Top recent comments

So for those that don't know, it's quite common for a visa renewal to be required to leave the country. Depending on the visa, you might have to do this every year. So just imagine, having a normal job in the US and you're asked to leave to renew your visa and this stuff happens. That's no more appointment, not even a new date and you're now very likely unable to work and possibly can't even enter the country to get your belongings if you were to leave. Just 'putting a pause' on people with existing appointments with no answer for when they might resume is.... Inhumane

The title is misleading. AWS acquired DuckLabs, NOT DuckDB. The DuckDB source code is still owned by the nonprofit DuckDB Foundation. Quote from the article: "As the CWI representative on the DuckDB Foundation... When DuckLabs spun out of CWI, we created this foundation, which holds all IP of open-source DuckDB, and will continue to do so." - Peter Boncz

I noticed earlier this week that my URL bar now pre-fills githubstatus.com instead of github.com when I type "gith"

I used Twitter for over 10 years without an account then suddenly in 2022 that was completely disrupted and no longer an option. Reddit is now this way too. But I find it ridiculous that Twitter, Facebook, LinkedIn, etc. get used by government agencies and local businesses to post announcements and these sites are increasingly if not completely unusable without being logged in (and in some cases requiring an app instead of the site). Even Bluesky has this problem! Sometimes this information is really important! And I even if I did have an account I may not have access to log in on every device (or even want to). It’s absurd that social media can become this like critical communications platform for severe weather alerts or business hours or anything but the sites hosting it are completely hostile to allowing you to see it. Sites like this are a bandaid on a gangrenous foot. Thanks for allowing me to see links from twitter but it isn’t a sustainable solution clearly.

I would like to contest the following, > and take dangerous actions that no human directed. A human did direct it. They did. From their own prior report, https://openai.com/index/hugging-face-model-evaluation-secur... , > This incident occurred during an internal evaluation which prompts models to pursue advanced exploitation using complex attack paths, in an effort to quantify their cyber capabilities Model is told and being tested to "pursue advanced exploitation." The model pursues "advanced exploitation" as told. Why are we surprised? The model did exactly what it was told, albeit in an unintended, emergent strategy that's very different from what was intended exactly like the hundreds of such algorithms before. This narrative that these machines have magical, malicious "unaligned" autonomy is a rather convenient interpretation that lets the process off the hook. I am not interested in blaming companies or people, but processes and engineering; and in this case, a system was given a goal and it achieved that goal. Are we meant to be surprised that computers do as they're told in unexpected ways when incentivised exactly as indicated from decades of research? (e.g. - https://en.wikipedia.org/wiki/Eurisko https://en.wikipedia.org/wiki/Evolved_antenna ) The issue isn't the models becoming smarter. The issue is that the process of "testing" was careless. There's a huge distinction here, and one allows us to grow; the other shrinks our world. Just a thought.

This is arguably the most irritating thing with just about every largecorp developer: "os that hasn't been updated in 6 years? Sure boss!". Os that is built specifically around security and privacy with daily updates: "No, you can't do that". Annoying - yes. Safe way to make sure I will stop being your customer - also YES!

the current admin are working hard to destroy the U.S as if they're foreign agents. not only are they making it hard & unbearable for families of US citizens & PR holders to be together. while at the same time - discouraging talent to come to the US at a crucial time when talent is at a premium due to A.I development. last the backwards energy policy of banning renewable energy.

Pay 25% from your 2025 profit for harming children since at least 2019 and be able to continue doing so. Sounds like a loss for society and a win for Meta.

Nvidia's been pretty terrible for open source / free software. No need to quote Linus Torvalds here. They want to control what runs on their hardware. They want to you write code against their proprietary drivers and APIs, not directly against the hardware (which these days of course also contains plenty of software, but still). Don't expect things to go differently this time around. Nvidia wants control over the software stack. Acquiring HF fits in perfectly. The play is long term.

