What I'm professionally good at * What I'm professionally _not_ good at #### Stating what I need or what I'm looking for Each section has an aside that details what I'm looking for next in that area. * Skills: Which skills I'm looking to make use of * Experience: What kind of opportunities I'm looking for next, such as mission, industry, or culture * Education: What I'm learning next Here's a list of needs that can be considered for inclusion: * Location, including openness to remote work and willingness to relocate * Visa sponsorship * Company mission and culture * Company size * Job type (full-time, contract, part-time, ...) * Job functions * Individual contributor and/or manager * Skills to make use of or skills to learn * How best to assess me, or interview accommodations * Compensation (I left this out myself because it felt too much of a taboo to publicly disclose this information. Maybe it can be described in vague terms like "enough to make a living") The hope is that the resume screening process will reject my application early on if there's a fundamental mismatch or, conversely, advance my application because of a particular detail that indicated a good match. `37:T554,md` #### The purpose of a resume is to get your foot in the door, nothing more That may well be the case now. It doesn't necessarily have to stay that way! #### Isn't this something covered by the Objective section? and it's fallen out of fashion I wonder about that too. I wonder why it has fallen out of fashion if that's true. In any case, the difference I see is that the two-way resume is more finegrained and structured along existing sections, making it easier for the reader to parse and understand. #### You're in a position that allows you to be vocal about your needs, which isn't the case for me Yes, I'm not a fresh graduate and I have developed some ideas about how I want my work life to be. I have not run out of money yet. I do believe — idealistically — that whoever you are and whatever stage in life you are at, no qualification is required to have your needs met, and that we all can get to a world where the needs of the job seekers are given equal consideration as that of the employers'. The first step towards that world may be to state those needs. #### Publicizing your weaknesses will lower your chance of getting a job My hope is to raise the chance of working at a place that is a good fit and that such places will be accepting or open-minded enough to be curious to learn more when they see my weaknesses upfront. `38:T592,md` ## Postscript There are more ways that the recruiting process resembles dating, or an example of sub-par communication if placed in a real-world context; * The needs of one side of the relationship being de-emphasized (at the resume screening stage at least) is unhealthy in the first place, not just in terms of lack of information * Assumptions without clarification; sometimes the interviewee gets only one chance to get an answer right. People's communication styles and thinking styles differ wildly — it's best to ask rather than to assume * Ghosting ### What can employers do? One thing I can think of is to revise job posts to include company mission, culture, and interview process, or any other information that interviewees may ask in interviews. [Lots of job boards are structured to only require little more than job requirements](https://observablehq.com/@ento/job-boards-comparison), but much more can be communicated if job posts are written with the mindset of shifting informaton sharing to the left. Also consider doing more of what's covered in interviews in writing upfront, especially if the team is remote-first or remote-friendly. It'll save everyone's time, take the time pressure off of interviewees, reduce variability coming from having different interviewers, and may be more reflective of how the actual work will be done. (I admit I'm biased here as I like written communcation.) `39:T1040,function cansNeedsBreakdown(resumeFieldGroups, jobPostFieldGroups) { function calculatePositions(fieldGroups) { return fieldGroups.groups.reduce((y, group) => { group.y = y; group.height = breakdownChartConfig.cellWidth * group.members.length; group.members = group.members.map((name, i) => ({ name, y: group.y + breakdownChartConfig.cellWidth * i })); return y + group.height + breakdownChartConfig.cellWidth * 0.5; }, 0); } function renderGroups(fieldGroups) { const g = d3.create('svg:g'); const labelsOnLeft = fieldGroups.align === 'right'; const { groupLabelsX, labelsX, cellsX } = (isLeft => { if (isLeft) { return { groupLabelsX: 0, labelsX: breakdownChartConfig.groupLabelsWidth, cellsX: breakdownChartConfig.groupLabelsWidth + fieldGroups.fieldLabelsWidth }; } else { const labelsX = breakdownChartConfig.cellWidth + 4; return { cellsX: 0, labelsX, groupLabelsX: labelsX + fieldGroups.fieldLabelsWidth }; } })(labelsOnLeft); g.append('text') .attr('x', labelsOnLeft ? cellsX + breakdownChartConfig.cellWidth : 0) .attr('y', 20) .attr('text-anchor', labelsOnLeft ? 