Some of the comments point toward genuine concern, some smell of gatekeeping.

It is interesting to see a flip in attitude toward GitHub.

There were ~trends similar to what appeared early in TikTok.

For example, early TikTok had the Boss Walk.

Sora had no big content trends split into many micro trends in some established ~universe.

Well, that stuff goes viral because it’s fun to imitate, all the dances and challenges provided a flywheel to get people creating more content, it’s fun to make the video.

If I see an AI video and my options to participate are… prompt another AI video? What’s the point

I think you’d have to participate, but this was a real thing and it was fun.

For example, there is one that was a making sort of Macy’s thanksgiving day-style parade floats.

And you could pick virtually any content type and see this interpreted as “real” floats.

It did not require a ton of effort, for example you could reply to the above example existing with the prompt “do this, but have the float theme be ‘meet the feebles’”

And if you know of that film and recognized the AI’s interpretation of it in that context and it was half decent it was entertaining.

Not all “trends” if you can call them that were so simple to do well with.

Often the prompts needed to be elaborate and required multiple generations to really get a feel for if you were on the right track.

I think they did have something here and probably someone will do it again and it will work.

You're supposed to press the button to receive dopamine. It's all just narrower and narrower Skinner boxes.

An AI video trend on Instagram as been Han from Tokyo Drift with different cars. People still want to share those on the platforms they are already locked into with their friends.

The models are not well trained on bringing products to market.

And even “product engineers” often do not have experience going from zero to post sales support on a saas on their own.

It is a skill set of its own to make product decisions and not only release but stick with it after the thing is not immediately successful.

The ability to get some other idea going quickly with AI actually works against the habits needed to tough through the valley(s).

Claude Code breaks production features and doesn't say anything about it. The product has just shifted gears with little to no ceremony.

I expect that from something guiding the market, but there have been times where stuff changes, and it isn't even clear if it is a bug or a permanent decision. I suspect they don't even know.

No IP will stand up to AI, from Star Wars to Linear. Things are about to change.

Indeed, like MagSafe charging—they simplify. Simplicity has a premium.

That said, my first pair failed out of Apple Care and resulted in a full cost replacement. The APM sub is littered with stories of the BT module failing.

I’m sure it is ludicrous to some but I often measure value by utility and I go through entire workdays wearing this product.

They will replace worn out fabric on the v1s with AppleCare. Have to send them in but it works.

I also do this. My primary use case is for reproducing page layout and styling at any given tree in the dom. So, capturing various states of a component etc.

I also use it to automatically retrieve page responsiveness behavior in complex web apps. It uses playwright to adjust the width and monitor entire trees for exact changes which it writes structured data that includes the complete cascade of styles relevant with screenshots to support the snapshots.

There are tools you can buy that let you do this kind of inspection manually, but they are designed for humans. So, lots of clickety-clackety and human speed results.

---

My first reaction to seeing this FP was why are people still releasing MCPs? So far I've managed to completely avoid that hype loop and went straight to building custom CLIs even before skills were a thing.

I think people are still not realizing the power and efficiency of direct access to things you want and skills to guide the AI in using the access effectively.

Maybe I'm missing something in this particular use case?

> There are tools you can buy that let you do this kind of inspection manually, but they are designed for humans.

You should try my SnipCSS Claude Code plugin. It still uses MCP as skill (haven't converted to CLI yet), but it does exactly what you want for reproducing designs in Tailwind/CSS at AI speeds.

https://snipcss.com/claude_plugin

its mostly because MCPs handle auth in a standardised way and give you a framework you can layer things like auth, etc on top of.

Without it youre stuck with the basic http firewall, etc which is extremely dangerous and this is maybe the 1 opportunity we have to do this.

And people forget, Claude Code isn’t the only Claude surface, and CLIs don’t help in other surfaces other than Cowork.

> My first reaction to seeing this FP was why are people still releasing MCPs?

MCPs are more difficult to use. You need to use an agent to use the tools, can't do it manually easily. I wonder if some people see that friction as a feature.

This presumes he knew he would be held that long.

Presuming he holds keys to vast wealth, the calculation would have shifted over time. Especially once he was serving his original sentence again starting a year ago.

Another consideration is that many go to jail longer with no upside once getting released.

Me too.

I used to report bugs, read release notes; I was all in on the full stack debug capability in pycharm of Django.

The first signs of trouble (with AI specifically) predated GitHub copilot to TabNine.

TabNine was the first true demonstration of AI powered code completion in pycharm. There was an interview where a jetbrains rep lampooned AI’s impact on SWE. I was an early TabNine user, and was aghast.

A few months later copilot dropped, time passed and now here we are.

It was neat figuring out how I had messed up my implementations. But I would not trade the power of the CLI AI for any *more* years spent painstakingly building products on my own.

I’m glad I learned when I did.