Oh good timing! I’m about to pivot back around to it.

"Amazon pointed at EPA data from 2017 that said Americans use 9 billion gallons of water daily just for landscape irrigation. This amounts to nearly 3.3 trillion gallons of water every year just for watering plants and gardens and doesn’t even include the irrigation needed for food production."

Amazon is comparing their use of water to cool machinery for corporate profits to humans using water to feed and nurture life. Ok.

I’m actually finding it hard to understand what has been lost. As you said, clearly something has. But, what that thing is, is hard to pin down. Maybe we can’t know until time passes and we have the benefit of looking at this time as “history”.

Nvidia. Pivoting to local first.

To be clear, the developer in me wants everything to be air tight and buttoned up. But, the human in me longs for the space between the lines. So many beautiful things over the years have been created out of the debates and relationships that result from the incompleteness and ambiguity in what we make. In our current AI predicament, I feel keeping that is almost essential.

If machines are to continue to serve our needs rather than us serving theirs, specifications and standards should at least reflect their origin: humans; gray areas and human phrasing over strictly defined, unit testable formats.

The more we make our writings mimic the machine, the more we fade. Struggling to understand ambiguity keeps us in control.

The <enclosure> tag (and RSS 2.0) is beautifully ambiguous. Yet, podcasting is a global force built upon it.

Pi is an amazing piece of software. The developer came from the game dev world so the UI is rock solid without the glitchiness you see in other harnesses.

I think the wonder of the earth bending part has now fully worn off for me. All I see now is the drudgery of the machine writing all the code and me, essentially, just filling out a survey.