'end' : 'start') .attr('font-weight', 'bold') .text(fieldGroups.title); const cells = g .append('g') .attr( 'transform', `translate(${cellsX},${breakdownChartConfig.cellsTop})` ); const labels = g .append('g') .attr( 'transform', `translate(${labelsX},${breakdownChartConfig.cellsTop})` ); const groupLabels = g .append('g') .attr( 'transform', `translate(${groupLabelsX},${breakdownChartConfig.cellsTop})` ); fieldGroups.groups.forEach((group, groupIndex) => { groupLabels .append('text') .text(group.name) .attr('font-size', breakdownChartConfig.fieldLabelsFontSize) .attr('x', labelsOnLeft ? breakdownChartConfig.groupLabelsWidth - 4 : 6) .attr('y', group.y + group.height * 0.5) .attr('dy', breakdownChartConfig.fieldLabelsFontSize * 0.4) .attr('text-anchor', labelsOnLeft ? 'end' : 'start'); groupLabels .append('path') .attr( 'd', labelsOnLeft ? `M ${breakdownChartConfig.groupLabelsWidth + 3} ${ group.y } h -3 v ${group.height} h 3` : `M 0 ${group.y} h 3 v ${group.height} h -3` ) .attr('fill', 'none') .attr('stroke', '#333'); group.members.forEach(member => { const cell = rc.rectangle( 0, member.y, breakdownChartConfig.cellWidth, breakdownChartConfig.cellWidth, { stroke: '#444', fill: breakdownChartConfig.cellColorByGroup(group.name), fillStyle: breakdownChartConfig.cellStyleByGroup(group.name) } ); cells.node().appendChild(cell); labels .append('text') .attr('x', labelsOnLeft ? fieldGroups.fieldLabelsWidth - 4 : 0) .attr('y', member.y + breakdownChartConfig.cellWidth - 4) .attr('font-size', breakdownChartConfig.fieldLabelsFontSize) .attr('text-anchor', labelsOnLeft ? 'end' : 'start') .text(member.name); }); }); return g; } const resumeBottom = calculatePositions(resumeFieldGroups); const jobPostBottom = calculatePositions(jobPostFieldGroups); const svg = d3 .create('svg') .attr('width', width) .attr( 'height', breakdownChartConfig.cellsTop + Math.max(resumeBottom, jobPostBottom) ); const rc = rough.svg(svg.node()); // render resume intention cells svg.node().appendChild(renderGroups(resumeFieldGroups).node()); // render job post intention cells svg.node().appendChild( renderGroups(jobPostFieldGroups) .attr( 'transform', `translate(${breakdownChartConfig.fieldGroupsWidth},0)` ) .node() ); return svg.node(); }3a:T1614,randomMatch = { function symbol(rc, parent, type, y, size) { const star = rc.path( d3.symbol(d3['symbol' + capitalize(type.description)], size)(), { stroke: '#666' } ); d3.select(star).attr('transform', `translate(0,${y})`); parent.appendChild(star); } function renderSymbolsColumn(rc, shapeTypes, hiddenIndices, x, size, scaleY) { const g = d3.create('svg:g').attr('transform', `translate(${x},0)`); shapeTypes.forEach((type, i) => { if (hiddenIndices.has(i)) { d3.select(g.node()) .append('text') .text('?') .attr('stroke', '#c30') .attr('font-size', '14') .attr('dy', scaleY(i) + 5) .attr('dx', -3); } else { symbol(rc, g.node(), type, scaleY(i), size); } }); return g; } function renderSymbols( rc, sizes, seekerSymbols, seekerHiddens, employerSymbols, employerHiddens ) { const g = d3.create('svg:g').attr('transform', 'translate(0,24)'); const scaleY = d3 .scaleLinear() .domain([0, seekerSymbols.length - 1]) .range([ sizes.cellSize, sizes.cellSize + seekerSymbols.length * sizes.cellSize ]); g.node().appendChild( renderSymbolsColumn( rc, seekerSymbols, seekerHiddens, sizes.leftX, sizes.symbolSize, scaleY ).node() ); g.node().appendChild( renderSymbolsColumn( rc, employerSymbols, employerHiddens, sizes.rightX, sizes.symbolSize, scaleY ).node() ); // Matching lines function visibleIndices(symbols, hiddens) { return symbols.reduce((m, sym, i) => { if (!hiddens.has(i)) { m.set(sym, i); } return m; }, new Map()); } const visibleSeekerIndices = visibleIndices(seekerSymbols, seekerHiddens); const visibleEmployerIndices = visibleIndices( employerSymbols, employerHiddens ); const lineIndices = Array.from(visibleEmployerIndices.entries()).reduce( (lines, [key, value]) => { if (visibleSeekerIndices.has(key)) { lines.push({ leftIndex: visibleSeekerIndices.get(key), rightIndex: value }); } return lines; }, [] ); lineIndices.forEach(line => { g.node().appendChild( rc.line( sizes.leftX + sizes.cellSize, scaleY(line.leftIndex), sizes.rightX - sizes.cellSize * 0.6, scaleY(line.rightIndex) ) ); }); return g; } // "S [count] E" function renderHeading(rc, parent, count) { parent .append('svg:g') .attr('transform', 'translate(0,16)') .call(selection => { selection .append('text') .attr('x', 8) .attr('font-size', 14) .attr('font-family', 'Caveat,sans-serif') .text('S'); selection .append('text') .attr('x', 28) .attr('font-size', 20) .attr('stroke', '#000') .text(count); selection .append('text') .attr('x', 54) .attr('font-size', 14) .attr('font-family', 'Caveat,sans-serif') .text('E'); selection.node().appendChild(rc.line(0, 6, 64, 6)); }); } // single round: heading + columns of symbols function round( rc, count, seekerSymbols, seekerHiddens, employerSymbols, employerHiddens ) { const g = d3 .create('svg:g') .attr('transform', `translate(${(count - 1) * 80},0)`); const sizes = { leftX: 8, rightX: 54, symbolSize: 32 }; sizes.cellSize = sizes.symbolSize / 3; // size to use in d3 scale g.node().appendChild( renderSymbols( rc, sizes, seekerSymbols, seekerHiddens, employerSymbols, employerHiddens ).node() ); // Heading text renderHeading(rc, g, count); return g; } function generateRandomIndices(array) { const indices = d3.shuffle(d3.range(0, array.length)); return new Set(indices.slice(0, d3.randomInt(1, indices.length)())); } function deleteAtRandom(set) { const count = d3.randomInt(1, set.size)(); d3.range(count).forEach(i => { const keys = Array.from(set.keys()); if (keys.length > 0) { set.delete(d3.shuffle(keys)[0]); } }); } // series of rounds function randomMatch(rc, name, nameWidth, seekerSymbols, numRounds) { const employerSymbols = d3 .shuffle(d3ShapeTypes.slice()) .slice(0, d3.randomInt(3, d3ShapeTypes.length)()); const seekerHiddens = generateRandomIndices(seekerSymbols); const employerHiddens = generateRandomIndices(employerSymbols); const g = d3.create('svg:g'); const roundsGroup = d3.create('svg:g').attr('transform', `translate(${nameWidth},0)`); g.node().appendChild(roundsGroup.node()); d3.range(1, numRounds + 1).forEach(count => { roundsGroup .node() .appendChild( round( rc, count, seekerSymbols, seekerHiddens, employerSymbols, employerHiddens ).node() ); deleteAtRandom(seekerHiddens); deleteAtRandom(employerHiddens); }); g.append('g') .attr('transform', 'translate(0,60)') .append('text') .attr('font-family', 'Caveat,sans-serif') .attr('font-size', '16') .text(name); return g; } return randomMatch; }3b:T5fa,matrixAnalysis = { const signalScale = { domain: ["none", "low", "mid", "high"], range: ["#fff", "#c5e8c3", "#82cec2", "#23479c"] }; const resumeChart = vl .markRect() .encode( vl .y() .fieldN("Job Post") .axis({ title: "Job Post (employer needs)" }) .scale({ domain: jobPostDomain }), vl .x() .fieldN("Resume") .axis({ title: "Resume (job seeker can do)" }) .scale({ domain: resumeDomain }), vl .color() .fieldO("ResumeSignal") .scale(signalScale) .legend({ orient: "bottom" }) ) .title([ "Do resumes give good signal that", "a job seeker can satisfy employer needs?" ]) .config({ axis: { grid: true, tickBand: "extent" } }); const jobPostChart = vl .markRect() .encode( vl .y() .fieldN("Job Post") .axis({ title: "Job Post (employer can do)" }) .scale({ domain: jobPostDomain }), vl .x() .fieldN("Resume") .axis({ title: "Resume (job seeker wants)" }) .scale({ domain: resumeDomain }), vl .color() .fieldO("JobPostSignal") .scale(signalScale) ) .title([ "Do job posts give good signal that", "an employer can satisfy job seeker needs?" ]) .config({ axis: { grid: true, tickBand: "extent" } }); return vl .hconcat(resumeChart, jobPostChart) .data(signals) .render({ renderer: 'svg' }); }3c:T605,signals = [ { Resume: "Location", "Job Post": "Location", ResumeSignal: "high", JobPostSignal: "high" }, { Resume: "Objective", "Job Post": "Company mission", ResumeSignal: "low", JobPostSignal: "mid" }, { Resume: "Objective", "Job Post": "Company culture", ResumeSignal: "low", JobPostSignal: "mid" }, { Resume: "Objective", "Job Post": "Role & responsibilities", ResumeSignal: "low", JobPostSignal: "mid" }, { Resume: "Objective", "Job Post": "Skill requirements", ResumeSignal: "low", JobPostSignal: "mid" }, { Resume: "Skills", "Job Post": "Role & responsibilities", ResumeSignal: "mid", JobPostSignal: "mid" }, { Resume: "Skills", "Job Post": "Skill requirements", ResumeSignal: "mid", JobPostSignal: "mid" }, { Resume: "Skills", "Job Post": "Company culture", ResumeSignal: "none", JobPostSignal: "none" }, { Resume: "Experience", "Job Post": "Role & responsibilities", ResumeSignal: "mid", JobPostSignal: "mid" }, { Resume: "Experience", "Job Post": "Skill