Yes, sorry! We're investigating, but my current theory is we got overloaded because I relaxed some of our anti-crawler protections a few days ago.
(The reason I did that is that the anti-crawler protections also unfortunately hit some legit users, and we don't want to block legit users. However, it seems that I turned the knobs down too far.)
In this case, though, we had a secondary failure: PagerDuty woke me up at 5:24am, I checked HN and it seemed fine, so I told PagerDuty the problem was resolved. But the problem wasn't resolved - at that point I was just sleeping through it.
I'll add more as we find out more, but it probably won't be till later this afternoon PST.
Like many others, the ability to run uBO is the main reason I use Firefox. Otherwise I'd use Chrome or Safari.
The interpretation is not the problem. Whether he will do it, is actually secondary to the fact that he thinks cutting adblock can bringing in money.
No, it will just kill the browser. The fact he thinks otherwise tells me how out of touch he is.
>> He says he could begin to block ad blockers in Firefox and estimates that’d bring in another $150 million, but he doesn’t want to do that. It feels off-mission.
> It may be just me, but I read this as “I don't want to but I'll kill AdBlockers in Firefox for buckerinos ”.
Yes, that does seem like a pretty uncharitable interpretation of that quote. I read it as "we won't do it, even though it would bring in $150M USD".
It's kinda frustrating that Mozilla's CEO thinks that axing ad-blockers would be financially beneficial for them. Quite the opposite is true (I believe) since a ton of users would leave Firefox for alternatives.
I don't think formal verification really addresses most day-to-day programming problems:
* A user interface is confusing, or the English around it is unclear
* An API you rely on changes, is deprecated, etc.
* Users use something in unexpected ways
* Updates forced by vendors or open source projects cause things to break
* The customer isn't clear what they want
* Complex behavior between interconnected systems, out of the purview of the formal language (OS + database + network + developer + VM + browser + user + web server)
> Large language models are something else entirely*. They are black boxes. You cannot audit them. You cannot truly understand what they do with your data. You cannot verify their behaviour. And Mozilla wants to put them at the heart of the browser and that doesn't sit well.
Am I being overly critical here or is this kind of a silly position to have right after talking about how neural machine translation is okay? Many of Firefox's LLM features like summarization afaik are powered by local models (hell even Chrome has local model options). It's weird to say neural translation is not a black box but LLMs are somehow black boxes that we cannot hope to understand what they do with the data, especially when viewed a bit fuzzily LLMs are scaled up versions of an architecture that was originally used for neural translation. Neural translation also has unverifiable behavior in the same sense.
I could interpret some of the data talk as talking about non local models but this very much seems like a more general criticism of LLMs as a whole when talking about Firefox features. Moreover, some of the critiques like verifiability of outputs and unlimited scope still don't make sense in this context. Browser LLM features except for explicitly AI browsers like Comet have so far had some scoping to their behavior, either in very narrow scopes like translation or summarization. The broadest scope I can think of is the side panels that show up which allow you to ask about a web page with context. Even then, I do not see what is inherently problematic about such scoping since the output behavior is confined to the side panel.
I really hope astral can monetize without a highly destructive rugpull, because they are building great tools and solving real problems.
It is us, developers, who convinced our management to purchase GitHub Enterprise to be our forge. We didn't pay any heed to the values of software freedom. A closed source, proprietary software had good features. We saw that and convinced our management to purchase it. Never mind what cost it would impose in the future when the good software gets bad owners. Never mind that there were alternatives that were inferior but were community-developed, community-maintained and libre.
The writing is in the wall. First it was UX annoyances. Then it was GitHub Actions woes. Now it is paying money for running their software on your own hardware. It's only going to go downhill. Is it a good time now to learn from our mistakes and convince our teams and management to use community-maintained, libre alternatives? They may be inferior. They may lack features. But they're not going to pull user hostile tricks like this on you and me. And hey, if they are lacking features, maybe we should convince our management to let us contribute time to the community to add those features? It's a much better investment than sinking money into a software that will only grow more and more user hostile, isn't it?
Okay results are in for GenAI Showdown with the new gpt-image 1.5 model for the editing portions of the site!
https://genai-showdown.specr.net/image-editing
Conclusions
- OpenAI has always had some of the strongest prompt understanding alongside the weakest image fidelity. This update goes some way towards addressing this weakness.
- It's leagues better at making localized edits without altering the entire image's aesthetic than gpt-image-1, doubling the previous score from 4/12 to 8/12 and the only model that legitimately passed the Giraffe prompt.
- It's one of the most steerable models with a 90% compliance rate
Updates to GenAI Showdown
- Added outtakes sections to each model's detailed report in the Text-to-Image category, showcasing notable failures and unexpected behaviors.
- New models have been added including REVE and Flux.2 Dev (a new locally hostable model).
- Finally got around to implementing a weighted scoring mechanism which considers pass/fail, quality, and compliance for a more holistic model evaluation (click pass/fail icon to toggle between scoring methods).
If you just want to compare gpt-image-1, gpt-image-1.5, and NB Pro at the same time:
https://genai-showdown.specr.net/image-editing?models=o4,nbp...
I can't help but feel that this article was written in a format that is the textual equivalent of thin desires…
Every sentence is separated into its own paragraph, like each one is supposed to be revelatory (or maybe tweet-worthy). It's pretty common design knowledge that if you try to emphasize everything, you end up emphasizing nothing. The result is that reading the article feels choppy, and weirdly unsatisfying, since the larger arc of each point is constantly being interrupted.
Why choose such an antithetical form, to what is otherwise an important and deep message?
The only answer that comes to mind is that the author's livelihood, or at least their internal gauge of success, is tied to manipulating readers' thin desires.
The article is missing this motivation paragraph, taken from the blog index:
> Graphics APIs and shader languages have significantly increased in complexity over the past decade. It’s time to start discussing how to strip down the abstractions to simplify development, improve performance, and prepare for future GPU workloads.
I got contacted by our rep a couple weeks ago, who informed me of this news. I thought it was a disaster and it really pissed me off. The rep couldn't even explain the reasoning well. It basically summed up to "because we can" and "where are you going to go?". He was shocked to find out that I didn't like it.
We currently self-host on kubernets/aws. The thing that really got to me isn't the new charge per se. It's the fact that GHA has a ton of problems. I can hold my nose and deal with them when it's free. But now that you're squeezing me, at least you could have created something like GHA 2.0 and added a charge for that. Instead, there are vague roadmap promises which don't even include things that I care about. Specifically:
- Jenkins had better kubernetes integration years ago. It's crazy that GHA can't beat that.
- "Reintroducing multi-label functionality" - yeah, so they first broke it. They did supply "reasons", which looked like they never talked to a customer. [1]
- Still no SDK of any kind.
- "Actions Data Stream" - or you can just fix your logging.
There are dozens more complains, which are easy enough to find. This kind of an approach just makes me want to make sure that I don't use GHA again. Even if I end up paying another vendor, at least I'll be treated as a customer.
[1] - https://github.com/orgs/community/discussions/160682#discuss...
For years I've thought about doing an "art project" to make people more aware of the fact they are being observed – but I never actually got up and did it.
The idea was to seek spots in the city where public web cams are pointed at, and paint QR codes on the ground at those spots (using a template), linking to the camera stream. So when curious passerbys scan the code, they see themselves in a camera stream and feel "watched".
That's not a move of a company that thinks it can still grow. That's a Netflix "we have 90% of the market, let's squeeze them" move. This is the beginning. We have all seen this pattern over the last 5+ years. You know their next few moves.
Having worked at Mozilla a while ago, the CEO role is one I wouldn't wish on my worst enemy. Success is oddly defined: it's a non-profit (well, a for-profit owned by a non-profit) that needs to make a big profit in a short amount of time. And anything done to make that profit will annoy the community.
I hope Anthony leans into what makes Mozilla special. The past few years, Mozilla's business model has been to just meekly "us-too!" trends... IoT, Firefox OS, and more recently AI.
