Hacker News Reader

ezston FreeCAD

I got to use SolidWorks, Catia, Inventor before having my hands on pre-1.0 versions of FreeCAD. I never really understood the argument that it's too complex. The UI may be what it is (and admittedly full of shortcomings), but I found FreeCAD to be very conventional in the sense that you build out of sketches, define constraints that are identical from every other tool, compose those through extrusions, revolutions, etc.

The fact that it crashed on me for everything and nothing all at once seemed a bigger problem than "complexity".

ezston Claws are now a new layer on top of LLM agents

Now we do computing like we play Sim City: sketching fuzzy plans and hoping those little creatures behave the way we thought they might. All the beauty and guarantees offered by a system obeying strict and predictable rules goes down the drain, because life's so boring, apparently.

ezston 15 years later, Microsoft morged my diagram

> What if code generation is copy-pasting GPL-licensed code in to your proprietary codebase?

This is obviously a big, unanswered, issue. It's pretty clear to me that we are collectively incentivised to pollute the well, and that it happens for long-enough for everything to become "compromised". That's essentially abandoning opensource and IP licensing at large, taking us to an unchartered era where intellectual works become the protected property of nobody.

I see chatbots having less an impact on our societies than the above, and interestingly it has little to do with technology.

ezston 15 years later, Microsoft morged my diagram

Waiting for the LLM evangelists to tell us that their box of weights of choice did that on purpose to create engagement as a sentient entity understanding the nature of tech marketing, or that OP should try again with quatuor 4.9-extended (that really ships AGI with the $5k monthly subscription addon) because it refactored their pet project last week into a compilable state, after only boiling 3 oceans.

ezston Running My Own XMPP Server

There's a room index at: https://search.jabber.network/

ezston Running My Own XMPP Server

For me too, ejabberd is the admin-friendlier/lower-effort one. Being more "monolithic", your calls will work straight out of the box because it ships a turn server properly configured out of the box, manage certificates over ACME for you, etc. Prosody isn't bad but has a reputation of needing attention to be paid to which incompatible modules not to enable together and overall more protocol knowledge. Both will run on a first gen RPi effortlessly.

ezston Running My Own XMPP Server

Re: Signal, it's even worse: they are openly opposed to federation and to letting alternative clients use their server. They demand control and obedience, which has always been suspicious-enough to defeat any goodwill effort on their side. Why would I have/want to trust them when XMPP is a viable federated alternative?

ezston Running My Own XMPP Server

Matrix has gotten to a complexity threshold that makes it near-impossible to have independent client/server implementations. Element is terrible, and many contenders are better in a way or another, but all lack some essential feature to turn them into practical alternatives.

ezston Running My Own XMPP Server

Same, years ago, I set up a Matrix server, because it was advertised as the-new-XMPP (and I had done XMPP as a user, a long time before, and thought it had been quirky enough to warrant a successor protocol).

What I found with Matrix was the same terrible experience you describe, so I gave old XMPP a new look, and it's been great and continuously improving since. I sleep much better at night having my whole family using XMPP over a self-hosted ejabberd than I can using Matrix to talk with them (and synapse... Forget using synapse federated).

ezston Editor's Note: Retraction of article containing fabricated quotations

So, the solution to too much AI is... Even more AI! You sound like you would fit just right at a LLM-shop marketing department.

ezston Matrix messaging gaining ground in government IT

I was like "oh common, that can't be a real comment, it's obvious to everyone how unstable this still is", then I saw that the comment was from Arathorn.

