I’ll be out of town when this sends so I won’t have my full week of content consumption to work with, but don’t worry, I still found plenty of interesting concepts to ramble about.

Topics for this week:

• divs.holding

• The Problem with AI Coding

• CrowdStrike Further Explained

• Seeking Easy

• Authenticity on the Internet

divs.holding

Divs is in a bit of a holding pattern at the moment as I’m waiting for things in the works, but I’m taking the time to evaluate how to move forward.

I’ve experimented with 2 different businesses now and I’m considering which to really focus on.

Not sure I can operate a web design service, a consulting service and still find time to build things I want to build, so there may be a shift to focusing just on divs.cloud for now and divs.design may have to patiently wait.

I do think a web design service would be fun, but it’s not the thing I think about all the time. It sounded more like a fun and “easy” business but in reality I’m guessing I’m just another victim of uninformed optimism which I’ll touch on more below.

The Problem with AI Coding

I’m having trouble finding the clip, but Thor of PirateSoftware makes a joke about using AI to generate code and then spending 3 hours to fix the code that was generated in 1 minute, and that sums up part of my day yesterday.

I’ve continued writing with Sonnet 3.5 to generate code as I hack away at my own project, but I was lulled into a false sense of security.

I asked for something simple, like adding a spinner to a page to show it was loading while I waited for a response from GPT, but the code that Claude provided was just a little bit off.

It gave me some extra styling that I didn’t need, which caused the spinner to not show up. It was on the page but it remained invisible.

After digging through the htmx documentation, I was able to figure out that I had duplicate CSS code that I didn’t need because it was already provided by htmx. Once I removed the extra code, it worked beautifully.

AI generated mostly correct code in a few seconds but took about 20 minutes of investigating to figure out it wasn’t quite right.

Not the best trade off considering I could have looked up the code and tested it in a few minutes rather than 20 minutes.

It was a harsh reminder of something I tell people a lot: AI is a tool, but it shouldn’t be the source.

I’m still optimistic about coding with AI, but it seems there’s still plenty of room for improvement before businesses exclusively use AI to write code instead of people.

CrowdStrike Further Explained

After last week’s CrowdStrike incident, Fireship neatly summed up the situation in a satirical 6 minute recap.

He points out the same question I did in wondering what happened in the testing process to let something like this happen which just goes to show how glaring of an issue it seems to be.

CrowdStrike did release a technical blog post about the incident, but it doesn’t give much more detail beyond pointing out the specific files that caused the issue.

I’m sure more details will slowly emerge as Fireship highlights, with more engineers digging into the issue, but for now you can start with his brief summary:

Seeking Easy

This short popped up in my feed and hit pretty hard.

Most people are seeking easy, and I completely understand, I’ve fallen for the trap more than I’d like to admit.

divs.design was rooted in the idea of easy. It sounded like something that would be “easy” to make money from, and listening to people that have been running successful design agencies for years make it sound easy because it has become easy for them.

Thinking it sounded like easy money, I was unknowingly tricked once again by the illusion of “uninformed optimism”.

“All these people are making money from this thing, it sounds so easy.”

Then once you jump into that thing you realize, “oh, this is more complicated than I thought”.

Suddenly it doesn’t feel so easy to make money from this new idea, it’s stressful and overwhelming.

But there’s this new idea I saw on Instagram. That sounded way easier. Better try that instead.

And the cycle repeats.

Tim Ferriss wrote about this Transition Curve 16 years ago from a start-up perspective, but with the evolution of the internet and “make money online” culture, the concept has also evolved to much smaller and faster ideas like making money with TikTok and selling t-shirts on Etsy.

Again, the allure of easy money is hard to resist so I completely understand it.

Money is kind of necessary to get by in life so it’s hard not to want to make easy money fast.

But the easy money will be the most competitive because it’s easy for anyone to get into.

And whatever that easy thing is quickly becomes not easy anymore due to competition or over-saturation.

But if you pick the hard path from the beginning, you immediately put yourself in a situation with much less competition just by the fact that it’s not easy.

Or if you stick with it long enough you reach the other side with informed optimism, telling everyone else how easy it is.

I’m being reminded to go back to the hard path rather than be tempted with yet another Transition Curve.

Hopefully that made sense, but in case it didn’t, Alex Hormozi sums it up quite succinctly:

Authenticity on the Internet

I was listening to Devin Nash’s mastermind call the other day, and he made a pretty wild prediction which I happen to agree with.

It’s only going to get harder and harder to tell what’s real on the internet.

AI will get better at producing text, voice and video, and relatively soon it will be extremely difficult to decide what’s real online and what’s AI generated.

With that, Devin recommends focusing on connecting through social media more than anything else, but also making it clear your social media account is run by you, a human being.

Get in the habit of posting videos of yourself raw and unedited, the more awkward the better, to prove that you’re a human and not generated content.

People will be much more likely to trust and work with an account that seems to be run by an actual human.

Authenticity will be more important than ever, not looking perfect and polished, but looking real and relatable.

It’s an interesting concept and I think it’s sound advice.

Time to get back to making some videos for jaypeters.dev…

Might be something to look into for yourself if you haven’t already!

And that’s it for this week! Deciding on a direction, seeking mastery, and standing out in an increasingly fake internet. Pretty dense for missing 2 days of content consumption.

Those are the links that stuck with me throughout the week and a glimpse into what I personally worked on.

If you want to start a newsletter like this on beehiiv and support me in the process, here’s my referral link: https://www.beehiiv.com/?via=jay-peters. Otherwise, let me know what you think at @jaypetersdotdev or email [email protected], I’d love to hear your feedback. Thanks for reading!