What Mozilla is good at, though, is taking complex things the average user doesn't really understand, and making it palpable and safe. They did this with web standards... nobody cared about web standards, but Mozilla focused on usability.
(Slide aside, it's not a coincidence the best CEO Mozilla ever had was a designer.)
I'm not an AI hater, but I don't think Mozilla can compete here. There's just too much good stuff already, and it's not the type of thing Mozilla will shine with.
Instead, if I were CEO, I'd go the opposite way: I'd focus on privacy. Not AI privacy, but privacy in general. Buy a really great email provider, and start to own "identity on the internet". As there's more bots and less privacy, identity is going to be incredibly important over the years.. and right now, Google defacto owns identity. Make it free, but also give people a way to pay.
Would this work? I don't know. But like I said, it's not a job I envy.
Introducing a separate charge specifically targeting those of your customers who choose to self-host your hilariously fragile infrastructure is certainly a choice.. And one I assume is in no way tied to adoption/usage-based KPIs.
Of course, if you can just fence in your competition and charge admission, it'd be silly to invest time in building a superior product.
> We are introducing a $0.002 per-minute Actions cloud platform charge for all Actions workflows across GitHub-hosted and self-hosted runners.
Charging for self-hosted runners is an interesting choice. That's the same cost as their smallest hosted runners [1]
[1] - https://docs.github.com/en/billing/reference/actions-runner-...
They also have the "extensions that can do real ad blocking" angle.
I bought a Kobo because it felt one of the most hackable e-readers but I ended up using the stock software with a single but very important change: I edited a configuration file to point the API to my BookLore instance. This gives me access to all of my ebooks and preserves all the functionalities of the stock software, which is perfectly adequate for my needs.
It turns out that what I wanted all along was the ability to seamlessly read books I buy from any source, not any deeper hacking of the OS.
Kudos to Kobo for keeping their system so open. These days it’s not that common
The root cause of the problem is that parents and children need to raise funds for cancer treatment in the first place.
We have the best mid-air collisions. Noone does it better, or so people tell me. We don't do sleepy silent disappearances over the Bermuda Triangle, that's SAD!! We blow em up, BIGLY, in someone else's airspace. A great PRESIDENT knows how to WIN at mid-air collisions. Thank you for your attention to this matter.
"The campaigns with the biggest apparent international reach were under the name of an organisation called Chance Letikva (Chance for Hope, in English) - registered in Israel and the US."
Chance Letikva is registered with the US IRS as a charity. They've filed a Form 990. Location is Brooklyn, NY. [1] Address is listed. It's a small house. It's also incorporated as CHANCE LETIKVA, INC. in New York State. Address matches. Names of officers not given. There's one name in the IRS filing, listed as the president.
Web site "https://chanceletikva.org" has been "suspended". Domain is still registered, via Namecheap.
Some on the ground digging and subpoenas should reveal who's behind this.
[1] https://projects.propublica.org/nonprofits/organizations/852...
It's expected for military operations to fly without transponder, they don't want to have their location visible. But it's crazy that they're also doing it in Curacao controlled airspace without agreeing a restricted area.
Even for training they set up restricted/military areas in airspace all the time. Not doing it here, in allied (Curacao is part of the kingdom of the Netherlands) airspace is unacceptable. They could have coordinated this in the normal ways so ATC would route civilian traffic around the military operations or talk to the military controllers (who can see both types of traffic) before sending an aircraft through the shared airspace.
This isn't new, it's how military operations are done all the time.
The F-150 is a great power bank that comes with a very useful set of 4 tires and a steering wheel for a truly portable charging experience.
They could also not invade a country that did nothing to attack them, but I guess that’s asking too much.
Because it’s flying near Venezuela, who we’re currently fucking with militarily.
What really worries me is that I keep hearing "cooling is cheap and easy in space!" in a lot of these conversations, and it couldn't be farther from the truth. Cooling is _really_ hard and can't use efficient (i.e. advection-based air or water cooling) approaches and are limited to dramatically less efficient radiative cooling. It doesn't matter that space is cold because cooling is damned hard in a vacuum.
The article makes this point, but it's relatively far in and I felt it was worth making again.
With that said, my employer now appears to be in this business, so I guess if there's money there, we can build the satellites. (Note: opinions my own) I just don't see how it makes sense from a practical technical perspective.
Space is a much harder place to run datacenters.
The certificate lifetime decrease, to 45 days, was discussed in: https://news.ycombinator.com/item?id=46117126
This isn't LE's decision: a 47 day max was voted on by the CA/Browser Forum.
https://www.digicert.com/blog/tls-certificate-lifetimes-will...
https://cabforum.org/2025/04/11/ballot-sc081v3-introduce-sch...
https://groups.google.com/a/groups.cabforum.org/g/servercert... - public votes of all members, which were unanimously Yes or Abstain.
IMO this is a policy change that can Break the Internet, as many archived/legacy sites on old-school certificates may not be able to afford the upfront tech or ongoing labor to transition from annual to effectively-monthly renewals, and will simply be shut down.
And, per other comments, this will make LE the only viable option to modernize, and thus much more of a central point of failure than before.
But Let's Encrypt is not responsible for this move, and did not vote on the ballot.
A decade of “personal cloud box” attempts has shown that the hard part isn’t the hardware, it’s the long-term social contract. Synology/WD/My Cloud/etc all eventually hit the same wall: once the company pivots or dies, you’re left with a sealed brick that you don’t fully control, holding the most irreplaceable thing you own: your data. If you’re going to charge an Apple-like premium on commodity mini-PC hardware, you really have to over-communicate what happens if Umbrel-the-company disappears or changes direction: how do I keep using this thing in 5–10 years without your cloud, your app store, your updates?
The interesting opportunity here isn’t selling a fancy N100 box, it’s turning “self-hosted everything” into something your non-technical friend could actually live with. That’s mostly about boring stuff: automatic off-site backup that isn’t tied to one vendor, painless replacement/restore if the hardware dies, and clear guarantees about what runs locally vs phoning home. If Umbrel leans into being forkable and portable across generic hardware, it has a shot at being trusted infrastructure instead of just another pretty NAS that people regret once the marketing site goes dark.
I love the quote the article starts with:
> Neither of us had prior experience developing mobile apps, but we thought, “Hey, we’re both smart. This shouldn’t be too difficult.”
I think, 40 years from now when we're writing about this last decade or so of software development, this quote is going to sum it all up.
>What is the expected compensation for participants? Compensation varies based on experience level and agency placement. Annual salaries are expected to be in the approximate range of $150,000 to $200,000. Benefits include health insurance, retirement plans, paid time off, and eligibility for performance-based awards.
>Tech Force will primarily recruit early-career technologists
So "early-career" but they're going to get paid GS-14/15 pay[1] in DC? New grad engineers in DC are going to be GS-7/9 at best. This is either a blatant lie, or created by someone who has no idea of how federal pay works (or both).
As an aside, I was a fed for >10 years and left last year for industry but stay in touch with friends still working federal jobs. Before this administration recruiting was extremely difficult and candidate quality was low. I've heard that it's nearly impossible now and in the last 18 months they've only been able to hire a single person. Federal jobs used to be considered stable, with good benefits, but low pay. Now they're unstable, the current administration is actively working to make benefits worse, and the pay is still really low.
[1] https://www.opm.gov/policy-data-oversight/pay-leave/salaries...
One of the dark consequences of America losing its city-upon-a-hill aspirations is we're less able to effectively call out evil abroad. Jimmy Lai should not have been allowed to this quietly.
You should consider dropping that instinct. If you look into how corporations have behaved historically you'd assume evil until proven innocent. Especially US corps.
Reminds me of "cancer alley" [1].
As somebody who's looked in to this a bit, the deeper I dug the more I ultimately moved toward the conclusion (reluctantly) that indeed big corporations are the baddies. I have an instinct to steel-math both sides, but not every issue has two compelling sides to it...
One example of them clearly being the baddies is them paying people to social media astroturf to defend the roundup pesticide online [2].