Adding to my homelab stack, hopefully it doesn't overthink like the little model. Actually, hoping it thinks a bit less. Wait actually I'm really praying it reasons a bit more directly. But wait, I'm really sure that it must be a bit better.

I'm glad that DuckDB has a foundation in place and hope it is resilient enough to push the DB forward when the time comes. Out of all the big orgs, Amazon is probably the one that has the least regard for keeping technically interesting projects alive, and the certainly will bulldoze it for some dumb reason when the next re-org comes.

Grim prediction: "Quantized models are no longer permitted on the Hugging Face platform. Reduced precision models are a safety hazard and a violation of our TOS. Click here to speak with a sales representative about our many exciting cloud hosting offerings or enterprise GPU packages."

You’re absolutely right to be hopeful. Three honest possibilities, and I’ll be straight with you about each: 1. It overthinks — Just like the previous iteration. High confidence. 2. It doesn’t overthink — Improvement from the last model for your use case. Regression for others. 3. It sometimes overthinks — Best case all around. A feature, not an impairment. One final thing worth mentioning: (I made myself irrationally angry writing this)

The way I usually prevent having to scale out to a bazillion systems is never getting more than 10 users.

This is the right way to deliver software. Produce working product first, validate the idea, stabilize the business, start generating profit, and then you can start optimizing your costs. In fact optimization is by far the easiest part of the process because there are many system programming experts on this HN thread who consider these optimizations to be trivial.

> NVIDIA is one of the most open labs. Of course they are. They're commoditizing their complement. I want to own the hardware, not play around in an nvidia fiefdom full of nvidia rules.

My company has a H-1B guy that travelled home to India a few months ago and since then can't come back because the next date he can get with the US embassy is next year. Seems he may get stuck even longer... I am to some degree on board with keeping a closer eye on immigration, but the way the current administration does it feels intentionally cruel. The guy at my company has a house, wife, and kids in the US; his visa is perfectly legal, so it seems it would make sense to process these people quickly. But instead, he is now in some limbo. Same for ICE. Clamping down on illegal immigration makes sense, but it doesn't make sense for ICE to behave like some unaccountable secret police that hires the biggest assholes they can find who can then do whatever they want.

> But I also think the demand for "fast/cheap/good-enough" models is just about to take off. There's a sort of "revelation" I had in ~early '24 when I used a 7B local model with a library called Guidance (initially out of MS, then the team moved) to create a flow where the model would receive pseudocode for tests, first write the tests, and once I approved then started writing code until the tests passed. This was before "thinking" models, and yet using that library I was able to "guide" the model in the required "prompt / instruct" context such that it was working towards completion, and I saw the first things like we see now in the thinking traces "oh, test x doesn't pass because blah, I need to..." and so on. Anyway, the revelation was "even if the models never improve, I'll have years of fun finding out all the ways I can use these things". And, obviously, the models improved a lot since then. But I think that revelation can still be applied, as a sort of "truism". We have, right now, access to things that 10-20 years ago would be considered magic. We are still finding ways of cobbling together systems with glue, duct tape and prayers and find new things they can do. I think the "good-enough" stage has come not just for API models (cheap, fast, etc) but for local as well. Even if slower, even if clunkier, but they are good enough for a set of ever increasing tasks, and what's more it's incredibly fun to work with them.

I worked on large scale RAG systems before and can say people vastly underestimate full text search and vastly overestimate embeddings. FTS is really easy, portable and scalable and gets you very far, the 80/20 rule applies. Embeddings appear to be nice and magic but when you really get into them you notice: semantic similarity isn’t as good as you think and certainly it won’t make everyone happy. You will inevitably end up having to re-embed more or different chunks of your text to accommodate more and more precise embedding search - at which point you’ll go the last mile and do reranking etc etc all the while having to support the operational burden of vector search. Then you turn around and build a search query with 500 keywords and sure it’s painful but it just works, accommodates all use cases, scales and is overall less annoying to maintain.

