Three-Ring Circus

December 4, 2024

It's been a full year since I gave notice at Seattle Convention Center. Being an Admission Attendant was my first front-of-house job, my first time in a union, and my first job wearing a uniform. The job had its charms, but I began to feel like I wasn't living up to my potential.

Since then, I've learned Power Platform. Earned two certifications. Won an award at the annual conference. Built a QR Cards solution.

And, best of all, I started this Model Citizen Developer newsletter, which just hit the 200-subscriber mark!

The first 100 were mostly people I know personally, which is lovely. I'm glad to be in regular contact with you.

The second 100 are mostly new people. That feels earned. I'm honored to have you here. Thank you for subscribing.

To celebrate, I'm going to switch it up a bit. I'd like to follow more of a newsletter format with a collection of articles instead of one big article. That seems more appropriate given the wide diversity in my activities, interests, and readers. Pick and choose or read them all! Here we go!

In this edition:

My Clown-Car Resume: At the outset of a job search, an inadvertent typo leads me to source control and a build process.
A Two-Bit Webpage: In the latest adventure with QR Codes, I'm seeking a Jamstack replacement for Power Pages.
Translating Office Work: A brief note on translation in the AI age, plus my own translation of a timely Julio Cortázar short story.

1. My Clown-Car Resume

Over the past few weeks, I've started to submit job applications.

In doing so, I've attempted to take a careful, dispassionate look at my own resume, trying to assess how others see it, how hiring managers might see it.

Here's how I want you to see it: Diverse range of interests with a broad knowledge base, a commitment to lifelong learning, interdisciplinary insights, exceptional communication skills.

Here's how I imagine you see it: "This resume is a clown car."

Yeah, I get the joke.

I have an undergraduate degree (Information Systems) with an unrelated minor (Music); two unrelated master's degrees, the first an MBA with a dual concentration (Finance and Accounting), and the second a Master of Liberal Arts (ALM -- it's Latin) with a concentration in Foreign Literature and Culture; let's see, what else, I studied the Japanese language for a year; wrote a peer-reviewed academic paper about Cervantes' talking dogs (look it up); and wrote a book (buy it or read a sample) about my father's experiences from boyhood during the Great Depression to joining the US Merchant Marine during WWII and becoming a ship's officer. I've taken courses at Jazz Night School (guitar in jazz combos), Gage Academy of Art (drawing, painting, sculpture, animation), and at Seattle Central College (cooking, translation & interpretation, and Spanish). There's more, I'm sure. And that's just the Education section.

But here's the thing -- I have confidence in myself and my abilities. I have always felt secure enough in my prospects to follow my interests wherever they may lead, regardless of what it looks like on my resume.

Perhaps it's because I've been a digital native since my first 300 baud modem, doing database consulting as a high school student. These skills are locked in, and based on my recent work, improving with age.

I wouldn't trade places with parallel-universe Ivan, the unitasker Ivan, the one who learned one thing and stuck with it. Especially now that artificial intelligence is transforming the world of work by making it more difficult for unitaskers to stake a claim to expertise. The multidisciplinary approach is the way to go.

But it's hard to explain that on a resume.

As I mentioned earlier, I'm sending out resumes, just a few, not too many, only the ones I believe to be an excellent fit. If you're a hiring manager reading this, please know that I selected your job posting with care and attention. (And if you can think of a good fit in your own organization, or in your network, let's connect!)

I know that many job-seekers have figured out how to scale job search using AI and automation to spam every hiring manager in existence. I'm not doing that, even if everyone else does.

Nevertheless, I do use AI and automation, but it's toward the goals of accuracy and targeting. That's fair game.

It's the difference between scale-out and scale-up. Scale-out adds machines to do more of the same thing. Scale-up gets a bigger machine to do better on the most important things.

Here's what I mean by that.

Resume Builder

A few weeks ago, I sent a resume with missing Information. Literally. Due to a trackpad error, I inadvertently deleted the word "Information" from my undergraduate degree, as follows:

Bachelor of Science, and Decision Systems, Minor in Music

Mind you, this was for a job with significant editorial responsibilities, where "attention to detail" was a core requirement.

I resolved to prevent that specific error from happening again, and by doing so, demonstrate my facility and comfort level with the Docs-as-Code approach to documentation.

Before:

Start with a Word document with my most recent resume.
For each new job application, "Save as" a new version.
Edit to address the job requirements.
Save as PDF.

After:

Organize resume content into modular sections.
Convert resume text into LaTeX, convertible to and from Markdown format.
Implement Git for version control to manage different versions of modules and documents
Establish a simple build process to assemble resume from reusable components, with output going directly to PDF.

Benefits:

When changing one section, you cannot break another.
Easier to assemble.
Single repository for all skills, education, and experience.
Ability to maintain branches for different skill areas.

2. A Two-Bit Webpage

To get beyond prototype for my QR Cards solution, I'll have to solve a front-end problem.