1. https://en.wikipedia.org/wiki/Cancer_Alley
2. https://galiherlaw.com/media-manipulation-comes-out-during-m...
Thousands of systems, from Google to script kiddies to OpenAI to nigerian call scammers to cybersecurity firms, actively watch the certificate transparency logs for exactly this reason. Yawn.
I'm all for antitrust, but it's a shame the Amazon acquisition was blocked.[0]
iRobot was in a distressed state then, and immediately laid of 1/3 of staff when the deal fell through. I knew a survivor of that mess. Now this.
0: https://www.theverge.com/2024/1/29/24034201/amazon-irobot-ac...
As an aside, it's been fascinating reading the comments here about news media.
People want journalists to publish quickly AND only publish what’s fully verified.
They want anonymous sources named "in the spirit of truth," without grappling with the reality that doing so would instantly dry up anyone risking their job, or worse, to provide information.
They expect journalists to release raw information as soon as they have it, while simultaneously acting as perfect filters; never amplifying rumors, or being wrong, even as new facts emerge.
They want neutrality, except when neutrality conflicts with their priors.
It's no wonder that morale among journalists is at an all-time low. Is any other profession held to such an impossible standard?
Tractors largely replaced human labour in farming about a hundred years ago. Should we have started taxing tractors?
I really have difficulties seeing AI as anything else than yet another type of machinery. If your argument is "but it's replacing ALMOST ALL human labour" - well, the same argument was valid for tractors a hundred years ago (when almost everyone was employed in agriculture).
> There were unspoken rules, commandments passed down from teacher to student, year after year. The first commandment? Thou shalt begin with a proverb or a powerful opening statement. “Haste makes waste,” we would write, before launching into a tale about rushing to the market and forgetting the money. The second? Thou shalt demonstrate a wide vocabulary. You didn’t just ‘walk’; you ‘strode purposefully’, ‘trudged wearily’, or ‘ambled nonchalantly’. You didn’t just ‘see’ a thing; you ‘beheld a magnificent spectacle’. Our exercise books were filled with lists of these “wow words,” their synonyms and antonyms drilled into us like multiplication tables.
Well, this is very interesting, because I'm a native English speaker that studied writing in university, and the deeper I got into the world of literature, the further I was pushed towards simpler language and shorter sentences. It's all Hemingway now, and if I spot an adverb or, lord forbid, a "proceeded to," I feel the pain in my bones.
The way ChatGPT writes drives me insane. As for the author, clearly they're very good, but I prefer a much simpler style. I feel like the big boy SAT words should pop out of the page unaccompanied, just one per page at most.
> I even checked one of his responses in WhatsApp if it's AI by asking the Meta AI whether it's AI written, and Meta AI also agreed that it's AI written
I will never understand why some people apparently think asking a chat bot whether text was written by a chat bot is a reasonable approach to determining whether text was written by a chat bot.
Our household (and I suspect many with us) bought a Roomba specifically to not give the Chinese government a roving camera in our home. Ouch!
According to the Financial Times, Roomba has sold more than 40 million robotic devices, most of them robotic vacuum cleaners.[a]
Many of those vacuum cleaners have cameras, can move around on their own, and are connected to the Internet. If they're taken offline, they stop working. Many have microphones too.
The new Chinese owner will get control of a network of tens of millions Internet-connected, autonomously mobile, camera/microphone-equipped robots already inside people's homes and offices.
More than 40 million is a lot. For comparison, the US has ~132 million households.
---
[a] https://www.ft.com/content/239d4720-aee4-443d-a761-1bd8bb1a1...
A prime example of premature optimization.
Permanent identifiers should not carry data. This is like the cardinal sin of data management. You always run into situations where the thing you thought, "surely this never changes, so it's safe to squeeze into the ID to save a lookup". Then people suddenly find out they have a new gender identity, and they need a last final digit in their ID numbers too.
Even if nothing changes, you can run into trouble. Norwegian PNs have your birth date (in DDMMYY format) as the first six digits. Surely that doesn't change, right? Well, wrong, since although the date doesn't change, your knowledge of it might. Immigrants who didn't know their exact date of birth got assigned 1. Jan by default... And then people with actual birthdays on 1 Jan got told, "sorry, you can't have that as birth date, we've run out of numbers in that series!"
Librarians in the analog age can be forgiven for cramming data into their identifiers, to save a lookup. When the lookup is in a physical card catalog, that's somewhat understandable (although you bet they could run into trouble over it too). But when you have a powerful database at your fingertips, use it! Don't make decisions you will regret just to shave off a couple of milliseconds!
No, not the AI. Just the owner of means of production like AI.
The fact that capital owners successfully avoid contributing to the financing of our states and social systems is, in my view, one of the fundamental problems of our time.
I'm CTO at a vertical SaaS company, paired with a product-focused CEO with deep domain expertise. The thesis doesn't match my experience.
For one thing, the threat model assumes customers can build their own tools. Our end users can't. Their current "system" is Excel. The big enterprises that employ them have thousands of devs, but two of them explicitly cloned our product and tried to poach their own users onto it. One gave up. The other's users tell us it's crap. We've lost zero paying subscribers to free internal alternatives.
I believe that agents are a multiplier on existing velocity, not an equalizer. We use agents heavily and ship faster than ever. We get a lot of feedback from users as to what the internal tech teams are shipping and based on this there's little evidence of any increase in velocity from them.
The bottleneck is still knowing what to build, not building. A lot of the value in our product is in decisions users don't even know we made for them. Domain expertise + tight feedback loop with users can't be replicated by an internal developer in an afternoon.
This is the cost of complacency. They were ahead for so long then the likes of Roborock just left them in the dirt. I remember the first time I tried one of the Roborock devices, and until then I have been a long time Roomba user (like, 20 years). I just couldn’t believe how much better it was. And iRobot just stubbornly refused to iterate on their fundamental products.
It's a positive for a nameless middle manager somewhere who can show their boss a graph with a line moving to the right and up with a title like "AI Adoption Across Platforms" and hit their bonus target.
A few things I've come to personally believe after spending years developing web and robotics software in Python/JavaScript then spending years having to maintain while constantly adding new features and dealing with company pivots:
- The types exist whether you write them down or not.
- If they're not written down, they're written down in your head.
- Your head is very volatile and hard for others to access.
- Typing is an incredibly good form of documentation.
- JSDoc and TypeScript are standards/formats for typing. Like any tools, they both have advantages and disadvantages. Neither is objectively better than the other.
- Make informed decisions on how you'll describe your types, and then be consistent and unsurprising.
- A type checker is the computer saying, "okay then, prove it" about your program's type validity.
- Not every program benefits from the same amount of "prove it."
- Too much can be as bad as too little. You're wasting resources proving throwaway code.
- I like languages that let you decide how much you need to "prove it."
Thank you! Opening an incident as soon as user impact begins is one of those instincts you develop after handling major incidents for years as an SRE at Google, and now at Anthropic.
I was also fortunate to be using Claude at that exact moment (for personal reasons), which meant I could immediately see the severity of the outage.
Hello, I'm one of the engineers who worked on the incident. We have mitigated the incident as of 14:43 PT / 22:43 UTC. Sorry for the trouble.
Props to them for actually updating their status page as issues are happening rather than hours later. I was working with claude code and hit an API error, checked the status page and sure enough there was an outage.
This should be a given for any service that others rely on, but sadly this is seldom the case.
An PWA primarily for my wife and my daughter. They can order their hot chocolate and their coffee as if they were going to grab something at a fancy café downtown, but instead it's at home and I'm the barista. It is quite nice to have for when my wife comes back from work and want something specific, or when we are waiting for the visit of a few friend, they can order exactly the available beverages and everything is ready when they're here.
It was also a good playground for me to implement Web Push notifications (to never miss new orders).
It's a basic Nuxt 3 app with Appwrite as the backend with rough edges, but much enough for our household use !
If you want to spam my phone with notifications, please visit my café : https://mytinycafe.com/alix
I think that the real power of spaced repetition is not in flashcard applications like this. It is in behavior modification.
