mamp
mamp on Hacker News
So you want to close down Chinatown?
I was at WWDC in 1996. Each presentation showing off a new feature of Copeland required booting a custom build of the binary, and even then it was flakey as hell because it didn't implement proper memory protection. I vividly remember getting a sudden realisation that Copeland was never going to ship.
The 18 Pro enables 'pro' features in the default camera app that finally allows you to set the white balance, shutter speed, aperture, but not the ISO.
I've never heard of anyone calling SOM 'better' at anything this century. I came across it when it was the foundation for OpenDoc at Apple, your comment brought back many bad memories
One doesn’t simply create a Warren Abstract Machine by accident
Strange article. The problem isn’t that everyone doesn’t know how everything works, it’s that AI coding could mean there is no one who knows how a system works.
Sadly, the US is more likely to at war with Europe than China
I haven't been writing Rust for that long (about 2 years) but every time I see .unwrap() I read it as 'panic in production'. Clippy needs to have harder checks on unwrap.
To be fair the performance of rules or Bayesian networks or statistical models wasn't the problem (performance compared to existing practice). DeDombal showed in 1972 that a simple Bayes model was better than most ED physicians in triaging abdominal pain.
The main barrier to scaling was workflow integration due to lack of electronic data, and if it was available, interoperability (as it is today). The other barriers were problems with maintenance and performance monitoring, which are still issues today in healthcare and other industries.
I do agree the 5th Generation project never made sense, but as you point out they had developed hardware to accelerate Prolog and wanted to show it off and overused the tech. Hmmm, sounds familiar...
Don’t attribute to jealousy that can be adequately explained by vanishing gradients.
BTW the ad hoc treatment of uncertainty in Mycin (certainty factors) motivated the work of Bayesian network.
I've been using Gemini 2.5 and Claude 3.7 for Rust development and I have been very impressed with Claude, which wasn't the case for some architectural discussions where Gemini impressed with it's structure and scope. OpenAI 4.5 and o1 have been disappointing in both contexts.
Gemini doesn't seem to be as keen to agree with me so I find it makes small improvements where Claude and OpenAI will go along with initial suggestions until specifically asked to make improvements.
When do you turn it off? I have a Mac M1 Studio and I just let it sleep. If things get weird I reboot. I think I recall using the power button about a year ago after returning from vacation after I had shut it down.
But unfortunately it’s not premature. It’s been a problem for so long!
I suspect it's not so much the "expert" instruction but the list of subjects of the expertise. These words will generate embeddings that have a better chance of activating useful pathways and relationships within the LLM's generation path.
My understanding is the goal is to prime the LLM with context that it will use when generating answers, rather than hoping it will infer connections in the feed forward layers when generating answers based on a sparse prompt.
Apple’s first in-house designed chip was the A4 in 2010.
I agree with your assessment.
The real challenge is how can organisations best leverage super stars to achieve great things. It's difficult because managers have to be smart and confident, and the super star can't be allowed to undermine others.
The high prevalence of insecure managers and arrogant super stars makes this hard to do.
With ChatGPT some of the limitations of the tech are handled by the user e.g. starting a new chat when you want to discuss a new topic. An assistant has to detect changes in user context somehow. Also, I think it would be harder to know what to inject in the prompt since conversations are more like context based RAG rather than topic (embedding) based.
Then you have all the usual generative issues: hallucinations, alignment, sticking within guardrails, no repeatable testing, drift. The potential for errors at that scale is pretty staggering.
ChatGPT has helped me survive learning Rust.
Yes, a real “Flowers for Algernon” vibe
It turns out to accurately predict the next word requires huge amounts of implicit contextual knowledge i.e. understanding about the world.
Next word prediction was the trigger, the optimisation for the task results in a broad (but currently unreliable) model of the world.
Elixir has a vector library Nx that’s designed for these use cases.
I read it as even if extra type information was there the BEAM doesn’t use type information to optimise across modules.
Actually I think de Dombal’s work showed better than expert performance in 1974 (one of the references in the book):
de Dombal, E T., Leaper, D. J., Horrocks, J. C., Staniland, J. R., and McCann, A. E 1974. Human and computer aided diagnosis of abdominal pain: Further report with emphasis on the performance of clinicians. British Medical Journal 1: 376-380.
Interestingly Mycin was used to demonstrate the limitations of rule based inference and with trying to handle uncertainty in Rule based systems by two of Shortliffe’s students in the fantastic paper “The myth of modularity” [1] which argued for probabilistic approaches to reasoning.
1. https://arxiv.org/abs/1304.3090
I think their hoping tab groups will reduce the 35 tab situation, but it’s too hard to organise when in information gathering mode. I hope they put a preference option to go back to the current interface. I’ll be filing a bug report.
Very cool project but definitely not best practice. You could reduce the number of State variables using one or more shared state objects and access via environment at certain points in the view hierarchy.
Generally it’s better to compose views, by that I mean you often have to because of the compiler. I’m surprised this file even compiles! The Swift compiler usually chokes on much smaller views.