Would Altman shine within Microsoft? Seems like raising capital is his main skill set, and theres no need for that now. But from Microsofts point of view this prevents a new competitor from popping up.

I tried that once, but paying changed the ui in some way that I found to be a huge degradation. Dont remember what exactly it was, but now Im using the adblocker again. Just ublock works fine btw, as long as you keep it and its filters up to date.

It would be useful if you can elaborate because as a paid user I've never spotted any UI or UX degradation and I'm now curious.

It might have been a ”continue where you left off” player that always showed on the main page or something like that.

Continuing is never desirable so I always had to waste a click closing that so it wouldnt obscure part of the screen.

Your family is not the intended audience for the code. If the reviewer isn’t either, then they have no business reviewing your code.

I can agree with that. But the intended audience for your code isn't as rigid and formalized a thing as review processes make the reviewers identities.

That actually means that review processes are usually wrong. But then, people experiences are about the wrong process, and that's what they react on.

This only happens when the users are the product or its a marketplace. Most business models dont suffer from shittification. But VCs love investing in the ones that do.

I would say the opposite. Consumers frequently reward the business models that do whatever it takes to lower prices in the short term, often sacrificing the long term.

It is a constant struggle to convince people that your higher quality product/service is worth the extra cost, and obviously, many times it is not.

But the formula for operating a successful, long term business is not as simple as “output the best quality product or service you can, and you will be rewarded”. It is more like “output the best quality product or service you can relative to prices of competing sellers, and at prices your clientele can afford”.

Which may or may not include sellers that have access to much cheaper money (VC, bigger companies with other revenue streams, etc), or sellers operating in different jurisdictions with lower costs.

It’s not that consumers don’t reward quality. It’s that more consumers reward lower prices, and once outside money is involved you have to chase growth instead of steady profits.

Family owned businesses are often able to maintain very high quality. They’ll just never make enough to satisfy people looking for huge returns, which is nearly anyone who doesn’t have some deeper connection to the company.

Nah it happens to all public companies (see hasbro with magic/d&d) even when users are not the product.

It happens when nobody wants to be the first to show flagging growth and investor expectations have not reset yet after a period of growth.

It’s just easier in tech to manufacture growth metrics you cant get checked on

Sure, but in those industries the users leave, the company suffers and the executives get fired. VC are attracted to moaty business models where users cant leave.

> the executives get fired

By fired, do you mean receive multimillion dollar payouts and jobs from their friends at another place where they can do it all again?

d&d has weak network effects. wotc overestimated the effect though

In Canada the once beloved brick and mortar retailer Mountain Equipment Coop shittified itself straight to (pandemic) bankruptcy hard and fast not long after the Harvard MBA types got their hands on the business. MEC was late to the party. Much of mall retail had already shittified itself ages ago.

It seems to be happening everywhere. Give the worst product you can get away with while using marketing to get people to keep upgrading before they realize how bad the product is. While progressively lowering the quality of the product each year. "Planned Obsolescence" is the physical product equivalent of digital enshittification.

For an example of literal enshittification see water companies in England. Not discharging raw sewage into rivers and the sea would interfere with dividends to the PE owners.

This architecture completely ruins your write performance. Theres a reason double databases, one OLTP and one OLAP, is the norm.

Its also that the longer journey increases the variation in arrival time thus increasing the risk of not making the transfer.

Ive gotten some use out of ChatGPT for system design. I describe my problem, it recommends solutions, I ask follow up questions and add more context etc… Just like I would work with an actual collegue. Then in the end I still have to go to the source docs but the option space is often so large that I find this initial back and forth very helpful.

Coding tools, like Copilot, are completely useless on the other hand. I think they would need to be integrated with the compiler and the IDEs indexes to be of real value. Perhaps you would also need another network architecture than transformers to really understand the tree shaped nature of code as well.

Theyre already noticably plataeuing when it comes to coding, which similarly requires you to navigate tree structures and plan how to connect existing things together using shortest paths rather than inventing things out of thin air.

I would love to see the evidence for plateauing when it comes to coding! Specifically, I'd like to see that new larger models are not achieving performance increases relative to previous smaller models.

If you don't want to publish events from uncommitted transactions you'll have to first store them in a local table and then move them to the queue after the commit. But if all consumers have direct access to the database anyway...

The problem is that it should be the caller that decides which category the error falls within, but most languages forces the callee to choose.