Let's take a real example to show how this works.
August 19, 2025. My wife called me in to help her decide what to do about a dentist that she thought was ripping her off. A couple of quick suggestions later, and she went to being mad at me about not having heard the problem through before trying to fix it badly. As soon as she was mad, I immediately connected with how stupid what I did was, and that this never goes well. But, of course, it was now too late.
Not a mistake I was going to make for a while. But, given my history, a mistake I was bound to make again.
I changed that. This time I stuck this into my spaced repetition system. Each time the prompt comes up, I remember that scene, holding in mind how it important it is to emotionally engage, not offer quick suggestions, and be sure to listen to the full problem in detail. It takes me less than 30 seconds. Reviewing this prompt, for my whole lifetime, will take less than 15 minutes of work. Just typing this up this time takes more work than I'll spend on it in the next several years.
This mistake hasn't happened since. Not once. And I believe it won't again in my life.
I have literally changed dozens of such behaviors. My wife says that it is like there is a whole new me. She can't believe the transformation.
All it took is looking at spaced repetition as general purpose structured reinforcement, and not as just a way to study flashcards.
Simplification of my digital self. Removed most of my online accounts. Removed all my VPS's. Removed most apps from my phone except core ones. Cancelled a lot of online subscriptions.
In the real world finally moved everything to USB-C. Gave all my old cables away. I have two chargers in my home and a handful of C to C cables. Everything connects to everything now.
Home is now downgraded to a dumb home. Lights work on physical toggles. No hubs or sensors anywhere. Heat and AC is with a dumb panel on the wall.
It feels freeing.
I'm working on porting KiCad to the browser. It's a lot of sweat and tears, multithreading issues and some more sweat. I've updated a port of WxWidgets and now I support all the features KiCad needs with ~200 tests.
Right now I have a build that loads in the browser, but I really want to have "multithreading" which means workers in the web. One can use asyncify with emscripten to translate blocking C++ to WASM, but that transition is not perfect, right now I'm debugging a bug where there's a race condition that halts all execution and the main thread runs in an infinite loop waiting for the workers to stand up. I guess I'll have a few of those ahead.
The main goal is to 1. just have fun 2. use yjs as a collab backend so multiple people can edit the same PCB. This will probably work with pcbnew, KiCad's layout editor, since it has a plugin system and AFAIK I can do the sync layer there. For the rest ( schematic, component editor etc. ) I'll have to figure out something.. KiCad does not sync automatically if you modify a file, I'll have to do some lifting there.
Anyway, it's a lot of fun, I really want this thing to exist, I'm hoping that I won't run into a "wellll, this is just not going to work" kind of issue in the end.
>Running npm install is not negligence. Installing dependencies is not a security failure. The security failure is in an ecosystem that allows packages to run arbitrary code silently.
No, your security failure is that you use a package manager that allows third-parties push arbitrary code into your product with no oversight. You only have "secutity" to the extent that you can trust the people who control those packages to act both competently and in good faith ad infinitum.
Also the OP seemingly implies credentials are stored on-filesystem in plaintext but I might be extrapolating too much there.
I really like this author's summary of the 1983 Bainbridge paper about industrial automation. I have often wondered how to apply those insights to AI agents, but I was never able to summarize it as well as OP.
Bainbridge by itself is a tough paper to read because it's so dense. It's just four pages long and worth following along:
https://ckrybus.com/static/papers/Bainbridge_1983_Automatica...
For example, see this statement in the paper: "the present generation of automated systems, which are monitored by former manual operators, are riding on their skills, which later generations of operators cannot be expected to have."
This summarizes the first irony of automation, which is now familiar to everyone on HN: using AI agents effectively requires an expert programmer, but to build the skills to be an expert programmer, you have to program yourself.
It's full of insights like that. Highly recommended!
American tech companies have been pushing the needle on privacy ever since Google. Then Facebook. They've gradually normalised that privacy does not exist, all for their own capital gain.
There are European alternatives but they need support.
IMHO it requires conscious choices by European citizens to choose more carefully which online services they dedicate their time and money to. Or expect unintended consequences.
The link in the last paragraph provides some data to back up the claim. https://mikelovesrobots.substack.com/p/wheres-the-shovelware... - If the goal is to increase the rate of software production, there isn't much evidence that AI has moved the needle.
Sure, code gen is faster now. And the industry might finally be waking up to the fact that writing code is a small part of producing software. Getting infinitely faster at one step doesn't speed up the overall process. In fact, there's good evidence it that rapid code gen actually slows down other steps in the process like code review and QA.
Everyone's suggestions feel designed to frustrate me. Instructions on how to cajole and plead that seem more astrology than engineering.
This is the pattern I settled on about a year ago. I use it as a rubber-duck / conversation partner for bigger picture issues. I'll run my code through it as a sanity "pre-check" before a pr review. And I mapped autocomplete to ctrl-; in vim so I only bring it up when I need it.
Otherwise, I write everything myself. AI written code never felt safe. It adds velocity but velocity early on always steals speed from the future. That's been the case for languages, for frameworks, for libraries, it's no different for AI.
In other words, you get better at using AI for programming by recognizing where its strengths lie and going all in on those strengths. Don't twist up in knots trying to get it to do decently what you can already do well yourself.
I'm a co-founder at WonderProxy, we didn't make their list (we target people doing application testing, not consumer VPNs).
We're in 100+ countries, and I'll stand by that claim. It's a huge pain in the neck. In our early years we had a lot of problems with suppliers claiming to be in Mexico or South America who were actually just in Texas. I almost flew to Peru with a rackmount server in my luggage after weeks of problems, that plan died when we realized I'd need to figure out how to pay Peruvian income tax on the money I made in country before I could leave.
We've also had customers complaining that a given competitor had a country we'd had trouble sourcing in the Middle East. A little digging on our part and it's less than a ms away from our server in Germany.
> Once the code for all destinations lived in a single repo, they could be merged into a single service. With every destination living in one service, our developer productivity substantially improved. We no longer had to deploy 140+ services for a change to one of the shared libraries. One engineer can deploy the service in a matter of minutes.
If you must to deploy every service because of a library change, you don't have services, you have a distributed monolith. The entire idea of a "shared library" which must be kept updated across your entire service fleet is antithetical to how you need to treat services.
I know multiple people who worked / working at Mullvad and they take their business, security and privacy _very_ seriously. Not surprised to see them shine here.
It's a bit dark, but I'm doing much better now, so happy ending. No need to wish me well or anything, I'm the happiest I've ever been (thankfully).
After reaching an age where bi-polar disorder goes full swing, I was unable to manage manic episodes; they'd spring up and I'd be awake for days and then crash horribly. I lost all hope that I'd be able to hold down a typical job ever again. I became a 24h/7d alcoholic with the goal of never being conscious and trying to sleep through life until it ended.
I was at the local shop where I bought my booze buying a bunch of beer and vodka around 7-8am. A guy near me at the counter made a comment about what a great party must be coming. I looked at him, probably dead-eyed, and said, "I'm an alcoholic."
He put his hand on my shoulder. He didn't say anything. It was just a moment of compassion. It was deeply kind. What was communicated was simply that someone cared and, to this day, I wish I had a way to thank him for that profound gesture.
Hi, I’m the creator of Gleam!
The comment you are replying to is correct, and you are incorrect.
All OTP APIs are usable as normal within Gleam, the language is designed with it in mind, and there’s an additional set of Gleam specific additions to OTP (which you have linked there).
Gleam does not have access to only a subset of OTP, and it does not have its own distinct OTP inspired OTP. It uses the OTP framework.
A friend of mine and I (we are both Aussies) had been staying at my Grandma's house in a rural village in UK and were trying to make our way back to London on Boxing Day. The fact that it was Boxing Day meant that no buses were running, so we started sticking our thumbs out to try and hitch a lift to the nearest town's train station. As you would expect, picking up two 19 year old blokes in the middle of nowhere was not an attractive proposition to your average passerby.
