> A redesign that gets replaced 2 years later is a catastrophe.

People forget how quickly Uber scaled, and the user impact of not being able to track your trips could be catastrophic to retention. There's a class of tech-influencer who think they can dissect past decisions on a blog post without being in the room when the technical constraints were being laid out. This is Monday morning quaterbacking at it's most grotesque.

> There's a class of tech-influencer who think they can dissect past decisions on a blog post without being in the room when the technical constraints were being laid out

The cost they are laying out are not that prohibitively expensive. I’ve known corporations where people spin up test clusters that cost 5K a month and forget about it. A business critical service can definitely ignore costs in the short term if they bring in customers. The standard practice is to just ship something quickly and optimize for the cost later if it helps bring in revenue/customers.

Besides, the napkin math isn’t always true. If you’re an enterprise customer for AWS, you get massive discounts, especially in the time frame they’re talking about. And when it comes to partnerships, I remember back in the day AWS used to let you do pretty much anything for free if it meant they could parade your project to other customers.

Yeah, I don't know the specific trade-offs here, but taking on some tech debt for two years of successful growth is not necessary a bad deal.

You can read the article and see it's not a tech debt trade off but someone not doing a back of the envelope guesstimate about how much DyanmoDB would cost to run their payments system on it.

I think X paid for itself, so it worked our for him.

  > Nikita Bier @nikitabier
  >  
  > If you’re seeing a bunch of Japanese posts, here are some fun facts:
  > Japan has more daily active users and more time spent on X than any other country in the world.
  > Over two thirds of the country is monthly active on X. 
  > X in Japan has one of the highest penetration rates of any social network in history.

I wouldn't be so sure when "any other country except US" usually apply to everything on the Internet, except Twitter after the power transfer

1: https://twitter.com/nikitabier/status/2037764895064867061

I'm pretty sure that claim about Japanese Twitter activity was true for most of the site's history pre acquisition

No. JP activity was always second to US, only the biggest "out there". Same is true for all Twitter-like social media, such as Mastodon and Bluesky. Even VRChat doesn't have a majority Japanese userbase. Japan actually becoming the top majority anywhere is an anomaly and a major reversal of power balance.

Still blows me away that Google had complete dominance in Brazil and then just threw it all away and shut it down a few years later.

Google Plus? I wouldn't be sure if that was a strategic blunder or if they were seeing something us in the public didn't. I remember it was more popular among not-so-tech savvy male of parental to retirement ages, which are still masses but not the sweet spot in terms of demographics. Besides they have YouTube and its comment section full of kids, which is the sweet spot.

Orkut, which nobody now remembers.

Orkut

Source?

It paid in influence, not dollars. Billionaires don't buy newspapers or social media platforms because they think they are good businesses.

Usually they would just use an off the shelf product and extend it, so they wouldn’t produce the absolute horror story described in the article, no.

I’m not even sure what your last comment means, are you contending that it is a good thing this company violated multiple laws with sensitive patient data?

> Usually they would just use an off the shelf product and extend it

AI does the same thing an agency or dev would do. Those vibe coding platforms have a template for these things which is usually Vite + React with Supabase for the backend, the same as a dev might use because surprise the LLM trained on the dev's work.

OP's point is that you're not guaranteed a good outcome hiring an agency or solo dev either, in fact I would say you're almost guaranteed a bad outcome either way.

Apparently your assumptions about AI are completely wrong, if you read the article it produced terrible code.

Right, because the user provided terrible prompts. A dev / agency wouldn't have done much better.

There's lack of security theater and there's:

> All "access control" logic lived in the JavaScript on the client side, meaning the data was literally one curl command away from anyone who looked.

They are not the same thing.

You've got to wonder from where did the "AI" parroted that.

A Stackoverflow wrong answer?

If a consultant made the same mistakes I'd expect the consultant to be held accountable, not the client business that hired the consultancy - they knew they didn't have the requisite skills and so outsourced to an "expert" (and therefore can't be judged for not knowing how to secure their software since they did everything possible)

In this case the "client" is fully liable for the security issues.

