terramauthe

yeah, but in 2026 the interaction data of the average user isn't adding much to training. only the really exceptional users.

Wait until people find out this is true of separatist movements in every country.

I use LLMs to find items for me all the time and it's great. I still have to use my human capacities for consideration and kindness as well as my deep understanding for the people in my life who I care about to suggest what gift I might want to get them. I still need to care. The people in my life have very specific needs and finding the right vendor for a gift involves a lot of searching and reading. Not to mention navigating the brilliant tariffs that have recently made shopping such a joy. LLM tools turn this 3-5h process into a 5 minute exploration.

I'm glad to see this being made mainstream. Hopefully nobody imagines that this will let them care less about people in their lives, since we need a lot more people to care about each other in my opinion.

[flagged]

I turned it on, and now geolocation works. That's how it's related.

They added a feature to enable better geolocation (if you want).

In the Starlink app: Settings > Advanced > Debug Data > Starlink Location > Allow access on local network.

This allows local devices to use the Starlink lat/long that dishy uses for satellite targeting.

I would argue that it is post-ironic, especially in the context of Tom Sachs's body of work. #nuance

Why did the title of this link get changed?

I'm sorry that has been your experience with agile. That doesn't sound very good. In my experience, if agile rituals take more than three hours per 2-week sprint, something is not aligned with the team or the project. Maybe the necessary skills aren't there. Maybe the team doesn't trust each other. Perhaps the project plan is not aligned with business goals, etc. None of these problems can be solved by adding more agile rituals.

How I prefer to run it:

- 1h per week alternating between estimating and planning

- (optional) 1h per sprint for backlog refinement

I like to keep two sprints worth of work in the backlog. More than that, the spec will change too much, and you will re-estimate. When a project has just started, or the requirements have become understood differently, to get four weeks of backlog work may require additional refinement. I typically do the bulk of this refinement myself (as a tech lead), and I try to avoid asking the team to have any extra meetings beyond what I mentioned above.

It can be good to have retros sometimes, but it's better to have a team that feels safe to share and discuss issues constructively without a mediator.

I'd also like to advocate for 1:1s. We are all humans and social organisms. Ideally, these 1:1s are 80% unrelated to a work task.