Eventually a guy comes along and picks us up. Tells us he hitched all the way across Europe back in the day so he empathized with us. Says he's on the way to pick up his son (our age) from work, a department store that happened to be on the way to the station.
His son gets into the car, understandably pretty bemused as to why his dad has brought two random stragglers with him!
We get to the station only to find that it's closed, because, yes, it's Boxing Day and trains weren't running either (we hadn't really thought this through). Guy says:
"Don't worry lads, all the family are around ours for Christmas dinner. My brother lives in West London so he can give you a ride there at the end of the night."
So we found ourselves, two foreign students, invited to a complete stranger's Christmas dinner party. We all had so much fun and drank so much that we completely abandoned the London idea and went back to my Grandma's at the end of the night.
And the kid who was our age that got picked up from work? He ended up being my Best Man when I got married 15 years later. True story!
I've got a weird one from a bank.
I was driving from South Carolina to Virginia, I was completely broke, and had exactly $20 in cash, and only a couple dollars in my checking account. I did my math wrong, and didn't have enough money for enough gas to make it home. I was trying to draft behind semi trucks and drive slow to conserve fuel, but it wasn't enough.
I called my bank at the gas station with my needle on empty and asked what would happen if I overdrew my account by $50, and the guy on the phone asked me to explain the situation. Afterwards he said I was good to go.
I asked, what does that mean? He said there's now $50 in your account. You can use it to fill up your car on your debit card.
I filled up my car and made it home. When I checked my account later, expecting to see an overdraft fee, there was a deposit of $50 from some account I didn't recognize. The guy had just transferred me $50 from his own account. I never figured out who this was, so ~18 years later, I'll take this opportunity to say: thank you sir.
I don't know that this is THE nicest thing anyone has ever done, but it was a small thing that made a huge difference in that moment.
Alas is right, China is poised to dominate battery, solar, and EV technology and to translate it to military technology as well. Meanwhile the Republicans are blowing up US alliances and sabotaging the battery/EV industrial development policy that was actually making progress in giving the US hope in catching up.
Hey, Boris from the Claude Code team here. A few tips:
1. If there is anything Claude tends to repeatedly get wrong, not understand, or spend lots of tokens on, put it in your CLAUDE.md. Claude automatically reads this file and it’s a great way to avoid repeating yourself. I add to my team’s CLAUDE.md multiple times a week.
2. Use Plan mode (press shift-tab 2x). Go back and forth with Claude until you like the plan before you let Claude execute. This easily 2-3x’s results for harder tasks.
3. Give the model a way to check its work. For svelte, consider using the Puppeteer MCP server and tell Claude to check its work in the browser. This is another 2-3x.
4. Use Opus 4.5. It’s a step change from Sonnet 4.5 and earlier models.
Hope that helps!
I live in Minnesota and do not own a snowblower. Probably my mistake, but I always joke that I get most of my exercise in the winter. Snow is really heavy for those without context.
A couple years ago we had a particularly bad snowfall. The plow has a nasty hate filled habit of dumping all its snow in my driveway. I had a drift at the end of my driveway about 4 feet high and 6 feet deep. Literally up to my chest. I had spent a solid hour just chipping away at it trying to get my car out and had made very little progress.
Right as I was about to give up in frustration, a man in a bobcat drove by. Moments later he turned around, came back, and asked "would you like me to clear that for you?" I told him that would be amazing. Took him a couple minutes and then he waved and drove off before I got a chance to offer him any money or even thank him.
I think about this guy pretty often, it's absolutely the random act of kindness in my life I have appreciated most.
A recent lesser snowfall for context:
5th grade, my best friend at the time was in a basketball team, just a small town league for kids. I never really played basketball, so I was planning to watch the game then we'd hang out. It was the first game of the season and my friend was getting his uniform from a table when a dad running things asked me what team I was playing on and I said no, I'm just here to hang out with my friend.
He shook his head and said, "No, that won't do. You're on his team, too" and handed me a jersey. Then he went ahead and paid my registration fee.
More than the money, it was the proactive nature of it that struck me at the time. The thing is, if I had asked my parents they probably would have signed me up. But it was one of those things where it would have never crossed my mind to ask. I ws as one of those kids that needed a push every now and then and rarely got one.
I never got very good at basketball but I never missed a game and had a great time with my friend. So not a tragic or desperate story, but still meaningful to me all these years later.
My best friend, his girlfriend, brother, sister and I piled into a minivan in July of 2018 and drove from Boston to SF. Best friend and I both took jobs there out of school and decided to make a trip of the life move.
We painted BOS > SF on the back window. At a gas station in Memphis a random guy walked up to us and said "Make sure you go to Graceland. Can't miss it."
We sort of smile and nod politely and then walk into the gas station to use the bathroom, reload on snacks etc.
10 mins later we come back outside and the same guy comes over "I bought you all tickets to Graceland, who can I text them to?"
Truly such a sick moment. Graceland was a highlight of the trip and to have someone just do such a random kind thing made it that much better. Long live Elvis, long live the King. Thanks again to whoever you are that did that. Respect.
The untapped answer is litigation. Call a lawyer and file against Apple. It may take several business days, and cost $$$$ but it will absolutely light a fire at Apple and get the attention of many-a-human. And if they ignore it, well, maybe a class action lawsuit awaits.
Lot of arrogant people here who think they are safe and better than anybody and blame OP.
It is totally normal in today’s world to depend on cloud services and reasonably difficult to do without it. In China: no WeChat you are practically dead. Here try to join meetings without account, try to send a message on WhatsApp without account, etc… a lot can go wrong very fast. What if you used your Apple account as SSO to other services ?
I back up regularly using Google Takeout and similar tools, but I don’t think it’s fair to shame this author . Even if you have backups , your recent and essential content and credentials will be locked out . 1% of your content is the most important
We all depend heavily on cloud storage and sso . Everything works fine until you are locked out .
And using them isn’t fully voluntary. They are necessary for collaboration . You end up using what your team uses .
You can try to be that “own cloud” snob but it only works if you live in a basement
Every normal person has content in Google , iCloud , OneDrive , Dropbox and maybe more. That’s 4+ single points of failure
You’re just not imaginative enough if you think you’re safe .
OPs only recourse is an insider or a lawyer
> I have escalated this through my many friends in WWDR and SRE at Apple, with no success. Ouch. If he can't get it fixed, it's scary
Here’s a great quote by him:
> In my 30’s, I exercised to look good. In my 50’s, to stay fit. In my 70’s, to stay ambulatory. In my 80’s, to avoid assisted living. Now in my 90’s, I’m just doing it out of pure defiance
>It's just insane that a gift card redemption can trigger this.
It's also the buying of gift cards that can get Apple accounts locked: https://old.reddit.com/r/apple/comments/r8b1lu/apple_will_pe...
If enough of these horror stories are publicized, people will learn to never buy/redeem Apple gift cards because of the real possibility of account bans.
- Don't give Apple gift cards to family and friends: You're potentially ruining the recipient's digital life if they redeem it.
- Don't buy Apple gift cards: You risk ruining your own digital life.
If you've been given an Apple gc for Christmas -- and you have paranoia of the risks -- don't buy anything online that's tied to your Apple ID. Instead, go to the physical Apple store to redeem it. And don't buy an iPhone with it because that will eventually get assigned to an Apple ID. Instead, get a non-AppleID item such as the $249 ISSEY MIYAKE knit sock.
I have thousands of credit-card reward points that could be traded in for Apple gift cards but I don't do it because Apple's over-aggressive fraud tracking means Apple's store currency is too dangerous to use.
To paraphrase an old saying: Live by Big Tech, die by Big Tech.
After nearly 30 years as a loyal customer
I've heard others say this (and was a "loyal advocate" of Windows for around 2 decades myself), but the reality is they simply do not care. You are merely a single user out of several billion.
Many of the reps I’ve spoken to have suggested strange things
That almost sounds like some sort of AI, not a human. But if I were in your situation I'd be inclined to print out that response as evidence, and then actually go there physically to see what happens.