requirements", ResumeSignal: "mid", JobPostSignal: "mid" }, { Resume: "Experience", "Job Post": "Company culture", ResumeSignal: "low", JobPostSignal: "low" }, { Resume: "Education", "Job Post": "Role & responsibilities", ResumeSignal: "low" }, { Resume: "Education", "Job Post": "Skill requirements", ResumeSignal: "low" } ]3d:Tb2d,intentions = [ { Document: "Job Post", Field: "Name", Intention: "Can", Signal: "low" }, { Document: "Job Post", Field: "Name", Intention: "Want", Signal: "low" }, { Document: "Job Post", Field: "Location", Intention: "Can", Signal: "mid" }, { Document: "Job Post", Field: "Location", Intention: "Want", Signal: "high" }, { Document: "Job Post", Field: "Role & responsibilities", Intention: "Can", Signal: "none" }, { Document: "Job Post", Field: "Role & responsibilities", Intention: "Want", Signal: "high" }, { Document: "Job Post", Field: "Skill requirements", Intention: "Can", Signal: "none" }, { Document: "Job Post", Field: "Skill requirements", Intention: "Want", Signal: "high" }, { Document: "Job Post", Field: "Company mission", Intention: "Can", Signal: "high" }, { Document: "Job Post", Field: "Company mission", Intention: "Want", Signal: "mid" }, { Document: "Job Post", Field: "Company culture", Intention: "Can", Signal: "high" }, { Document: "Job Post", Field: "Company culture", Intention: "Want", Signal: "mid" }, { Document: "Job Post", Field: "Interview process", Intention: "Can", Signal: "none" }, { Document: "Job Post", Field: "Interview process", Intention: "Want", Signal: "mid" }, { Document: "Job Post", Field: "Compensation", Intention: "Can", Signal: "high" }, { Document: "Job Post", Field: "Compensation", Intention: "Want", Signal: "none" }, { Document: "Resume", Field: "Name", Intention: "Can", Signal: "none" }, { Document: "Resume", Field: "Name", Intention: "Want", Signal: "none" }, { Document: "Resume", Field: "Location", Intention: "Can", Signal: "none" }, { Document: "Resume", Field: "Location", Intention: "Want", Signal: "high" }, { Document: "Resume", Field: "Objective", Intention: "Can", Signal: "low" }, { Document: "Resume", Field: "Objective", Intention: "Want", Signal: "high" }, { Document: "Resume", Field: "Skills", Intention: "Can", Signal: "high" }, { Document: "Resume", Field: "Skills", Intention: "Want", Signal: "none" }, { Document: "Resume", Field: "Experience", Intention: "Can", Signal: "high" }, { Document: "Resume", Field: "Experience", Intention: "Want", Signal: "none" }, { Document: "Resume", Field: "Education", Intention: "Can", Signal: "mid" }, { Document: "Resume", Field: "Education", Intention: "Want", Signal: "none" } ]3e:T45c,{ const signalScale = { domain: ["none", "low", "mid", "high"], range: ["#fff", "#b4c8d6", "#5e97bf", "#2f6790"] }; const chartForDocument = (docType, fieldDomain, title) => { return vl .markRect() .data(intentions) .transform({ filter: `datum.Document == '${docType}'` }) .encode( vl .y() .fieldN("Field") .scale({ domain: fieldDomain }), vl .x() .fieldN("Intention") .scale({ domain: ["Can", "Want"] }) .axis({ orient: 'top' }), vl .color() .fieldO("Signal") .scale(signalScale) .legend({ orient: "bottom" }) ) .title(title) .config({ axis: { grid: true, tickBand: "extent" } }); }; return vl .hconcat( chartForDocument("Resume", resumeDomain, [ "How well do resumes capture intention?", "" ]), chartForDocument("Job Post", jobPostDomain, [ "How well do job posts capture intention?", "" ]) ) .render({ renderer: 'svg' }); }3f:T14e1,{ const svg = d3 .create('svg') .attr('width', width) .attr('height', 240); const rc = rough.svg(svg.node()); function renderIntentions(title, domain, fieldSignals, labelsOnLeft = true) { const g = d3.create('svg:g'); const labelsX = labelsOnLeft ? 0 : latticeChartConfig.cellWidth * 2 + 4; const cellsX = labelsOnLeft ? latticeChartConfig.yLabelsWidth : 0; g.append('text') .attr('x', labelsOnLeft ? cellsX + latticeChartConfig.cellWidth * 2 : 0) .attr('y', 20) .attr('text-anchor', labelsOnLeft ? 'end' : 'start') .attr('font-weight', 'bold') .text(title); const cells = g.append('g').attr('transform', `translate(${cellsX},0)`); ["Can", "Want"].forEach((intention, x) => { cells .append('text') .text(intention) .attr('font-size', latticeChartConfig.labelsFontSize) .attr( 'transform', `translate(${latticeChartConfig.xScale(x) + latticeChartConfig.cellWidth - 5},${latticeChartConfig.yScale(0)}) rotate(-65)` ); domain.forEach((field, y) => { const cell = rc.rectangle( latticeChartConfig.xScale(x), latticeChartConfig.yScale(y), latticeChartConfig.cellWidth, latticeChartConfig.cellWidth, { stroke: '#444', fill: latticeChartConfig.cellColorScale( fieldSignals[field][intention] ), fillStyle: latticeChartConfig.cellStyleScale( fieldSignals[field][intention] ) } ); cells.node().appendChild(cell); }); }); const labels = g.append('g').attr('transform', `translate(${labelsX},0)`); domain.forEach((field, y) => { labels .append('text') .attr('x', labelsOnLeft ? cellsX - 4 : 0) .attr('y', latticeChartConfig.yScale(y + 1) - 2) .attr('font-size', latticeChartConfig.labelsFontSize) .attr('text-anchor', labelsOnLeft ? 