> ## Content Index
> Fetch the complete content index at: https://www.uwsdad.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# The family tech stack in five questions
- URL: https://www.uwsdad.com/family-tech-stack-sorted/
- Published: 2026-10-31T13:51:25.000Z
- Updated: 2026-10-31T13:51:25.000Z
- Description: Two months into the school year and only one missed publishing week, the tools are multiplying. Here is a five-question framework for deciding what comes into your house - and what to do with what's already in it.
- Author: John Ratcliffe-Lee
- Tags: newsletter, tech-dad, fall-2026

Two months of school are in the books. The folder is full, conference day is this week, and the tech is multiplying: the school rolls out a portal you never asked for, a grandparent gifts a screen, an AI assistant promises to organize your family, and suddenly you are the household IT department nobody elected.

This issue is the tool support for that. Not a gadget review. A framework: five questions, asked in the same order, every time a new device, app, or agent shows up at your door. I'll run it on what's actually in our house right now, because the framework is only good if it survives contact.

## The five questions

**1\. Where does the data go?**  
  
Not "is it private" - that is unanswerable - but "where does it go, and who holds it?" A calendar entry about a therapy appointment, a photo of your kid, a voice note: each one lands somewhere, with some company, under some policy. Know the destination before you type.

**2\. Who can see it?**  
  
Household products love the word "family." Check what that word means in the settings: can both parents see everything, can the kid see the parents, can an account you invited for one thing see another?

**3\. What does it cost to leave?**  
  
Every tool is easy to join. The honest test is the exit: can you export your data? Does the device stop working if you cancel? A tool you cannot leave is not a tool you bought; it is a place you moved into.

**4\. What job does it own?**  
  
The failure mode in a two-parent house is tool sprawl: a shared calendar, a chore app, a group chat, a photo frame, each half-owning "family organization." Each tool should own one job and clearly lose to another tool for every other job.

**5\. What's the exit plan?**  
  
Not for you - for the technology. What happens when the company pivots, the subscription lapses, or your kid outgrows it? Decide in advance what "done" looks like, because the default is that nothing ever gets deleted and the household runs on legacy systems forever.

## Worked example: the family calendar

Our house runs on a [Skylight Calendar on the paid Calendar Plus plan](https://myskylight.com/calendar/?ref=uwsdad.com) (no referral code because this is the best there is on the market, not the best product it could be) - the countertop family screen that syncs everyone's schedules. Run it through the framework:

- **Data:** calendar events, chore lists, reward points - low-sensitivity by design, which is why it fits the job. The rule in our house: the family calendar holds *logistics*, never *context*. "Dentist 3:30" is fine; why the dentist visit is happening is not.
- **Visibility:** both parents see all; kids see what the screen shows. For its job, that's right.
- **Exit:** if we cancel, we are back to paper and phone calendars - annoying but survivable. That's a tool you bought, not a place you moved into.
- **Job:** it owns *the day's surface* \- what's happening today, whose turn it is. It does not own reminders, meal planning, or anything that needs a thought rather than a glance.

That last line is the whole framework in miniature: the tool owns one job, and you know exactly which jobs it lost.

## Worked example: the kid's audio player

Our six-year-old has a [Yoto - the screen-free audio player](https://refer.yotoplay.com/inbox4!f27f8e73e8!a?locale=en&ref=uwsdad.com) (referral code because I recommend this IRL) that reads cards you slot into it. Framework view:

- **Data:** mostly nothing - it plays audio. The account holds what he listens to, which is about as low-stakes as family data gets.
- **Job:** owns *independent listening*. Not background TV, not screen time - the thing a kid can operate himself at 6:45 a.m. without waking the house.
- **Exit:** the cards work offline; the platform could vanish tomorrow and the good content is still on the shelf. That's rare in kids' tech.
- **The hidden bonus:** Yoto has an open development surface - an official API, a "Make Your Own" card system, and an active community building custom content. That matters when the answer to question 3 is "you can leave easily" AND "you can build on it." Few kid products get both.

Which points at where this is going: our household experiment for the winter is a small pipeline that takes the day's schedule from the family calendar and turns it into a morning audio briefing or a printable page the kid can read at breakfast. The framework decided *that* the experiment is safe - the data is logistics, the visibility is household, the exit is a single script. Whether it's *worth it* is a test for December, and if it works, you'll read about it here.

## Worked example: family AI, done deliberately

The newest pressure on this stack: personal AI agents and assistants flooding the market, each offering to hold your family's context - schedules, preferences, the thing you told it last week. The framework's verdict, run honestly:

- **For disposable work** \- homework explanations, a rainy-day list, "explain tectonic plates like a subway map" - use the market tools freely. The data is throwaway; convenience wins.
- **For sensitive context** \- the family's actual record: health notes, school details, the stuff you would not put on a postcard - the market tools' retention policies are somebody else's decision. That context belongs in something you control, on a timeline you set, where deletion means deletion. Roll your own for that layer, even if it's just a document and discipline.

The insight worth keeping: **the question is never "AI or no AI." It is "which layer of family information does this system get to hold?"** Same framework, new category - which is the point. When the next thing arrives (and it will, likely a school app or an agent promising to "run your family"), you already know the questions. It takes ten minutes, and it keeps you from being the household IT department by accident.

## The week, in one place

- **Sun Nov 1:** clocks fall back at 2 a.m. - set Saturday bedtime 30 minutes later and the Sunday wake-up is survivable
- **Tue Nov 3:** Election Day - school is in session *remotely*; charge the device and test the login Monday night
- **Thu Nov 5:** parent-teacher conferences, afternoon and evening (elementary) - ten minutes with a plan beats ten minutes without one
- **Sat Nov 14:** STREB opens at the New Victory (booked? book it)
- **Sun Nov 29:** Beatles for Kids at Brooklyn Bowl, Thanksgiving weekend
- **Ongoing:** holiday food orders - if you haven't placed yours, this is the week

What tool in your house would fail question 3 - the one you couldn't leave if you tried? Replies stay private; the aggregate answer becomes a future issue.

*Sources: product details verified against vendor pages (skylightframe.com, yotoplay.com, yoto.dev) this week; pricing stated from our own household subscriptions.*