I was expecting they'd send a reminder per domain that's expiring and not some general mail like this.

Imho the only one that works is Basecamp, but somehow it doesn't reach audiences beyond the West.

Linear is also up there too and I WISH more comps used it, but the pricing doesn't cut it.

The US tech scene also mostly ignores regional pricing and support, so there's that too.

I agree with this, especially for person-to-person communication.

Use AI as you see fit to make it easy to understand or learn, but once you're hitting reply or about to hit send, the text box better be written by hand. Put it in your own words.

The only exception here is a language barrier.

I always think: what does India, China, Indonesia, Brazil and other countries think about the stuff you're strongly pushing for? Do you really think they share the same sentiments?

I feel it's like when AI art became common and people had to "filter the noise" and learn to spot original vs AI.

It's sad, because I know some folks use AI not for clout but to overcome language barriers and join in discussions but they're also likely caught in this.

I agree with both points.

We've tried plain issue tracking before but we ended with Jira anyway because more people could use it and knew its value.

Having Jira brought everyone together. Unlike getting GitHub / GitLab where the tech people were separated from management.

I've been noticing this more on Reddit and beyond.

It's especially frustrating when it's clearly pasted from ChatGPT complete with emoji headers.

I picked up LM Studio because they actually got MLX working and didn't store their models in some godawful structure. Ollama failed in both.

I was also pleased that UX stuff from before were improved. Even more than what I expected.

The relaxed stance on work use is good too.

Review with kindness imho is the difference for people helping out or not at all.

I did a PR before where I wanted to help but the maintainer just commented and closed it because they didn't agree on parts of it that could've been easily corrected in a commit.

The outright dismissal turned me off.

I ranted more of this here: It's especially ironic since I was writing in rage while the editor I used (Notion) is equally annoying in this regard.

Impressive. It's uncanny but imho it's getting closer to "good enough" stage.

I feel it's uncanny since it has a 😮 energy or perhaps head movements are too stable.

I wonder how this looks compared to IRL to properly judge. Also if it looks much better while seeing it with the Vision Pro on.

Oof. Yeah, I can relate to this strongly. It should be common courtesy to message first if you're calling for the first time but some folks really just don't.

And they know I'll pick up anyway since it could be something else, like a delivery.

Great interview. I like the use of the term "jagged frontier". I've been trying to explain to peers the same stuff and this helps.

Nice prompt btw. I've been using something like "make a mindmap in bullets" but I've never thought of asking links / timestamps and best quotes like this.

Depends on how you define "watch".

It's safe to assume they check anything and everything since Content ID became a thing. Then it expanded when speech-to-text and Gemini became common.

Maybe humans weren't involved before, but with AI and how it feeds on feedback? It's now debatable.

Man, this really captures what I think about the AI discussion as well and yeah, I dislike it when things are either no longer civil or untrue or both, and on both tribes.