You know, for half of the time you spend commenting over here to save face (or something), you could work with your users and see their firsthand experience for yourself.

ezston GitHub is down again

Or maybe the mindless rush to host it in azure?

ezston uLauncher

From your comment I wanted to give Niagara a try, most useful features are apparently behind a pay wall, and that was completely fine... Until google asked me for my credit card details for the 7-days trial. This is probably none of Niagara's fault, but enough of a discouragement for me to give it a real chance.

ezston European Commission Trials Matrix to Replace Teams

And XMPP. Which probably will remain a better Matrix than Matrix ever will be, the venture capital put aside, that is.

ezston European Commission Trials Matrix to Replace Teams

It's so slow because it's so badly designed as a protocol, E2E isn't really the problem (the slowness is roughly equivalent for non-encrypted rooms)

ezston Data centers in space makes no sense

> colonize mars

Oh, that crap again.

ezston France dumps Zoom and Teams as Europe seeks digital autonomy from the US

Yeah, and once the precedent is settled, you can bet that the private sector will follow, and give birth to a bunch of local service companies to deploy and support those solutions in an healthier and fairer manner than the current GSuite/MSOffice duopoly.

ezston Norway EV Push Nears 100 Percent: What's Next?

So we are playing semantics. I don't think "Kicking-out" is universally understood as "getting rid of the entirety of something", you are encouraged to prove me wrong, I certainly don't care enough to die on this hill.

ezston Norway EV Push Nears 100 Percent: What's Next?

The above-linked protests led to re-drawing urban development plans for most major Dutch cities with the effect of reducing the prevalence (and necessity) of cars in city centers. Whole housing blocks were getting destroyed as to build highways, canals were dried so cars could run in their place. All that was undone, and more, and nowadays Dutch cities are a shining example of how liveable urban centers are with fewer/close to no cars. Those are well documented facts, I don't know what you are even arguing about.

ezston Microsoft 365 now tracks you in real time?

One more reason it sucks to be American? I know of several counties where that violates more than a couple labour and constitutional laws.

ezston Norway EV Push Nears 100 Percent: What's Next?

https://www.smartcitiesdive.com/ex/sustainablecitiescollecti...

https://bicycledutch.wordpress.com/2025/03/05/a-rare-glimpse...

https://old.reddit.com/r/videos/comments/ittddm/how_a_busy_r...

https://www.iamexpat.nl/expat-info/dutch-news/history-amster...

ezston Norway EV Push Nears 100 Percent: What's Next?

The Netherlands is peculiar in that cyclists reclaimed their rights to the cities by kicking cars out of them.

ezston Project Genie: Experimenting with infinite, interactive worlds

> intelligence. Human or LLM doesn't matter.

Being enthusiastic about a technology isn't incompatible with objective scrutiny. Throwing-up an ill-defined "intelligence" in the air certainly doesn't help with that.

Where I stand is where measured and fact-driven (aka. scientists) people do, operating with the knowledge (derived from practical evidence¹) that LLMs have no inherent ability to reason, while making a convincing illusion of it as long as the training data contains the answer.

> Sorry, but I just get the picture that you have no clue of what you're talking about- though most probably you're just in denial.

This isn't a rebuttal. So, what is it? An insult? Surely that won't help make your case stronger.

You call me clueless, but at least I don't have to live with the same cognitive dissonances as you, just to cite a few:

- "LLMs are intelligent, but when given a trivially impossible task, they happily make stuff up instead of using their `intelligence` to tell you it's impossible"

- "LLMs are intelligent because they can solve complex highly-specific tasks from their training data alone, but when provided with the algorithm extending their reach to generic answers, they are incapable of using their `intelligence` and the supplemented knowledge to generate new answers"

¹: https://arstechnica.com/ai/2025/06/new-apple-study-challenge...

ezston Project Genie: Experimenting with infinite, interactive worlds

> You really need to be obstinate in your convictions if you can dismiss LLMs at the time when everyone's job is being turned around by them.

I'm factual. You are the one with the extraordinary claim that LLMs will find new substantial markets/go through transformative breakthrough.

> Everywhere I look, everyone I talk to, is using LLMs

And everywhere I look, I don't. It might be the case that you stand right in the middle of an LLMs niche. Never did I say that one doesn't exist or that LLMs are inadequate at parroting existing code.

