Hacker Newsnew | past | comments | ask | show | jobs | submit | filereaper's commentslogin

As a Canadian, we should work on improving our productivity and incorporate more automation and tooling. Whatever impacts our GDP in times of tariffs and economic uncertainty.

Not collectively waste time on the useless debates on how to spell things.


Llamafile did great work on improving cpu inferencing and generosly upstreamed it into llama.cpp

Wonderful project, please check it out.


There's almost an underlying assumption that taxpayers will bail you out, or you that you won't go to jail for all these securities frauds, or miscalculating risk on pension funds assets is okay.

Rip away all these away and the Emperor is called out to be naked very quickly.


We've just started to roll out our MCP Servers and if Anthropic and the community has already moved on, we'll wait till all this churn subsides till switching over next time.


I don’t really see how this replaces MCP tbh.

MCP gives the LLM access you your APIs. These skills are just text files with context about how to perform specific tasks.


You don't need MCP if you can instead drop in a skill markdown file that says "to access the GitHub API, use curl against api.github.com and send the GITHUB_API_KEY environment variable in the authorization header. Here are some examples. Consult github-api.md for more."


I would much, much rather provide discreet APIs directly to the LLM via MCP than just tell it to hit the api and figure it out from the docs.


> You don't need MCP

Depends on who the user is...

A difference/advantage of MCP is that it can be completely server-side. Which means that an average person can "install" MCP tools into their desktop or Web app by pointing it to a remote MCP server. This person doesn't want to install and manage skills files locally. And they definitely don't want to run python scripts locally or run a sandbox vm.


Am I the only person left that is still impressed that we have a natural language understanding system so good that its own tooling and additions are natural language?


I still can't believe we can tell a computer to "use playwright Python to test this new feature page" and it will figure it out successfully most of the time!


Impressing, but I can't believe we went from fixing bugs to coffee-grounds-divination-prompt-guessing-and-tweaking when things don't actually go well /s


That's going to be a lot less efficient context-wise and computing-wise than using either a purpose-built MCP or skill based around executing a script.


Strong agree here


Is any of it even churn? I feel like almost everything is still relevant, basically everything was a separate card which they're using to build up a house. Even RAG still has it's place

Now wherever they're able to convert that house of cards into a solid foundation or it eventually spectacularly falls over will have to be seen over the next decade.


I would also like to recommend Aerospace which is closer to i3 on Linux:

https://github.com/nikitabobko/AeroSpace


Aerospace is great, but macos is not, and unfortunately, there are enough immutable things (given the discussions in the GH issues around them in the Aerospace repo) about macos' walled garden/closed approach that annoyed me/got in my way enough after a few weeks of using it that I have concluded (for myself/my purposes at least) that macos is not tiling/custom WM workable now or in the forseeable future.

How Aerospace (and others) degrade under heavier system load, etc (again, not b/c they are poorly written, but b/c macos inherently does not want these kinds of packages to even exist and makes things very hard to say the least, if not impossible, requiring some very awkward gymnastics) made this incredibly frustrating at the worst possible times.


Adds support for Wayland with its latest release.

It also supports the downstream forks of lan-mouse and input-leap.


Not so sure about the 9-1-1 point.

>"All wireless phones, even those that are not subscribed to or supported by a specific carrier, can call 911. However, calls to 911 on phones without active service do not deliver the caller’s location to the 911 call center, and the call center cannot call these phones back to find out the caller’s location or the nature of the emergency. If disconnected, the 911 center has no way to call back the caller."

Source: https://www.911.gov/calling-911/frequently-asked-questions/

Seems they're deliberately distorting their responsibilities.


That sounds like something to fix, then.

Requiring carriers to enable whatever side-channel is necessary to transmit location info to the 911 PSAP shouldn't be a heavy lift.

The callback issue seems harder to resolve, but even if a handset has no assigned phone number, there are other ways to address it (e.g. IMEI). Carriers should be required to build in a capability to make this work.


> Requiring carriers to enable whatever side-channel is necessary to transmit location info to the 911 PSAP shouldn't be a heavy lift.

Carriers can and they should.

> The callback issue seems harder to resolve, but even if a handset has no assigned phone number, there are other ways to address it (e.g. IMEI). Carriers should be required to build in a capability to make this work.

Carriers can't, because the 3GPP standards do not make this possible.


Total aside.

Has anyone found an alternative to Synergy? Its the last thing that's anchoring me to X11.

The Synergy team has been saying for years they'll support Wayland but no release has been made, I'm a paid customer.


IIRC by design applications can't get the absolute mouse pointer position in wayland, so something like Synergy probably is not easy to do in Wayland without 'hacks'.


Isn't that something libei is designed to solve?


Have never heard of it, but seems like it should solve that


Do any Synergy forks like Barrier do it?


Barrier was abandoned and no contributions are made IIRC.


