Substrate

A living collection of notes, ideas, and reflections from Doug Belshaw.

reflections updated

On 'stability'

I'm trying to stay off social networks, but occasionally people DM me on LinkedIn, I get an email, log in, and inevitably have a quick scroll of my home feed.

So I saw this by Dave Gray, who is absolutely correct when it comes to "having a job" vs. "running your own business or being a freelancer/contractor/entrepreneur". You can see the risk in the latter situation, whereas it's entirely hidden from you in the former.

reflections

Quotations from 'How to Do Nothing'

I'm re-reading Jenny Odell's excellent book How to Do Nothing at the moment. Here are some of my highlights (emphasis, where it appears, is mine).

Nothing is harder to do than nothing. In a world where our value is determined by our productivity, many of us find our every last minute captured, optimized, or appropriated as a financial resource by the technologies we use daily. We submit our free time to numerical evaluation, interact with algorithmic versions of each other, and build and maintain personal brands. For some, there may be a kind of engineer’s satisfaction in the streamlining and networking of our entire lived experience. And yet a certain nervous feeling, of being overstimulated and unable to sustain a train of thought, lingers. Though it can be hard to grasp before it disappears behind the screen of distraction, this feeling is in fact urgent. We still recognize that much of what gives one’s life meaning stems from accidents, interruptions, and serendipitous encounters: the “off time” that a mechanistic view of experience seeks to eliminate.

reflections updated

TechFreedom pilot cohort feedback

We had five participants in our pilot TechFreedom and four gave us feedback after the final session. using Tom's Drift forms which has a built-in AI summariser:

The responses show remarkable consistency in quantitative ratings, with all participants scoring 5/5 for clarity of technology dependencies, 4/5 for confidence using lenses, and 4/5 for overall value. Lock-in and cost exposure emerge as the most commonly selected relevant lenses (3/4 responses each), suggesting these are universal concerns across organization types. The most significant pattern is the gap between programme satisfaction and implementation likelihood: despite high NPS scores (two 10/10s, two 8/10s), only one organization has begun acting on their roadmap, with others citing lack of time/capacity as the primary barrier. Participants particularly valued the discussion format and visual/collaborative elements, with multiple requests for enhanced peer interaction through breakout rooms or conversation time outside main sessions. There's a clear demand for better resource distribution and continuity mechanisms—participants want materials emailed, pre-session agendas, and post-programme support like peer networks or periodic prompts. Technical feedback centers on the Stackmap tool needing improved saving/exporting functionality. The Deep Risk Assessment received one piece of negative feedback as less useful than other exercises. Overall, the programme appears to successfully build awareness and provide frameworks, but participants need additional scaffolding to bridge from learning to implementation within their organizational constraints.

reflections updated

Building a world instead of an audience

I guess this is what I'm trying to do with Substrate - a digital garden ecosystem:

instead of “building an audience,” build a world. build a digital garden-ecosystem, that exists — first and primarily — for itself. a world that doesn’t need likes, traffic, subscribers, or clicks — in order to validate its existence.

reflections updated

Using AI to write about the dangers of AI

We shouldn't unquestioningly believe "AI detectors" but nor should we be so naive to think that anyone's immune form the temptation of using AI to write very long documents.

It’s possible that AI was used to write parts of Pope Leo XIV’s latest encyclical about AI’s impact on humanity. An analysis by Linch Zhang posted on the forum LessWrong found certain paragraphs of Magnifica Humanitas to be between 40 percent and 100 percent written by AI, according to the popular AI detector Pangram.