> Non-developers I know use them […]

among those are:

- things that have nothing to do with LLMs/AI

- things that you should NOT use LLMs for the reason that they will give you confidently wrong and/or random answers (because it's not in their training data/cut-off window, it's non-public information, they don't have the computing abilities to produce meaningful results)

- things that are low-value/low-stakes for which people won't be willing to pay for when asked to

> The list of uses is endless

no, it is not

> And I'm only talking of work-related usage

and we will get to see rather sooner than later how much business actually value LLMs when the real costs will be finally passed on to them.

ezston The Sovereign Tech Fund invests in Scala

Happy to throw some of mine for that. Scala may not be as popular as some other languages, but it's relevant for use cases as diverse as those for which you would otherwise use Java/C#/Go, JS/TS, C/C++/Rust (via scala-jvm, scala-js and scala-native), and it's a very nice language to work with at that.

ezston Project Genie: Experimenting with infinite, interactive worlds

Since a hot take is as good as the next one: LLMs are by the day more and more clearly understood as a "local maximum" with flawed capabilities, limited efficiency, a $trillion + a large chunk of the USA's GDP wasted, nobody even turning a profit from that nor able to build something that can't be reproduced for free within 6 months.

When the right move (strategically, economically) is to not compete, the head of the AI division acknowledging the above and deciding to focus on the next breakthrough seems absolutely reasonable.

ezston Android’s desktop interface leaks

Who in their right mind would put Windows on such a device, though? This is THE hardware generation that Microsoft painted a target on with their planned obsolescence strategy. I don't know if there can be a clearer signalling that it is to be avoided at all costs.

ezston I made my own Git

In some ways, the legacy of mercurial lives through jujutsu/jj and offers some sanity and familiarity on top of git's UI. But with that said, mercurial is far from dead, major "under-the-hood" works are going strong (including a rewrite in rust), the hosting situation is getting good with heptapod (a branch of gitlab with native mercurial support).

I really don't see any downside to recommending mercurial in 2026. Git isn't just inferior as a VCS in the subjective sense of "oh… I don't like this or that inconsistent aspect of its UI", but in very practical and meaningful ways (on technical merit) that are increasingly forgotten about the more it solidifies as a monopoly:

- still no support for branches (in the traditional sense, as a commit-level marker, to delineate series of related commits) means that a branchy-DAG is border-line useless, and tools like bisect can't use the info to take you at the series boundaries

- still no support for phasing (to mark which commits have been exchanged or are local-only and safe to edit)

- still no support for evolve (to record history rewrites in a side-storage, making concurrent/distributed history rewrites safe and mostly automatic)

ezston Parametric CAD in Rust

Not sure either but sharing the sentiment.

ezston Prism

Sure, but now we can't even assume that such research is submitted in good faith anymore. There just seems to be no perfect solution.

Maybe something like a "hierarchy/DAG? of trusted-peers", where groups like universities certify the relevance and correctness of papers by attaching their name and a global reputation score to it. When it's found that the paper is "undesirable" and doesn't pass a subsequent review, their reputation score deteriorates (with the penalty propagating along the whole review chain), in such a way that:

- the overall review model is distributed, hence scalable (everybody may play the certification game and build a reputation score while doing so) - trusted/established institutions have an incentive to keep their global reputation score high and either put a very high level of scrutiny to the review, or delegate to very reputable peers - "bad actors" are immediately punished and universally recognized as such - "bad groups" (such as departments consistently spamming with low quality research) become clearly identified as such within the greater organisation (the university), which can encourage a mindset of quality above quantity - "good actors within a bad group" are not penalised either because they could circumvent their "bad group" on the global review market by having reputable institutions (or intermediaries) certify their good work

There are loopholes to consider, like a black market of reputation trading (I'll pay you generously to sacrifice a bit of your reputation to get this bad science published), but even that cannot pay off long-term in an open system where all transactions are visible.

Incidentally, I think this may be a rare case where a blockchain makes some sense?