'end' : 'start') .text(field); }); return g; } // render resume intention cells const resumeFields = signalByIntentionByField('Resume'); svg .node() .appendChild(renderIntentions('Resume', resumeDomain, resumeFields).node()); // render job post intention cells const jobPostFields = signalByIntentionByField('Job Post'); svg.node().appendChild( renderIntentions('Job Post', jobPostDomain, jobPostFields, false) .attr('transform', `translate(${latticeChartConfig.jobPostLeft},0)`) .node() ); /* const resumeFieldIndex = arrayToIndexMap(resumeDomain); const jobPostFieldIndex = arrayToIndexMap(jobPostDomain); // @type { resumeFromJobPost: {[resumeField]: {[jobPostField]: [signal]}}, jobPostFromResume: ... } const wantSources = signals.reduce( (out, signal) => { const resumeField = signal['Resume']; const jobPostField = signal['Job Post']; if ( signal.JobPostSignal && signal.JobPostSignal != 'none' && resumeFields[resumeField]['Want'] != 'none' ) { out.resumeFromJobPost[resumeField] = out.resumeFromJobPost[resumeField] || {}; out.resumeFromJobPost[resumeField][jobPostField] = signal.JobPostSignal; } if ( signal.ResumeSignal && signal.ResumeSignal != 'none' && jobPostFields[jobPostField]['Want'] != 'none' ) { out.jobPostFromResume[jobPostField] = out.jobPostFromResume[jobPostField] || {}; out.jobPostFromResume[jobPostField][resumeField] = signal.ResumeSignal; } return out; }, { resumeFromJobPost: {}, jobPostFromResume: {} } ); function renderWantSatisfiers( links, sourceFieldIndex, targetFieldIndex, sourceX, targetX, offsetY, strokeColor ) { Object.entries(links).forEach(([targetField, incoming]) => { const targetY = intentionChartConfig.yScale(targetFieldIndex[targetField]) + intentionChartConfig.cellWidth * 0.5 + offsetY; svg.node().appendChild(rc.circle(targetX, targetY, 4)); Object.entries(incoming).forEach(([sourceField, strength]) => { const sourceY = intentionChartConfig.yScale(sourceFieldIndex[sourceField]) + intentionChartConfig.cellWidth * 0.5 + offsetY; const path = d3 .linkHorizontal() .source(() => [sourceX, sourceY]) .target(() => [targetX, targetY])(); svg.node().appendChild( rc.path(path, { fill: 'none', stroke: strokeColor, strokeWidth: intentionChartConfig.strokeWidthScale(strength) }) ); }); }); } renderWantSatisfiers( wantSources.resumeFromJobPost, jobPostFieldIndex, resumeFieldIndex, intentionChartConfig.jobPostLeft + intentionChartConfig.cellWidth * 0.5, intentionChartConfig.resumeRight - intentionChartConfig.cellWidth * 0.5, 2, '#600' ); renderWantSatisfiers( wantSources.jobPostFromResume, resumeFieldIndex, jobPostFieldIndex, intentionChartConfig.resumeRight - intentionChartConfig.cellWidth * 1.5, intentionChartConfig.jobPostLeft + intentionChartConfig.cellWidth * 1.5, -2, '#006' ); */ return svg.node(); }40:[{"id":"bf024ef868524e6160d313f263d2c8561dbd48782532ea6372776e19aded8c554a43fda88ef6d326715eb25143ccd3fdee0564b497c23d7bec7b9214b57e070b","href":"https://static.observableusercontent.com/files/bf024ef868524e6160d313f263d2c8561dbd48782532ea6372776e19aded8c554a43fda88ef6d326715eb25143ccd3fdee0564b497c23d7bec7b9214b57e070b","name":"layout-sketch.png","type":"image/png","lastModified":1606778559181.379,"size":140864}] 2f:{"id":"870329b8d2a4bc25","slug":"the-two-way-resume","title":"The Two-Way Resume","thumbnail":"7a3de9abaa559703a19161381b1ed2ff413c93afc9d0a99bed006185d95e08d0","access_level":"public","create_time":"$D2020-10-12T05:55:55.723Z","update_time":"$D2023-07-14T07:35:39.609Z","view_time":null,"first_public_time":"$D2020-11-01T18:55:24.385Z","fork":false,"stars":24,"starred":false,"owner":{"type":"team","id":"29c3cae0b7a46ab0","login":"ento","name":"ento","avatar_url":"https://avatars-next.observableusercontent.com/243e0c0d5a7636b3c2ec7be518b54666640203cca9008428e1f02a36bcd0488b","tier":"starter_2024"},"version":1464,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"9e583a9c2f29bc70","login":"ento","name":"ento","avatar_url":"https://avatars-next.observableusercontent.com/243e0c0d5a7636b3c2ec7be518b54666640203cca9008428e1f02a36bcd0488b","tier":null}],"members":[{"type":"individual","id":"9e583a9c2f29bc70","login":"ento","name":"ento","avatar_url":"https://avatars-next.observableusercontent.com/243e0c0d5a7636b3c2ec7be518b54666640203cca9008428e1f02a36bcd0488b","tier":null,"role":"editor"}],"comments":[],"custom_thumbnail":true,"trashed":false,"subscribed":null,"forks":4,"collection":{"id":"e9532638ed3f76a7","slug":"maybe-the-real-treasure-was-the-tools-i-made-along-the-way-of-j","title":"Maybe