It's just insane that a gift card redemption can trigger this. What's the rationale? It would make more sense if they just locked the person out of redeeming gift cards or something, not the entire account.
But reading horror stories like this is is why I only use the very bare minimum of any of these cloud services. Keep local copies of everything. For developer accounts, I always create them under a separate email so they're not tied to my personal. At least it can minimize the damage somewhat.
It sucks that I have to take all these extra precautions though. It's definitely made me develop a do not trust any big corp mindset.
This is one of the worst stories I’ve seen yet. It sounds like they were “all in” on Apple with zero backups, which shows some questionable judgment, but still, this sort of thing shouldn’t be possible any more than a bank deciding to take all your money with no recourse. (They can close your account, but they can’t keep your money.) Maybe hosts should be required to mail you a hard drive with your data on it when they close your account. Regardless, never assume cloud data is in safe hands.
I've given dozens of talks, but this one seems to have struck a chord, as it's my most popular video in quite a while. It's got over 14k views in less than a day.
I'm excited so many people are interested in desktop UX!
You shouldn't have to hack it, you should have the right to repair the software on your device. Hopefully the Vizio lawsuit will help with that for Linux based devices, signs are looking good though.
https://sfconservancy.org/copyleft-compliance/vizio.html
It’s crazy how Anthropic keeps coming up with sticky “so simple it seems obvious” product innovations and OpenAI plays catch up. MCP is barely a protocol. Skills are just md files. But they seem to have a knack for framing things in a way that just makes sense.
I ran into this issue when porting term.everything[0] from typescript to go. I had some c library dependencies that I did need to link, so I had to use cgo. My solution was to do the build process on alpine linux[1] and use static linking[2]. This way it statically links musl libc, which is much friendlier with static linking than glibc. Now, I have a static binary that runs in alpine, Debian, and even bare containers.
Since I have made the change, I have not had anyone open any issues saying they had problems running it on their machines. (Unlike when I was using AppImages, which caused much more trouble than I expected)
[0] https://github.com/mmulet/term.everything look at distribute.sh and the makefile to see how I did it.
[1]in a podman or docker container
[2] -ldflags '-extldflags "-static"'
For a hacker news article, it misses the crucial option - hacking a smart TV! I have LG OLED jailbroken using rootmy.tv, it was pretty trivial. It's basically a linux computer with a huge screen, you can customize it, SSH into it, map any commands to the remote, etc.
Before I only used monitor, simple DP/HDMI input is all I wanted. But being able to take full control of the tv and connect it with other devices in the house I would normally get Rpi for is pretty convenient!
It’s impressive how every iteration tries to get further from pretending actual AGI would be anywhere close when we are basically writing library functions with the worst DSL known to man, markdown-with-english.
I had a bunch of fun writing about this one, mainly because it was a great excuse to highlight the excellent news about Kākāpō breeding season this year.
(I'm not just about pelicans.)
Is it me or does this seem like naked corruption at its worst? These tech CEOs hang out at the White House and donate to superfluous causes and suddenly the executive is protecting their interests. This does nothing to protect working US citizens from AI alien (agents) coming to take their jobs and displace their incomes.
For a bit more context, those posts are using pipeline parallelism. For N machines put the first L/N layers on machine 1, next L/N layers on machine 2, etc. With pipeline parallelism you don't get a speedup over one machine - it just buys you the ability to use larger models than you can fit on a single machine.
The release in Tahoe 26.2 will enable us to do fast tensor parallelism in MLX. Each layer of the model is sharded across all machines. With this type of parallelism you can get close to N-times faster for N machines. The main challenge is latency since you have to do much more frequent communication.
I follow the MLX team on Twitter and they sometimes post about using MLX on two or more joined together Macs to run models that need more than 512GB of RAM.
A couple of examples:
Kimi K2 Thinking (1 trillion parameters): https://x.com/awnihannun/status/1986601104130646266
DeepSeek R1 (671B): https://x.com/awnihannun/status/1881915166922863045 - that one came with setup instructions in a Gist: https://gist.github.com/awni/ec071fd27940698edd14a4191855bba...
Where possible, I've stopped picking fonts that don't distinguish lowercase l and uppercase I. Words virtually always have redundancy (or context in the sentence) and it's fine in 98% of cases, but too often someone sends a token, password, name, or other string where you need to copy it out to another application to see it and just... why? Why bother?
I/O test for Sans Flex: https://snipboard.io/wXCQq5.jpg
It passes the O0 distinction but not the Il one
Example of a font that passes, Ubuntu: https://fonts.google.com/specimen/Ubuntu?preview.text=10%20I... (custom license but looks similar to GPL in that you can do what you want besides relicensing it as proprietary or removing credits)
Another one, Nunito Sans, using the Open Font License: https://fonts.google.com/specimen/Nunito+Sans?preview.text=1...
IBM Plex Sans is another Open Font License option: https://fonts.google.com/specimen/IBM+Plex+Sans?preview.text... (it has an unusual capital Q style though)
If the company's existence depends on the unfair exploitation of its staff, its foreclosure is inevitable and justified, and that is simply the price everyone involved must pay to maintain equilibrium.
Good for them. During economic downturns, when fewer resources are available for redistribution, collective action across population groups can help address worsening power imbalances.
Ugh, I looked this up and it completely ruined it for me:
https://refuga.com/karl-bushby-the-man-who-chose-to-walk-aro...
This is how much he had to sacrifice. Leaving his only son when he was just five and not being able to watch him grow up like any other normal father. He also sacrificed a father/son relationship that may never be restored. “Out of everyone I knew in this world, I knew my son least of all.” Karl didn’t have any means of communication with his son for years but managed to reach him after contacting one of his friends on Facebook. While he was away, his son was suffering from depression and self abuse and had to use medication and therapy.
That's not sacrifice, that's abandonment. I have a young son not far from that age and trying to imagine how he'd feel if daddy just walked off nearly brings me to tears.
In my opinion, every manufacturer of a programmable device should not be allowed to prevent the buyer from reprogramming it.
I don't believe iOS app reviewers actually do any of that, even if on paper they do.
It is because OpenAI bought 40% of the world’s production capacity overnight. RAM is like toilet paper during Covid now.
To avoid misunderstandings, this repository is about a project at Cornell University named the Flexible Extensible Digital Object Repository Architecture (FEDORA), not a Red Hat one.
There’s a fundamental trade-off between performance and privacy for onion routing. Much of the slowness you’re experiencing is likely network latency, and no software optimization will improve that.
When Karl was preparing to cross the ice from Alaska to Russia, I worked with him a bit on a kite-flown camera system to help him get a Birds Eye view of the flows to chart his course. I engineered a ruggedized wireless camera in an aluminum housing, I don’t remember much about it other than I was doubtful that the resolution would be able to give him the data he needed on on small low resolution screen. (This was before consumer drones were common or affordable). We built some devices, not sure if he ever used them or if they helped. I urged him to do a lot of testing to make sure they would be worth the weight.
We spent a lot of time at college coffee house in Fairbanks Alaska working over the ideas and overall design.
Nice fellow, strange aspirations, indomitable spirit. I’m glad to see his trek is nearing completion, and I wish him well on his further adventures. Good luck and Godspeed, Karl.
A key limiting factor for dietary use of single cell protein is the high mass fraction of nucleic acid, which limits daily consumption due to uric acid production during metabolism. High rates of RNA synthesis are unfortunately necessary for high protein productivity.
The paper notes:
>It is important to note that MP products often contain elevated levels of nucleic acids, constituting ~8% of the dry weight [17], which necessitates consideration when assessing their suitability for human consumption. To address this, a heat treatment process is employed at the end of fermentation that reduces the nucleic acid content in the fermented biomass to below 0.75/100 g, while simultaneously deactivating protease activity and F. venenatum biomass. However, this procedure has been observed to induce cell membrane leakage and a substantial loss of biomass, as evidenced in the Quorn production process [17], which also utilizes F. venenatum as the MP producer. Our experimental trials have encountered similar challenges, achieving a biomass yield of merely ~35%, and observed that heating process increased the relative protein and chitin content (Figure 2D,E), which may be related to the effect of membrane leakage, while the intracellular protein of the FCPD engineered strain was less likely to be lost to the extracellular. Thus, concentrating the fermentation broth to enhance protein and amino acids content in successive steps to produce a highly nutritious water-soluble fertilizer appears to be an effective strategy for adding value to the process (Figure 1).