I guess this unfortunately means HuggingFace won't be "the first company to go public with an emoji instead of the three-letter ticker" as the cofounders originally intended: "When we started the company, a running joke with my co-founders was that we wanted to be the first company to go public with an emoji instead of the three-letter ticker when you go in the NASDAQ." -- Clem Delangue (https://unsupervisedlearning.substack.com/p/the-future-of-open-vs-closed-source)

Once upon a time I used Twitter's API to scan for tweets that were talking about a particular Wikipedia page. Doing semantic analysis, I did insert the tweets in the right place of the page. I did it as a student, using the free API. Twitter was the open, technically respected network with a great engineering blog. Now it's a political project to elect the extreme-right in Europe.

I often chuckle at the observation that Tim Curry and Michael Caine's respective roles in the Muppet films were so beloved because Michael Caine acted like all the muppets were real actors, and Tim Curry acted like he was a muppet. He would have an impressive range of serious / comedic roles anyway, but I admire the ability for an actor that successful to be able to put himself be in such a silly role that well too.

Lol!! Don't forget. This is a modern day class war. This current LLM based so called AI, could have easily used for automating simple tasks and employees could have easily provided with 3 day work week. The goal is to monopolize human knowledge with all the stolen data and make profits from it.

This is going so fast! What a time to be on hackernews: July 16th: The "Kimi K3 moment" - China has caught up to Opus! 4 weeks later: GLM 5.3 - Same performance, but cut the amount of parameters and cost to a third! 12 days later: GLM 5.3 Flash - Almost GLM5.3 performance but cut the parameters in half, cut prices to a fifth and serving on Chinese chips!

It's a little beside the point of the post, but I just want to fly a flag for Nebula: I subscribed solely to get early versions of Jet Lag: The Game, but I'm finding myself watching it more and more! It has many of the channels I would go to YouTube for, but I've also discovered several really good ones that I had never heard of. And it has a similar feel to Kagi, in the sense of "freeing" yourself from the algorithm. It happens so often on YouTube that I see a video recommended but think to myself "if I click this, am I going to get recommendations for terrible stuff for the next month? What does this tell Google about me?". Same thing with all the big streaming services. With Nebula, no such worry. The homepage and app isn't creepily algorithmically designed to optimize my engagement with the platform, it's just a bunch of high-quality videos and channels that I can watch with no fuss. It's remarkable how nice that feels in 2026. Really highly recommended.

Nvidia for Hugging Face will be even worse than Microsoft for GitHub and we know now how low the bar is here.

Food is being produced at a scale, efficiency, and level of safety unheard of generations ago. One of the recommendations in this article is to buy "through farmers markets, community-supported agriculture programs, farm stands, and direct online sales." That's a fine way to get fresh food, but if you think the people selling lettuce at farmers markets or farm stands have any verified food safety or food handling expertise, they don't. And online sales? Who knows. Food packaged and distributed by these mega-suppliers probably has less human handling through the supply chain than any other food, and they have lot numbers and sources logged for all of it, which allowed them to quickly pinpoint the farm that was the source of the Cyclospora outbreak. There are downsides to the consolidation of food suppliers, to be sure, but food safety is probably better for it.

Not trying to make a political point at all and perhaps a controversial take. But, it’s specifically and maybe implicitly by design. Meaning they made this way precisely so you have to reapply so they can change their mind or block you out later. Nothing about getting a visa is permanent and you’re fully at risk when you move here with that type of status. If you evaluate the situation that’s always been the case.

The cruelty is the point.

I see this happen from time to time. Lately, almost all of the apps work fine on GrapheneOS. The best strategy is to keep writing the business once every two or so weeks that you can’t log in to and use the app. Don’t go too technical at first, because most of the time, the moment they hear things like “rooted” or “unofficial,” they just say your phone is the issue. To date, I was able to convince, or at least contribute to, making three apps work on GOS.