The Real Treasure Was The Tools I Made Along The Way Of Job Searching","owner":{"id":"29c3cae0b7a46ab0","login":"ento","name":"ento"}},"type":"notebook","body":{"title":"The Two-Way Resume","cells":[{"id":0,"value":"$31","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":59,"value":"coarse(dot`digraph G {\n node [shape=Square];\n rankdir = LR;\n subgraph cluster_job_seeker {\n label = \"job seeker\";\n seeker_can [label=\"can do\"];\n seeker_needs [label=\"needs\"];\n }\n subgraph cluster_employer {\n label = \"employer\";\n employer_needs [label=\"needs\"];\n employer_can [label=\"can do\"];\n }\n seeker_can -> employer_needs;\n employer_can -> seeker_needs;\n}`)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":865,"value":"md`\nLike they say, it's just like dating. Like dating, communication is the key to learning about each other.\n\nBut, lots of resumes and job posts are heavily skewed toward what the employer needs and how the job seeker can meet their needs:\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":889,"value":"cansNeedsBreakdown(\n {\n title: 'Resume',\n groups: [\n {\n name: 'Can',\n members: ['Skills', 'Experience', 'Education'],\n color: 'high'\n },\n {\n name: 'Need',\n members: ['Location'],\n color: 'mid'\n },\n {\n name: 'No Signal',\n members: ['Name'],\n color: 'none'\n }\n ],\n align: 'right',\n fieldLabelsWidth: 70\n },\n {\n title: 'Job Post',\n groups: [\n {\n name: 'Need',\n members: ['Location', 'Role & responsibilities', 'Skill requirements']\n },\n {\n name: 'Can',\n members: ['Compensation', '(Mission)', '(Culture)']\n },\n {\n name: 'Low Signal',\n members: ['Name']\n }\n ],\n align: 'left',\n fieldLabelsWidth: 130\n }\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":723,"value":"$32","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":891,"value":"md`### Lack of information, longer processes, more work for everyone\nIt's only natural that the dating analogy extends beyond the fact that it's a two-way street; it's a long and gradual process of unraveling what either party needs or whether one party can meet the other's needs.`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":516,"value":"single_rounds = {\n const svg = d3\n .create('svg')\n .attr('width', width)\n .attr('height', 120);\n const rc = rough.svg(svg.node());\n\n const seekerSymbols = randomSymbols(6, 6);\n const row = randomMatch(rc, `Assessment rounds`, 110, seekerSymbols, 4);\n svg.node().appendChild(row.node());\n return svg.node();\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":577,"value":"md`_Whether the job seeker (S) meets what the employer (E) needs and whether E meets what S needs are sussed out over multiple rounds of assessments. Here, I count resume screening as the first round._\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":567,"value":"md`According to a blog post for soon-to-be-graduates, it takes [_100-200+ applications to receive one job offer_](https://www.hirelehigh.com/post/how-many-applications-does-it-take-to-get-a-job#:~:text=Recent%20stats%20(from%20Talent%20Works,to%20get%20one%20job%20offer.). Let's picture it:`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":275,"value":"$33","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1394,"value":"md`\nIf job seekers didn't have to send out hundreds of applications, employers could spend less time reviewing them. If there could be fewer interview rounds, it'd save time for everyone. The sooner everyone's matched with a place they feel safe, productive and appreciated, the happier we'll all be.\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":544,"value":"md`\n## Problem statement\n**What can a job seeker do to make this arduous process more efficient and spend less time searching for a job that's a good fit?