kristianc

It’s probably still there, but it’s 26,000 light years away, so the light we’re seeing today left Sag A in the Paleaolithic era.

Creates another expensive bar for prospective challengers to clear - regulation nearly always ends up favouring incumbents, especially if they have a hand in shaping it.

> Fact is, vibe-coded projects devolve over time into an unmaintainable mess. The reason is simple, yet hard to fix: code maintainability and good architecture don’t have good measurements that we can apply, because it takes months, years even, to notice the effects of bad architecture or of unmaintainable code.

Oh boy do I have some news for you about legacy codebases.

I’m not even sure that his axiom is true. A project started with a 2024/5 model can subsequently be worked on by more capable models (who don’t, unlike some humans, have a deep aversion to paying down technical debt). I’ve seldom seen a human written legacy application spontaneously acquire a better engineering team every six months.

That’s the bargain Dropbox made when it took VC and then went public. Public markets don’t price a company purely on today’s cash flow, they price the expected stream of future cash flows. If growth stalls and the attempts to create new growth vectors don’t work, the valuation adjusts accordingly, and at some point an acquisition can become the rational outcome.

Dropbox will be one of the next Bending Spoons acquisitions. Low growth, none of the upsell or product extension things they've done have really worked out, and no-one's idea of what the future looks like.

For me it's a trust gap. Because I know the website UI has been automatically generated, I can't be sure of how much of the product is actually real, or how closely the claims on the page have been audited.

Whose strategy is explicitly to hold good stocks for a very long time.

It's interesting as grill-me and the other Matt skills are very much positioned to people who would consider themselves as developers. In fact, I'm not sure if people who were completely new to development would have heard of him at all.

Ed West’s longstanding intellectual project is fairly conventional reactionary/retvrn/conservative pessimism, and he has no particular technology background of all to speak of.

The argument is basically here is what happened to civilisation X when it abandoned virtue/cohesion/tradition; therefore contemporary liberalism, immigration, feminism, secularism, etc. are leading us towards the same fate.

No interest really in people whose project it is to launder reactionary opinions through ancient history.

To me it has a writerly New Yorker vibe to it, as in the magazine which reads as “polished” and probably performs well in RL but is totally exhausting to read in long sessions and completely inappropriate for coding where precision is paramount above all. In writing terms its called purple prose.

https://en.wikipedia.org/wiki/Purple_prose

Reading between the lines, you kind of get the picture...

2006 A spectacular turnaround sees a 10% sales decline turned into growth of 20% year-on-year by March 2006, making Iceland the UK’s fastest-growing food retailer.

2005 The Big Food Group is taken private and Iceland returned to the management of Malcolm Walker and other senior executives who had been ejected in 2001.

2004 The Big Food Group is nearing bankruptcy as provisions made in 2001 come close to exhaustion.

2002 Iceland-Booker is renamed The Big Food Group and launches a grandiose recovery plan (Click here to read the saga of ‘The one, two, three, four, five year recovery plan’) but customer numbers and sales remain in steady decline while costs escalate.

2001 New Iceland chief executive Bill Grimsey issues a massive profit warning, and Malcolm Walker and other senior managers are forced to leave the company

Score-settling by the ousted (and then returning) CEO Malcolm Walker. After Iceland merged with Booker in 2000, Walker was pushed out and a new management team led by Bill Grimsey took over. Sales declined, head office staff ballooned and they hired a lot of McKinsey types. Walker's point is that they'd overcomplicated what was basically quite a simple business.

As a buddy in construction used to say, concrete erasers are expensive.

Sure, but I'm sure China isn't keen for people to get the idea that there's a playbook for every ambitious Chinese founder on how to move Chinese-developed IP and talent outside Beijing’s control.

Almost certainly Chinese national-security and technology-transfer restrictions around strategic AI tech. China’s NDRC formally prohibited the acquisition and ordered the transaction unwound.

I suspect the issue there was Claude, especially Opus 5, which just hates the idea that it could ever be wrong about anything.

It feels like I might need 130 million years to watch the whole thing play out

“Eventually I realised restaurants are about diners, not knives.”

You can still create enormous value modelling things that are well understood, and even boring. Most LOB SaaS wasn't created from deep Google-level engineering, it was taken in modelling an ugly workflow in CRUD and creating customer records, job scheduling, worker assignments, quotations, recurring contracts, invoices, inspection checklists. Most Construction Tech startups for instance today is this.

Most democratic socialist societies are capitalist, capitalism just remains politically subordinate rather than able to govern society unchecked.

He had the good fortune to cut just as AI was beginning to mean you needed fewer software engineers for the same level of output. And yes, software companies across the industry had accumulated in many instances people whose business was not, and was sometimes orthogonal to, building software.

>> It opens a pull request >> Then keeps iterating as review comments land, in the same thread, with the same context.

My experience, particularly with Sol is that agents are generally really bad at knowing when to stop 'iterating' and will continue covering off ever-more obscure edge cases. How does Hoplite solve for this?

Stack Overflow conveniently defining the problem in a way that makes SO the answer. The irony is that the kind of code review that LLMs make possible would never have been doable on SO lest you be accused of trying to 'outsource the work'.

Now agentic coding produces the full 100 line change, and SO says the real crisis is that nobody has reviewed it properly. SO was fully responsible for creating the culture where no one wanted to use it.

Would it be uncharitable to say we are reinventing all of the awful single pane of glass / enterprise orchestration platforms of the 2010s but making them 'fun'?

Quietly eh?

To me, the "who manages the agents" shtick has the same grifty smell as “your developers are downloading from random Docker registries / random NPM packages” and "Shadow IT".

None of it is wrong exactly, but it feels like same enterprise-security machine finding the next anxiety surface than a "world is on fire right now" concern.

All of it always ends as a priesthood and a six-figure governance platform, rather than just taking practical steps to improve process.

Between 2002 and 2013, Hartlepool in North East England elected a local football mascot known as H'Angus the Monkey as their mayor, winning three elections. I think he did ditch the monkey costume between elections, maybe Binface could do the same?

> I just installed this. I am very confused. I no longer have a Codex app on my computer. ChatGPT is now Codex.

I ... hope we have an option for opting out of this? Otherwise I'm back on T3 Code.

They've discovered it's a good marketing strategy. Whenever there's an outage, or a new launch, there's often a reset with it, which helps keep people engaged with OAI / Tibo and reduces churn.

They've also introduced banked resets, which are really clever. If you have a $200/month plan and three banked resets, you're not churning because you will overweight giving up those resets (loss aversion theory).

Yes that's what I meant, the per token cost is higher but as you say the efficiency levels it out / works slightly in Codex's favour.