Power Pages is fantastic except for the licensing model, which starts at $75 for 500 anonymous users/site/month. If I hand out QR Cards to 1,000 people and it costs $150 per month to keep those cards active, that's $0.15/card per calendar month, plus printing costs. That cost model may work for high-end events, but it's not suited for supporting large-scale deployment for tourism and events. I need it to be pennies per card, not an entire quarter-dollar.

I need a front-end solution with:

· Javascript to handle widgets like maps, videos, image galleries, and interactivity.

· APIs to tap into Dataverse and/or other database engines

· Markup to render webpage content describing destinations and activities

That's the definition of Jamstack: Javascript, API, Markup.

To make this work, I have to figure out a few things: Static Site Generators (SSG), Jekyll, Node.js, Azure Static Web Apps, Azure Active Directory, Dataverse APIs, and a bunch of JavaScript libraries.

Wish me luck -- and if you have any advice to offer, I'm all ears!

QR Cards

Here's a brief explanation of my QR Cards solution:

· A ticketing system generates unique QR Codes, each pointing to a unique webpage.

· An invoicing system shows an "trip ticket" for each QR Code, repurposing the line items to display content elements from the CMS.

· A content management system (CMS) contains all the possible content elements that can be referenced via the invoicing system.

Your guest scans the QR Card (ticket) to see a journey (invoice) pre-loaded with destinations and activities (CMS listings). The QR Cards are pre-printed with default journeys, and since each card is unique, you can customize any card by adding destinations and activities.

How it works: First, design your cards. Next, define simple rules for which content elements you want to associate with each type of card. Then, generate a print-ready PDF that has with all your QR Cards -- and custom landing pages -- ready to send to your local print shop.

3. Translating Office Work

I'm not a professional translator, although I did consider that line of work not too long ago. Perhaps if I had been better at a single foreign language, faster as a writer, and focused on a single industry -- maybe, just maybe, I could have had a machine-aided translation practice to be swept up in the AI automation wave for LLM translation.

It's going to be rough going for translators, and for interpreters too.

At Ignite last month, I saw a demo featuring a pair of translator glasses that shows the wearer real-time translations from any language projected on the inside of the lenses. The glasses themselves are giant and clunky, resembling the military GI glasses ("BCG"), not the sleek futuristic look of Google Glass or look-how-cool-we-are Facebook Ray-Bans. Oh, and if you want to communicate in another language, you speak, your smartwatch translates for you.

And yet I still work on translation. It's one of the best ways to learn a language, especially if you want to read in it.

I expect literary translation to survive the AI revolution. It's a combination of curation, poetry, and empathy -- you have to identify something worth sharing in another language, discover the correct phrasing in your target language, and intuit how your intended reader will receive the work. It's no small task to get all those things right. My translation below is clunky, the work of a language student seeking exact correspondences. I'll get better with practice.

This is my own translation of a Julio Cortázar short story about an overeager secretary who takes over the office, including the very next word you're about to write. It's easy to imagine a contemporary version -- after all, it's the world in which we now live.

Office Work

by Julio Cortázar

My faithful secretary is one of those who follows her duties to the letter, which means crossing to the other side, invading territories, putting her five fingers in the glass of milk to take out a measly little hair.

My faithful secretary takes care of everything -- or would like to take care of everything. We spend the day waging a cordial battle of jurisdictions, a smiling interchange of mines and countermines, advances and retreats, captures and ransoms. But she has time for everything, not only seeking to take over the office, but also meticulously completing her functions. If a dispensable adjective comes to the tip of my tongue -- [e.g.] "Because all of those born outside of the orbit of my secretary, and in a certain sense me myself -- she is already there, pencil in hand, catching it and killing it withing giving it time to connect to the rest of the phrase and survive by neglect or custom. If I leave it in, if in that same instant I leave in in, she will throw those pages in the trash, infuriated. She is so resolved that I live an ordered life that with any unexpected movement, she straightens up, all ears, all erect tail trembling like a wire in the wind. I have to pretend, and under the pretext that I'm writing a report, fill some little sheets of red or green paper with the words that I like with their games and leaps and jumps and furious quarrels. My faithful secretary in the meantime arranges the office, in appearance distracted but ready to jump. Midway through a verse, born so contented, the poor thing, I hear her begin her horrible shriek of censure, and then my pencil gallops again to the forbidden words, hurriedly crosses them out, puts disorder in order, fixed, clean and showing splendor, and that which remains is probably very good, but for this sadness, this taste of betrayal on the tongue, this face of the boss with his secretary.

"Trabajos de oficina" by Julio Cortázar, in Cuentos completos 1: 1945-1966 (Debolsillo, 2016), originally published in Historias de cronopios y de famas (1962).

Discussion questions:

"Faithful secretary" (fiel secretaria) -- faithful to whom?
"...taste of treason on the tongue" (gusto a traición en la lengua) -- treason against what or whom?
How might a totalitarian censorship nightmare evolve with today's technology? What possible mitigation strategies exist?
How would you implement IoT ears and a tail for a chatbot? Provide network diagrams.