The closest I’ve found in terms of description is https://github.com/feschber/lan-mouse, but the lack of encryption on the connection has discouraged me from using it.

I’m also a paid Synergy user, and it’s frankly comedic how long Wayland support has been on their road map. I’m not convinced it’s ever coming, which means I’m probably only 1-2 distro updates from being forced to use something else


I'm in the same boat.


Synergy core (aka synergy 1) just had Wayland support merged this week. “Next year” has finally come to pass


Genuinely trying to understand.

Didn't the SR71 fly between LA and NY at above supersonic speeds? Don't military jets fly at supersonic speeds above land?

What makes this allowable in terms of sonic booms? Is it that they're much smaller and the their booms don't cause as much of a "wake" on the ground?


> Didn't the SR71 fly between LA and NY at above supersonic speeds? Don't military jets fly at supersonic speeds above land?

I think not routinely in either case? Most supersonic aircraft can't actually sustain supersonic speeds for a significant period; the Concorde and Tu-144 are, as far as I know, the only aircraft ever designed with supersonic flight as the effective _default_ mode of operation. For both planes, _getting_ to supersonic was rather expensive, and required the use of afterburners, so an ideal flight profile would involve ramping up to supersonic once, and then staying that way for the rest of the flight.

The SR71 was, unusually, _capable_ of sustained long-range supersonic flight, but I can't imagine it was routinely used above land when not actually on missions; why would it be?


> The SR71 was, unusually, _capable_ of sustained long-range supersonic flight, but I can't imagine it was routinely used above land when not actually on missions; why would it be?

Everything a military plane does is a mission, and the SR-71 absolutely did supersonic training missions over the US.

And as I understand, the military can and still does do supersonic training missions over land, but they do it at higher altitudes and in locations where it isn't disruptive to population centers.


The F-22 can also cruise at supersonic speeds without afterburners. https://en.wikipedia.org/wiki/Supercruise.


_Can_, but does it routinely outside of training/deployment situations? Like if you're just moving one from one side of the US to the other, you're probably going subsonic.


a couple things:

1. smaller lighter, and higher flying reduce the sonic boom size

2. way fewer planes

3. the military gets some latitude to ignore the rules that everyone else has to follow


> Don't military jets fly at supersonic speeds above land?

It can happen, but it is disruptive at best. It isn't something you'd want happening around a populated area on a routine basis in peacetime. And it typically is not, unless under emergency circumstances.

https://www.youtube.com/watch?v=G8G-mgKycUM


Thd only time I've ever heard a sonic boom was as a kid, on holiday in the south of France. They would regularly fly Mirage jets over the beach resorts. I suspect these flights were for when Gen. de Gaulle was in residence at the Chateau d'If.


smaller, and SR71 could cruise at 85000 feet


The European Politician would have to introspect and answer exactly how it is that they add value.

All they can do is enact regulations and decrees in a centralized way like the former USSR.

Also as a entrepreneur in Europe, what incentives are there to create anything. There's the existing crippling regulations, employment laws are tough to try things out and just go bust, and at the end of it all if you're successful, the taxes remove whatever gains you should rightfully get for jumping through all these hoops.

All you have now are $200k designer bag manufacturers because of legacy.

I hope to god they do whatever is needed to keep ASML and the like competitive.


So what are you proposing? Joining the race to the bottom to the Chinese surveillance state? Sell out "OCP"-style to big tech like the US?


I won't propose a solution since I'm not in charge, but my guess for what will happen regardless is the EU outsources its tech innovation to some other place where they can still reap the rewards. Kinda like how they've imported energy from countries with lower environmental standards.


> but my guess for what will happen regardless is the EU outsources its tech innovation to some other place where they can still reap the rewards.

The AI laws don't work that way. Just like Google can't say: "We'll just put all of our servers in the US so we don't have to comply with the GDPR and start serving the EU market".

> Kinda like how they've imported energy from countries with lower environmental standards.

That doesn't happen. Importing energy from another country doesn't make any difference for a countries emission rights and second of all the energy grid of the EU is connected only in a very limited way to countries outside the EU (like Turkey).


> AI laws don't work that way

The companies being in America and selling demanded services in the EU will force the laws into one configuration, whether through lobbying by customers or direct pressure from the U.S.

I’m not seeing, for example, any effort to comply with EU regulations among any of the smaller AI startups, most of whom have customers in the EU. (Which is rational. Once they get bigger they can weigh fines against exiting the market. And if it becomes widespread, it can escalate to a political priority for State.)


Other way. An EU-based or EU-related business can serve non-EU customers differently.

I'm referring to natural gas and other energy, not the electrical grid.


(or it can serve EU customers but be offshore enough that it's not subject to EU countries' labor laws or taxes)


> Joining the race to the bottom to the Chinese surveillance state?

Look up chat control and Digital Services Act.


Could you name some of those crippling regulations you're talking about? Just so Im better informed as a european


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: