sandbags

I find it rather disingenuous of them to claim these things they train into their models arising in their models.

I don’t disagree with your reasoning but I would have thought this pipe would be in an appropriately named function (at least that’s how I’d use it in Elixir) to help understand the result.

Thanks for the info.

Interesting. Can I use this with local LLM's? I use LM Studio which exposes an OpenAI compatible API endpoint. Btw your contact link on your website doesnt seem to go anywhere for me.

Years ago I used Backblaze and it twice "lost" my entire backup with their support staff shrugging and suggesting I start it again. At that point I didn't have fibre so a full backup was weeks if not months. Needly to say I opted out of their service. I am not overly surprised that this is not a solved problem.

Another publisher who hates their content so much they make it impossible to read with multiple flying adverts and so on. I was interested but gave up fighting to read it.

I’ve been wanting to write this somewhere and this seems as good a place as any to start.

Is it just me or is MCP a really bad idea?

We seem to have spent the last 10 years trying to make computing more secure and now people are using node & npx - tools with a less than flawless safety story - to install tools and make them available to a black box LLM that they trust to be non-harmful. On what basis, even about accidental harm I am not sure.

I am not sure if horrified is the right word.

PathFinder is the only one I can think of. I find it overly complicated when all I really need is Finder - the endless jank.

Possibly QuickSilver?

Anyone with an opinion can be labelled biased. Also I’m not clear what you mean by Marcus “riding the anti AI wave” but infer that you mean it negatively. He has been writing informed criticism for several years and about cognitive psychology for considerably longer.

How did you arrive at this model?

Right. Their fork is just a fork though, so far I can see it has no commits of its own. So, same question, what payoff do they get forking that fork?

17 years ago I wrote this plugin for Rails 2 to version database tables.

At the time it scratched an itch for a simpler alternative to acts_as_versioned. It was even used in Github for a little while IIRC.

17 years later what I can't understand is why a steady stream of people are still forking it. It's a rare week that goes by that I don't get a notification. 2 in the last day or so.

What's the game here? I can't think of any legitimate reason to fork a 17 year old plugin for a truly ancient version of Rails. So… what's the payoff?

A question that was raised, but not answered in the post comments. It's easy to turn off "Apple Intelligence" but does turning if off also turn off Siri? I.e. is Siri now "Apple Intelligence"?

"We are building machines that can think and reason."

I'm looking for the evidence of "thinking" or "reasoning".

Brave Hungarian peasant girl that she was.

mud + rainbow = pig ? I’m not sure I am seeing the point.

I couldn’t answer this from the site. Will this let me run macOS Catalina on an M2 Mac Studio with usable graphics performance? Because that would give me back a bunch of 32-bit games I didn’t want to give up.

I would be surprised if GPT-3 uses 700GB of RAM. It may be true, I don't know. But I am running 70B parameter models (quantized to 5 or 6 bits, biggest is 48GB loaded) on my 64GB Mac M2 Max Studio now and they are usable and the machine is still usable too. With an M2 Ultra and 192GB of RAM I imagine you could do a lot more.

I'm not arguing that these models hold up against GPT 3.5 and I still use GPT 4 when it matters. But they work and it's more like the difference between Premier League & Division 1, rather than PL & a five-a-side team from Bracknell.

Even a few years ago I could not have imagined this.

Given the pace of work on optimisation and my assumption that the M3 Studio I buy next will probably have 256GB of RAM at much the same power levels as I use now, it seems eminently possible it's a year or two away.

As a developer who came to Elixir from Clojure I had the same reaction to fun.(), “this sucks”. Then someone explained why. So, okay, there is a good reason but “bleh.” I think that lasted maybe a day, by which time I had accepted it as part of the language. It’s just not a big deal when the language offers so much in return. And I liked Clojure.

Unless you are arguing that HN is arbitrarily prioritising Elixir related content then it’s getting exactly the amount of attention it deserves.

In the Ruby language you are implicitly in the context of the Main object (IIRC). If scripting was it's own context then something like this could perhaps work for Elixir. You're already in a module context implicitly.

Same. As Douglas Adam’s said “I love deadlines. I love thé whooshing sound they make as they go by” but if a deadline involves other people those consequences get me to care.

Bought it. It’s very good. Indexed my 15,000 photos in a few mins (iPhone 14). Search seems fast, a few seconds only.

Am I missing a way to jump from the photo preview to the photo in the Photos app?

This. Why does my browser not keep history forever, I surely have space for it? Why doesn’t it archive every page it can? And build a full-text index of every page so that I can be reasonably sure I can lay my hands on something I’ve seen, even if the original is now gone. I can only conclude that the people working on browsers aren’t interested in what goes on in a browser.

“Growth without purpose is the ideology of the cancer cell” is one of my favourite quotes by the exceptionally quotable Edward Abbey.

And I tend to agree with you that we would be better off if companies could define “enough” and maybe disband at that point rather than stagnating.

The thing lurking in my mind that maybe makes this a pipe dream is the Red Queen hypothesis.

I’ve worked in dozens of startups in my career that all failed, including 2 of my own. I started another and I turned 50 this year. I’m older, slower, and have less stamina. I’m also more crafty, more self-aware, more experienced. I’m biased but I see many benefits. If you want to keep starting companies it’s not age that will hold you back. M

I am listening to Eric Schlosser’s book “Command and Control” at the moment and it has quite a bit of detail on this and other nearly disastrous incidents and the complete lack of serious command & control around nuclear material/weapons. We are very lucky not to have had several accidental or unwanted nuclear detonations to this point. I’ve not finished the book but I am not optimistic the situation is very much better today.

I programmed in Clojure for many years and now a couple in Elixir (which, though not a Lisp does have excellent macro support).

I think I wrote three macros in my time with Clojure and none with Elixir. That’s because mostly I don’t need to, I can solve it with functions.

However my work in both languages rests on a lot of macros included in the stdlib and other libraries. Those macros make my life much easier and, when you really need macro power yourself it is right there.

Writing good macros can be challenging and supporting them after even more so. So it’s a power I use sparingly. But I depend upon macros all the time.

This seems like it could be interesting if it could generate migrations for specific database frameworks as model diffs.