The challenges of developing economic single cell protein products, that are suitable for human consumption, are described in chapter 3 here:
https://www.researchgate.net/profile/Martin-Hofrichter-2/pub...
In my experience, the best models are already nearly as good as you can be for a large fraction of what I personally use them for, which is basically as a more efficient search engine.
The thing that would now make the biggest difference isn't "more intelligence", whatever that might mean, but better grounding.
It's still a big issue that the models will make up plausible sounding but wrong or misleading explanations for things, and verifying their claims ends up taking time. And if it's a topic you don't care about enough, you might just end up misinformed.
I think Google/Gemini realize this, since their "verify" feature is designed to address exactly this. Unfortunately it hasn't worked very well for me so far.
But to me it's very clear that the product that gets this right will be the one I use.
Traffic to my blog plummeted this year and you can never be entirely sure how it happened. But here are two culprits i identified.
1. Ai overview: my page impressions were high, my ranking was high, but click through took a dive. People read the generated text and move along without ever clicking.
2. You are now a spammer. Around August, traffic took a second plunge. In my logs, I noticed these weird queries in my search page. Basically people were searching for crypto and scammy websites on my blog. Odd, but not like they were finding anything. Turns out, their search query was displayed as an h1 on the page and crawled by google. I was basically displaying spam.
I don't have much control over ai overview because disabling it means I don't appear in search at all. But for the spam, I could do something. I added a robot noindex on the search page. A week later, both impressions and clicks recovered.
Edit: Adding write up I did a couple weeks ago https://idiallo.com/blog/how-i-became-a-spammer
It would be impossible to do without taking breaks, as explained in the article:
> Due to visa limits, Bushby has had to break up his walk. In Europe, he can stay for only 90 days before leaving for 90, so he flies to Mexico to rest and then returns to resume the route.
Given that he literally swam across the Caspian Sea in order to avoid Russia and Iran because of legal issues, nevermind bring imprisoned in Russia due to what sounded like bureaucratic BS, it's more impressive than I first thought.
The N900 was my peak “mobile computing is awesome” device.
I went to see District 9 in the cinema in Helsinki. Uh oh, the alien parts are only subtitled in Finnish and Swedish and my Finnish is not up to that.
I installed a BitTorrent client, found the release on Pirate Bay, successfully torrented just the subtitle file, and used an editor to read the subtitles for scenes with a lot of alien.
The N9 had much better UI, but there was something of the cyberpunk “deck” idea in that thing, it was great.
GNU's version of Yacc is called Bison. Pine Is Not Elm (even though that was never an official acronym). UNIX was UNICS which was a pun on MULTICS. I couldn't for the life of me tell you what dd stands for. nano is a copy of pico which was the "PIne COmposer". Postfix is a completely opaque portmanteau of post (as in mail) and "bug fix". C++ is "C incremented", and C is the successor of B, which is the successor of BCPL.
Developers haven't "lost the plot", we never had it in the first place.
Inversely, Clang, LLDB, jq, fzf, loc are modern projects perfectly in line with the author's notion of a good name. "mise-en-place" is the perfect metaphor for what mise does.
On a related note, they built their digital ID so that third parties could verify attributes (it's NOT just a single-service login across government + a linking ID across government services, which is how it was sold by the BBC).
They're pretty close to completely de-anonymising the internet for UK citizens. Say they introduce an Australian-style social media ban for under 16s, then requires all social media to link their accounts to digital IDs for this verification.
Naturally the only remaining loophole is if a UK citizen manages to avoid being flagged as British ever by using a VPN, so I expect they will focus on that going forwards. Keep in mind the UK already arrests and imprisons vast numbers of people for speech offences, there's no slippery-slope argument here because the UK is already at the bottom of the slope as an ultra-authoratitarian anti-speech nation.
React Server Components always felt uncomfortable to me because they make it hard to look at a piece of JavaScript code and derive which parts of it are going to run on the client and which parts will run on the server.
It turns out this introduces another problem too: in order to get that to work you need to implement some kind of DEEP serialization RPC mechanism - which is kind of opaque to the developer and, as we've recently seen, is a risky spot in terms of potential security vulnerabilities.
> grep (global regular expression print), awk (Aho, Weinberger, Kernighan; the creators’ initials), sed (stream editor), cat (concatenate), diff (difference). Even when abbreviated, these names were either functional descriptions or systematic derivations.
If you asked someone unfamiliar with unix tools what they thought each of these commands did, diff is the only one which they would have even the slightest chance of guessing. It's ridiculous to complain about "libsodium" and then hold up "awk" as a good name.
I suppose this is as good a place as any to mention this. I've now met two different devs who complained about the weird responses from their LLM of choice, and it turned out they were using a single session for everything. From recipes for the night, presents for the wife and then into programming issues the next day.
Don't do that. The whole context is sent on queries to the LLM, so start a new chat for each topic. Or you'll start being told what your wife thinks about global variables and how to cook your Go.
I realise this sounds obvious to many people but it clearly wasn't to those guys so maybe it's not!
This comment section is strange, a lot of people trying to discredit Snowden, saying he shouldn't have released the files, should be in prison, etc. 12 years ago this was HUGE news and had a major impact on the internet and everyone thanked Snowden for these documents! I certainly am thankful. Disappointed in my country that they literally said that "spying between friends is a no-go" but then did nothing and intimidated journalists and legalized it instead. And thanks to the author for giving the documents another look, found it very interesting. There is also part 2: https://libroot.org/posts/going-through-snowden-documents-pa...
This keeps coming up and we keep having the same debates about what Age Verification isn't.
For the folks in the back row:
Age Verification isn't about Kids or Censorship, It's about Surveillance
Age Verification isn't about Kids or Censorship, It's about Surveillance
Age Verification isn't about Kids or Censorship, It's about Surveillance
Without even reaching for my tinfoil hat, the strategy at work here is clear [0 1 2]. If we have to know that you're not a minor, then we also have to know who you are so we can make any techniques to obfuscate that illegal. By turning this from "keep an eye on your kids" to "prove you're not a kid" they've created the conditions to make privacy itself illegal.
VPNs are next. Then PGP. Then anything else that makes it hard for them to know who you are, what you say, and who you say it to.
Please, please don't fall into the trap and start discussing whether or not this is going to be effective to protect kids. It isn't, and that isn't the point.
0 https://www.eff.org/deeplinks/2025/11/lawmakers-want-ban-vpn...
1 https://www.techradar.com/vpn/vpn-privacy-security/vpn-usage...
2 https://hansard.parliament.uk/Lords/2025-09-15/debates/57714...
> Only downsides are in the polish department
What an understatement. It has me thinking „man, fuck this“ on the daily.
Just today it spontaneously lost an entire 20-30 minutes long thread and it was far from the first time. It basically does it any time you interrupt it in any way. It’s straight up data loss.
It’s kind of a typical Google product in that it feels more like a tech demo than a product.
It has theoretically great tech. I particularly like the idea of voice mode, but it’s noticeably glitchy, breaks spontaneously often and keeps asking annoying questions which you can’t make it stop.
I have been using chatGPT a ton over the last months and paying the subscription. Used it for coding, news, stock analysis, daily problems, and a whatever I could think of. I decided to give Gemini a go when version three came out to great reviews. Gemini handles every single one of my uses cases much better and consistently gives better answers. This is especially true for situations were searching the web for current information is important, makes sense that google would be better. Also OCR is phenomenal chatgpt can't read my bad hand writing but Gemini can easily. Only downsides are in the polish department, there are more app bugs and I usually have to leave the happen or the session terminates. There are bugs with uploading photos. The biggest complaint is that all links get inserted into google search and then I have to manipulate them when they should go directly to the chosen website, this has to be some kind of internal org KPI nonsense. Overall, my conclusion is that ChatGPT has lost and won't catch up because of the search integration strength.