**\n\nLet's break it down a little bit — the total time spent searching for a job, in very simplified terms, can be given like:`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":600,"value":"md`\n\\`\\`\\`\nTimeSpentJobSeaching = TimeSpentOnOpportunity▽ × NumberOfOppotunitesConsidered▽△\n\\`\\`\\`\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":606,"value":"md`\nSo\n* We want to minimize \\`TimeSpentOnOpportunity\\` (\\`▽\\`)\n* We want to minimize \\`NumberOfOpportunitiesConsidered\\` _that are not a good match_ (\\`▽\\`)\n* We want to maintain or increase \\`NumberOfOpportunitiesConsidered\\` _that are a good match_ (△) — by expanding the search radius, we'll be closer to picking the 'best' possible opportunity out there\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":66,"value":"$34","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":878,"value":"$35","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1172,"value":"$36","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1279,"value":"md`### Points counter points`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1209,"value":"$37","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1239,"value":"md`\n## And The Result...? Yet To Come\nI haven't put this two-way resume format to test in the wild yet. Wish me luck!\n
\n
\nUpdate: I found employment within a month after writing this notebook. Hard to say whether the resume helped or hindered overall, but I'm still with the company after 2+ years.\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":555,"value":"$38","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1321,"value":"md`
\n# Appendix\n\n## Prior art\nRecruiting services that let companies search job seeker profiles, such as [AngelList Talent](https://angel.co), [Hired.com](https://hired.com), or [StackOverflow Jobs](https://stackoverflow.com/jobs), allow you to register your job preferences similar to the list above. Two-way resume is like unlocking that ability outside of the confines of these closed platforms.\n\n[Applied](https://www.beapplied.com/) is transforming the recruiting process in the same direction: ask specific questions upfront in writing before getting into the interview stage. [Toggl Hire](https://toggl.com/hire/) seems to be focused on verifying hard skills through automatic assessments. [Product Hunt has a list](https://www.producthunt.com/alternatives/toggl-hire) of alternatives.\n\n## Resources\n[KeyValues](http://keyvalues.com/) has a nice list of cultural values to peruse and pick from.\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1339,"value":"md`
\n## Implementation`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":56,"value":"dot = require(\"@observablehq/graphviz@0.2\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":274,"value":"rough = require('roughjs@3.0.0/dist/rough.js').catch(() => window.rough)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":411,"value":"caveatFont = html``","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":625,"value":"import { coarse } from '@ento/coarse'","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":247,"value":"d3 = require(\"d3@6\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":788,"value":"$39","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":964,"value":"breakdownChartConfig = {\n const config = {\n cellWidth: 16,\n fieldGroupsWidth: 200,\n cellsTop: 40,\n groupLabelsWidth: 60\n };\n config.fieldLabelsFontSize = config.cellWidth - 4;\n config.cellColorByGroup = d3\n .scaleOrdinal()\n .domain([\"No Signal\", \"Low Signal\", \"Can\", \"Need\"])\n .range([\"#fff\", \"#fff\", \"#b4c8d6\", \"#2f6790\"]);\n config.cellStyleByGroup = d3\n .scaleOrdinal()\n .domain([\"No Signal\", \"Low Signal\", \"Can\", \"Need\"])\n .range([null, null, \"hachure\", \"cross-hatch\"]);\n return config;\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":240,"value":"d3Render = require('d3-render@v0.2.4')","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":241,"value":"render = d3Render.default","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":550,"value":"md`#### Assessment-rounds chart implementation`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":511,"value":"$3a","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":419,"value":"function capitalize(string) {\n return string.charAt(0).toUpperCase() + string.slice(1);\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":506,"value":"randomSymbols = (minCount, maxCount) => d3\n .shuffle(d3ShapeTypes.slice())\n .slice(0, d3.randomInt(minCount, maxCount + 1)());","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":501,"value":"d3ShapeTypes = [\n Symbol('circle'),\n Symbol('cross'),\n Symbol('diamond'),\n Symbol('square'),\n Symbol('star'),\n Symbol('triangle'),\n Symbol('wye')\n]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1002,"value":"outtakes = md`## Outtakes\nHerein you can see how this notebook became a hodgepodge of visualization tools.