Ah, well I certainly agree with the general notion you’ve stipulated. My question for data was regarding the quality of the CVEs, not the concept.

That’s a nice hot take. Got any data to back that up?

You know what, I’ve been a Firefox user since the days of beta Phoenix. I’ve expressed your view for years now - Firefox was great, but held back by a commercial org that wanted fancy SF offices, fancy SF compensation etc.

But I tried out Helium a little while ago and in truth, I’m not sure Firefox is all that good these days. Helium is basically ungoogled-chromium + extension support + super lean UI.

The upside is that you no longer have to contend with sites that have given up on Firefox testing. The downside is manifest v3. But all in all, I find it a better browser. Firefox is so busy telling its story, and finding revenue, that it just seems to have left behind product market fit.

Tell me more about your memory. I’m on automem and it’s ok, but slow.

But your analogy also breaks down because once an AI has written the code it’s very much concrete, right? The AI may choose to modify that concreteness into something differently concrete, but it’s not like you are in any doubt about what the AI comes out with (if you bother to check it, which you should).

I use to write asset systems during the AAA gaming part of my career and the elusive “[my] perfect file format” always eluded me.

Over the years I spent many a showers and long drives thinking about what the “perfect file format” needs and what kind of architecture would support this. I started formulating these into long text files and over time things crystallised into “yes this would work”.

I never had the time to write the code fully. A couple of false starts, coupled with “I should do this in rust even though I’ve been in C++ my whole life” meant it lingered and lingered.

The architecture felt right though.

Enter AI; feeding the spec text files to an AI made me resolve a couple of “ah, yes, but in practice this bit of information isn’t available yet” gotchas but the code worked. I worked in architectural layers, carefully checking input, internals and output of each before proceeding to higher layers. And tests and tests, that’s you would never have bothered to write yourself, kept each layer verifiable, even out in the far corners of reasonable.

It worked and it was real affirmation to see 20 years of pontification made real.

But then I started profiling it. And hot damn, an LLM knows nothing about cache, file systems nor does it think about ways memory allocators and virtual memory systems work.

There was 200-300% speed ups to gain from hand tuning it and iterating with a proper profiler.

And all of this happened at code level, with real concrete code.

So yeah, an AI can be a wonderful tool. And the code it produces and iterates on is concrete. It’s not like the AI starts from scratch every time you have a thought. So the very, very first pass is not deterministic. But the thing you deliver forwards very much is.

Of this I agree. It may very well be the end of us.

Having said that, LLMs produce far better code and architecture than most developers.

Flame on!

… then you win

I was deeeep in the OneNote eco-system. I forced myself out of it and spent a lot of time finding the best alternative so I can save you some time: The answer is Obsidian.

But to get to even 60% of what OneNote does, you need a fair amount of plugins, which lowers stability somewhat and, having tried the three biggest sync systems, you might as well go straight to paying for their sync. The self-hosted sync options do not work reliably.

Even better: Your advancement is critical to ensure the American intelligence industry doesn’t fall behind.

I mean it’s OK - you’re not copying it, you’re using it to train an intelligence.

People who think 60% is too high haven’t been to any mid-sized business in a while, I think.

The high of cracking an intractable problem (pre-AI) was just unreal. I’d struggle to sleep, awaiting the morning so I could get back at it.

It’s the challenge of it though.

For root drives (where recovery may be required) and database drives it still serves a purpose, IMHO.

Wait, are we talking Linux ISOs or “Linux ISOs”?