> This would be career suicide in virtually any other technical field.
This article would certainly disagree with you:
https://en.wikipedia.org/wiki/List_of_U.S._Department_of_Def...
> the Golden Gate Bridge tells you it spans the Golden Gate strait.
Is that even a meaningful distinction? Does anyone think, "Gee, I'd really like to cross the Golden Gate strait?" or do they think "I want to get to Napa?".
> The Hoover Dam is a dam, named after the president who commissioned it, not “Project Thunderfall” or “AquaHold.”
It was actually called the "Boulder Canyon Project" while being built, referred to as "Hoover Dam" even though finished during the Roosevelt administration, officially called "Boulder Dam", and only later officially renamed to "Hoover Dam".
The fact that Herbert Hoover initiated the project tells you nothing meaningful about it. Would "Reitzlib" be a better name than "Requests"?
> If you wrote 100 CLIs, you will never counter with a cobra.
But out in the real world, you could encounter a Shelby Cobra sports car, Bell AH-1 Cobra chopper, USS Cobra (SP-626) patrol boat, Colt Cobra handgun, etc.
> No chemist wakes up and decides to call it “Steve” because Steve is a funny name and they think it’ll make their paper more approachable.
When you open your medicine cabinet, do you look for a jar labeled "acetylsalicylic acid", "2-propylvaleric acid", or "N-acetyl-para-aminophenol"? Probably not.
It's a bad sign when all of the examples in an article don't even agree with the author's point.
I've benchmarked it on the Extended NYT Connections benchmark (https://github.com/lechmazur/nyt-connections/):
The high-reasoning version of GPT-5.2 improves on GPT-5.1: 69.9 → 77.9.
The medium-reasoning version also improves: 62.7 → 72.1.
The no-reasoning version also improves: 22.1 → 27.5.
Gemini 3 Pro and Grok 4.1 Fast Reasoning still score higher.
Oh hey this is using my go sqlite vfs module[0]. I love it when I find out some code I wrote is useful to others!
[0]: https://github.com/psanford/sqlite3vfs
That would be a laudable goal, but I feel like it's contradicted by the text:
> Even on a low-quality image, GPT‑5.2 identifies the main regions and places boxes that roughly match the true locations of each component
I would not consider it to have "identified the main regions" or to have "roughly matched the true locations" when ~1/3 of the boxes have incorrect labels. The remark "even on a low-quality image" is not helping either.
Edit: credit where credit is due, the recently-added disclaimer is nice:
> Both models make clear mistakes, but GPT‑5.2 shows better comprehension of the image.
Yep, the point we wanted to make here is that GPT-5.2's vision is better, not perfect. Cherrypicking a perfect output would actually mislead readers, and that wasn't our intent.
Is it me, or did it still get at least three placements of components (RAM and PCIe slots, plus it's DisplayPort and not HDMI) in the motherboard image[0] completely wrong? Why would they use that as a promotional image?
0: https://images.ctfassets.net/kftzwdyauwt9/6lyujQxhZDnOMruN3f...
I think we have different opinions on what's fun and what's boring!
I have had this conversation with several people. I feel like I used to be able to type with a fairly low error rate on a smaller screen with old iPhones. Now I feel that it is constant exercise in frustration as I will hit a letter and the keyboard will decide to pick the letter next to it. It is evolving backwards.
Claude is really good at specific analysis, but really terrible at open-ended problems.
"Hey claude, I get this error message: ", and it'll often find the root cause quicker than I could.
"Hey claude, anything I could do to improve Y?", and it'll struggle beyond the basics that a linter might suggest.
It suggested enthusiastically a library for and it was all "Recommended" about it, but when I pointed out that the library had been considered and rejected because , it understood and wrote up why that library suffered from that issue and why it was therefore unsuitable.
There's a significant blind-spot in current LLMs related to blue-sky thinking and creative problem solving. It can do structured problems very well, and it can transform unstructured data very well, but it can't deal with unstructured problems very well.
That may well change, so I don't want to embed that thought too deeply into my own priors, because the LLM space seems to evolve rapidly. I wouldn't want to find myself blind to the progress because I write it off from a class of problems.
But right now, the best way to help an LLM is have a deep understanding of the problem domain yourself, and just leverage it to do the grunt-work that you'd find boring.
The negativity towards this is wild. A company followed relatively widely accepted industry practice (lots and lots of other games also have huge sizes on disk for the exact same reason), then eventually they decided to do their own independent testing to check whether said common practice actually makes things better or not in their case, found that it didn't, so they reversed it. In addition, they wrote up some nice technical articles on the topic, helping to change the old accepted industry wisdom.
This seems great to me. Am I crazy? This feels like it should be Hacker News's bread and butter, articles about "we moved away from Kubernetes/microservices/node.js/serverless/React because we did our own investigation and found that the upsides aren't worth the downsides" tend to do really well here. How is this received so differently?
I actually straight up don't think they give a shit anymore.
I think decorum works in an environment where decorum is the norm, but we have entered a political moment where that is no longer the case. And I think that this kind of thing bleeds so heavily into culture that they no longer give a shit about having their characters next to it.
They have enough plausible deniability; they did not create the content. I think that's enough for them, in this moment.
buried the lede:
> As part of the agreement, Disney will make a $1 billion equity investment in OpenAI, and receive warrants to purchase additional equity.
I say this with no snark or disdain: Sam has mastered the art of the flywheel.
Re licensed ai videos, if anyone wants to see the perspective the C-suites are being sold on, check out this episode of Belloni's The Town, in which they discuss the vision for AI + IP https://overcast.fm/+AA4DU9JreIE
This will not end well for Disney, there were certain historical characters removed from Sora 2 because people kept making racist videos that are hard to censor, and it became increasingly unhinged. This feels like another circular investment where Disney is hoping to make money back I'm sure. On the other hand, assuming they do the freemium stuff, I look forward to making a few videos of my daughters favorite Disney princesses "talking" to her.
OP is the original upload, but the agency reposted it with English subs after it got popular outside of France: https://www.youtube.com/watch?v=iLERt5ZkpQ4
I think we all can agree that Comic Sans MS reflects the current US government best, both spiritually and aesthetically.
Our studio, LucasFonts, designed Calibri. Here are our CEO Luc(as) de Groot’s thoughts on the matter:
The decision to abandon Calibri on the grounds of it being a so-called “wasteful diversity font” is both amusing and regrettable. Calibri was specifically designed to enhance readability on modern computer screens and was selected by Microsoft in 2007 to replace Times New Roman as the default font in the Office suite. There were sound reasons for moving away from Times: Calibri performs exceptionally well at small sizes and on standard office monitors, whereas serif fonts like Times New Roman tend to appear more distorted. While serif fonts are well-suited to high-resolution displays, such as those found on modern smartphones, on typical office screens the serifs introduce unnecessary visual noise and can be particularly problematic for users with impaired vision, such as older adults.
Professional typography can be achieved with both serif and sans-serif fonts. However, Times New Roman—a typeface older than the current president—presents unique challenges. Originally crafted in Great Britain for newspaper printing, Times was optimised for paper, with each letterform meticulously cut and tested for specific sizes. In the digital era, larger size drawings were repurposed as models, resulting in a typeface that appears too thin and sharp when printed at high quality.
Serif fonts are often perceived as more traditional, but they are also more demanding to use effectively. While a skilled typographer can, in theory, produce excellent results with Times, using it in its default digital form is not considered professional practice.
Calibri, by contrast, incorporates extensive spacing adjustments and language-specific refinements. The digital version of Times New Roman, developed in the early days of computing, offers only minimal kerning and letter-pair adjustments. This is especially evident in words set in all capitals—such as “CHICAGO”—where the spacing is inconsistent: the letters “HIC” are tightly packed, while “CAG” are spaced too far apart. Microsoft cannot rectify these issues without altering the appearance of existing documents.