\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":246,"value":"{\n const svg = d3\n .create('svg')\n .attr('width', width)\n .attr('height', 100);\n let i = resumeDomain.length;\n while (true) {\n render(\n svg,\n resumeDomain.slice(0, i % (resumeDomain.length + 1)).map((d, i) => ({\n append: \"g\",\n transform: `translate(0, ${10 * i})`,\n children: [\n {\n append: \"circle\",\n stroke: \"black\",\n cx: 10,\n cy: 5,\n r: 5,\n fill: \"transparent\"\n },\n {\n append: \"text\",\n text: d,\n x: 20,\n y: 10\n }\n ]\n }))\n );\n\n i++;\n yield svg.node();\n await Promises.tick(2000);\n break;\n }\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":7,"value":"$3b","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":82,"value":"resumeDomain = [\n \"Name\",\n \"Location\",\n \"Objective\",\n \"Skills\",\n \"Experience\",\n \"Education\"\n]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":79,"value":"jobPostDomain = [\n \"Name\",\n \"Location\",\n \"Role & responsibilities\",\n \"Skill requirements\",\n \"Company mission\",\n \"Company culture\",\n \"Compensation\",\n \"Interview process\"\n]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":10,"value":"$3c","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":174,"value":"$3d","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":2,"value":"import { vl } from '@vega/vega-lite-api'","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":224,"value":"vl\n .markBar()\n .data(scores)\n .encode(\n vl.column(\"Intention\").title(\"\"),\n vl\n .x()\n .fieldN(\"Document\")\n .axis({ title: \"\" }),\n vl.y().fieldQ(\"Score\"),\n vl\n .color()\n .fieldN(\"Document\")\n .legend({ title: \"\" })\n )\n .title([\"How expressive are job match docs?\", \"\"])\n .height(100)\n .render()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":221,"value":"scores = {\n const rv = [];\n [\"Resume\", \"Job Post\"].forEach(docType => {\n [\"Can\", \"Want\"].forEach(intention => {\n rv.push({\n Document: docType,\n Intention: intention,\n Score: score(docType, intention)\n });\n });\n });\n return rv;\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":214,"value":"score = (docType, intention) =>\n intentions\n .filter(d => d.Document === docType && d.Intention === intention)\n .map(d => ({ high: 1, mid: 0.5, low: 0.1, none: 0 }[d.Signal]))\n .reduce((a, b) => a + b, 0)\n .toFixed(1)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":176,"value":"$3e","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1018,"value":"md`\n

I wish the job matching process would be based more on data, not guesswork.

\n\n_Me, in a feedback to a recruiting service startup._\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":646,"value":"$3f","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":759,"value":"latticeChartConfig = {\n const config = {\n cellWidth: 16,\n xLabelsHeight: 65,\n yLabelsWidth: 80,\n jobPostLeft: 200\n };\n config.labelsFontSize = config.cellWidth - 4;\n // config.resumeRight = config.yLabelsWidth + config.cellWidth * 2;\n\n config.xScale = d3\n .scaleLinear()\n .domain([0, 1])\n .range([0, config.cellWidth]);\n config.yScale = d3\n .scaleLinear()\n .domain([0, 1])\n .range([config.xLabelsHeight, config.xLabelsHeight + config.cellWidth]);\n\n config.cellColorScale = d3\n .scaleOrdinal()\n .domain([\"none\", \"low\", \"mid\", \"high\"])\n .range([\"#fff\", \"#b4c8d6\", \"#5e97bf\", \"#2f6790\"]);\n config.cellStyleScale = d3\n .scaleOrdinal()\n .domain([\"none\", \"low\", \"mid\", \"high\"])\n .range([null, \"hachure\", \"zigzag\", \"cross-hatch\"]);\n /*\n config.strokeWidthScale = d3\n .scaleOrdinal()\n .domain(['low', 'mid', 'high'])\n .range([1, 2, 2.5]);\n */\n return config;\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1087,"value":"// Convert `intentions` to a data structure that's easier for this chart.\n// This may become unnecessary once we trim down on which charts are in this notebook.\n// @return {[field]: {[intention]: [signal]}}\nfunction signalByIntentionByField(docType) {\n return intentions\n .filter(d => d.Document === docType)\n .reduce((fields, d) => {\n fields[d.Field] = fields[d.Field] || {};\n fields[d.Field][d.Intention] = d.Signal;\n return fields;\n }, {});\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":1449,"value":"{\n const sketch = await FileAttachment(\"layout-sketch.png\").image();\n sketch.style.width = `${width * 0.5}px`;\n sketch.style.height = 'auto';\n return html`

\n ${sketch}\n

Sketch of possible layouts

\n`;\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"}],"files":"$W40","stdlib":"1","titling":"auto"}} 30:"$2f:body"