It is possible. If you select consulting that you know nothing about, and they know nothing about programming and vibe coded it for you... and maybe you dont even have a contract to held them responsible and maybe they dont really have a company either... Then I can imagine something like this.

It is physically possible for a consultant to write bad code. But you'd hope that a consultant could understand that medical data is extremely important to keep secure, and actually write it to have some level of security

Sure, but you'd hope that the LLM could understand that too.

And yet it seems it didn't

I'll bet you anything it did. If you ask any recent coding model to do this project is will strongly advise you to give up.

I appreciate a good UX improvement, and I respectfully appreciate you building this for your own use case.

But Dawg? Really!

If we don't innovate, someone else will. This is the very nature of being a human being. We summit mountains, regardless of the danger or challenge.

>If we don't innovate, someone else will.

Terrible take. You don't get to push the extinction button just because you think China will beat you to the punch.

>This is the very nature of being a human being. We summit mountains, regardless of the danger or challenge.

No, just no... We barely survived the Cold War, at times because of pure luck. AI is at least as dangerous as that, if not more. We have far exceeded our wisdom relative to our capabilities. As you have so cleanly demonstrated.

Humanity stopped germ-line human genetic engineering (possible since the early 1970s) and humanity can (and should) stop OpenAI, Anthropic, etc.

Datacenters that use literal gigawatts of electricity are not exactly easy to conceal from the authorities.

Until recently Claude wasn't building itself. A group of people with agency were.

Isn't it a little weird that we trust this app to help us build some of the most important parts of our business and the company that vends this app keep breaking it in unique ways.

At my workplace we have been sticking with older versions, and now stick to the stable release channel.

I like dogfooding. You can use Azure if you want infra that is clearly not being used, tested, and pushed to the limits by its own creators.

Yeah, sounds like a skill issue.

There are a plethora of models that you can use with open code. Anthropic is well within its rights to limit third-party usage of services that violate the TOS. As a Claude code user I’d much rather have the very best experience on Claude code than the largest supportability matrix for Anthropic models.

as someone who has used codex/claude-code/opencode I can confidently say that "the very best experience on Claude" is not the one that Anthropic provides software for.

Well, the challenge for Anthropic's users is: While Anthropic has fantastic coding models right now, they have a bad harness in claude code and the claude desktop app. The best experience using Anthropic's models is OpenCode and Cursor. This wasn't true ~3 months ago, and may return to being untrue in ~3 months, that's how fast these things change, but right now this is the case. Unfortunately, Anthropic models in OpenCode/Cursor are tremendously expensive; and that gets at some of the leading theories on why CC has degraded recently; that Anthropic has been forced to dynamically route more of the agentic process onto Sonnet/Haiku, or reduce the Opus thinking budget.

For all of these reasons, currently, the meta is ChatGPT subscription on Codex or OpenCode. But, again, these things change every few weeks.

I don't think this is as clean-cut as just saying "Anthropic is in their rights" etc. Of course, they are, to whatever degree they are; the bigger problem is that you've got $100/mo and $200/mo Claude subscriptions who are actively saying "the $20/mo Codex subscription is better in every way", possibly because of these thinking budget/routing changes people suspect have happened this month. In other words: Anthropic is at-capacity after the DoW incident, they need to load shed, and they've chosen to harm their high-paying power users and Enterprise over just temporarily slowing growth a bit by hurting the $20/mo plan. And, frankly, they might be right: because for every $200/mo user that jumps over to ChatGPT, half of them will be back once Anthropic can scale capacity, and if they can gain 20x $20/mo users who only use half their sub, that's a win.

Yeah, the article is slop. Not even AI slop. More like guilty conscience slop.

I ran an experiment where I set DDG as my default on all surfaces. About 3 - 4 months in, I actually started hating searching, and a few weeks later most of my queries had the prefix !g

Gmail is hands down the best. I pay for Gemini, and Gemini outside gmail is much much better than Gemini inside gmail. I pay for ChatGPT, but for some reason, I trust Gemini with my email rather than ChatGPT.

They could likely increase their budget slightly and run an LLM-based judge.