rogerkirkness

7-14 drinks a week is seriously bad for you (e.g. 20 percent higher risk of premature death).

I think the reality is that software engineering, and soon all knowledge work, are going to transform profoundly like you're noticing.

Where we go depends on the philosophy we apply, and there's way too little philosophy for what to do when AI upends your intellectual craft profession.

All businesses exist to serve a customer, and to help them achieve the outcomes they desire, so to the extent that you can apply these skills to those outcomes then you'll continue to be gainfully employed in a business context.

I think zero sum applications of software like internal tools with fixed scopes will rapidly be automated away with these tools. By comparison, positive sum product engineering of novel technologies seem like they are still hiring.

Would focus on bridging your technical skills into product engineering and deeply understanding a specific customer domain, invest in EQ through coaching and therapy and continue adopting AI tools to optimize your work.

PSA: I did a similar thing using Claude about buying a 2012-2016 era Ford F150 and it was so good at providing information like this about each model year. I went with a 2014 since the V8 was most mature and 2012-2013 issues resolved, and got there in 2 minutes instead of having to surf through forums for hours.

The book 'The One Device' covers this in really thorough detail.

Lots of information about how Signal spends donations here: https://signal.org/blog/signal-is-expensive/

iCloud Mail (if you have paid iCloud...) is free to add custom domains.

We signed an annual pricing deal in fall and now our price is going up materially during the annual term because of this change, because we didn't buy Gemini.

The minutes and stock grants take about 2 minutes at the very end.

The real purpose is an assessment of product/market fit, customer adoption, product development and financial results. Each are taken up qualitatively and qualitatively based on an artifact (either memo or slides) created by the CEO and discussed during the meeting.

The CEO creates the agenda, proposes it to the board members who give feedback, then creates the artifact based around the agenda in order of importance. Discussion ensues about the artifact and how much progress is being made, how satisfying that is, and changes/decisions for the CEO to take in follow-up.

It's typically a discussion meeting rather than a decision meeting. It should force reflection and progress with the CEO contriving and making decisions about changes to make to financial, customer and product decisions in follow-up solo. Then the next meeting you check on what effect that had.

Hell yeah, screw TikTok and the horse it rode in on.

(Canadian founder in unrelated domain)

Our premise as a startup is that we should want CEOs that use AI to make higher quality decisions, than either status quo human only CEOs or AGI CEOs that do not have direct liability. The analogy is that planes are seen as safe using autopilot because the human pilot gets on board with you. Societally I think the same thing is true of CEO decision making AIs.

Anyone who has used Claude Sonnet 3.5 New will probably agree. I burned out on annoying uncanny valley ChatGPT issues and Claude absolutely paves over them with buttery smooth high quality results nearly every time in a way that is shocking.

Claude 3.5 Sonnet (New) is meaningfully better than ChatGPT GPT4o or o1.

Commoditize your compliment baby.

Most of the top 1000 e-commerce sites are built on Salesforce, but most of the rest are built on Shopify and it's a very long tail. Salesforce acquired Demandware which was strong at very custom requirements. As time goes on, the number of companies who need that level of customization is closer to 100 than 1000, so customers number 100-1000 is really what this is about.

Disclosure: worked at Shopify

I have two iPhone 13 Minis and I'm just going to keep buying them on Facebook Marketplace as needed. Hopefully Apple has usage retention data :)

His job is to hit the number, so this makes sense.

I find it weird how unserious manufacturing companies are about dogfooding. This is like Larry Ellison saying 'Yeah we actually think Postgres is better so we use it internally but feel free to buy our database if you want, just not my preference' it's unthinkable. Tim Cook wearing Meta Quest...

Adguard + Safari blocks pretty much everything, but helpful context on uBlock.

Using Adguard and it works well.

Already switched to Safari a while ago over this decision by Google. You know that the privacy stuff works because third party Javascript only sometimes does lol. No longer just the best browser to download Chrome with...

Bill Gates still runs product at Microsoft, bad counter example...

When you run out of the will or resources to continue, and when you have run out of ideas to make the situation better.

Apparently Steve's posthumous roadmap focused on the idea that personal computers get 'smaller and closer to you' as time goes on. So the idea that an Apple Watch and AirPods could be all you need when travelling, etc. follows that premise.

Customer development is when founders cold email (or Tweet, whatever is appropriate) and talk to customers trying to test the following in the following orders:

1. Is the problem that you're going to solve important? Do people respond to cold email about a hypothesis to solve it? (testing problem relevance)

2. Is the solution that you put forward realistic sounding to customers? Do they believe that you'll be able to build it, and want it if you do? (testing solution relevance)

3. When you show customers an early version of what you're building, do they ask to use it (testing problem/product fit)

4. When early users get access to your sketchy prototype, do they usage retain? Regardless of funds changing hands, do they keep using it?

If you have a product where you can email people about a problem, who consistently reply, see a demo of your product and ask to use it, then adopt it and continue using it without pressure, you should figure out pricing.

Casablanca

How do you think the S&P500 was able to increase profit per employee by 5x?

Just tax revenue or profit per employee, which has risen ~5x in the last 25 years among the S&P 500.

When all your problems can be solved by money, it buys happiness. When none of your problems can be solved by money, the perception is it isn't impacting your happiness, but I bet getting rid of it would impact your happiness a lot...

Impressed by the confidence of 'to AGI and beyond'.

Highly ineffective CEO responsible for destroying a huge amount of value that he originally played a major role in creating. But arguably destroying more value than he himself created. Should probably have made Zach Kirkhorn CEO instead.