sxp
Joined 14 years ago
Karma
5,842
Submissions
711
Member since
Jun 25
Virtual Rōnin wandering the Metaverse. Ex-Oculus. Ex-Daydream.
View on Hacker NewsRecent submissions
Comment · 19 days ago
Wow. This is a pretty bald-faced lie by Google. They're defining "datacenter" very narrowly in a way that lets them hide their actual footprint. E.g, they're ignoring the various caches they have in the SF Bay Area, NYC, and other metropolitan locations. Back when they were running Stadia, their claim was that they had edge nodes single-digit milliseconds from major population centers. They also run their servers in datacenters which are technically not owned by "Google". That's why Google Cloud has an us-west2 near LA not on this map but their GCP site makes different claims from this website: https://cloud.google.com/about/locations
Comment · 23 days ago
Note that this appears to be about green cards/permanent residency rather than h1b.
Labor Secretary Keith Sonderling said other companies that would be suspended from the Permanent Labor Certification Program included Cognizant Technology Solutions Corp., Infosys Ltd., Tata, Wipro Ltd., HCL Technologies Ltd., and Capgemini SE. He said there were multiple active federal investigations ongoing into the matter.Comment · 27 days ago
This is one of my favorite books and it's fun to read for anyone interested in the AI consciousness debate. https://hpmorpodcast.com/blindsided-00-announcement/ is a podcast that read the book last year. And https://blindsight.space/ is a Watts' approved website that has a short film/movie trailer for the book.
Comment · 1 month ago
Since these articles tend to actually bury the data that was scraped:
> .. the information likely included one's "public profile, page likes, birthday and current city" ..
This information was limited to what was publicly accessible on the site rather than some private data as many outlets tend to claim.
Comment · 1 month ago
This is a shitty article. It doesn’t mention that the ban on social ads in the EU is due to their anti-free speech rules rather than a Meta specific policy. Meta is in favor of social activism ads in general but restricts them in the EU: https://about.fb.com/news/2025/07/ending-political-electoral...
Comment · 1 month ago
Earth is round so the map should also be round.
Comment · 1 month ago
Booo...they should have gone with https://en.wikipedia.org/wiki/Mollweide_projection
Comment · 1 month ago
Note that San Mateo closed off the street to all traffic and turned it into a pedestrian walking/eating area. So it's used for events. Market Street still allows traffic so people can't wander around in it.
Comment · 2 months ago
The Google blog post is more useful: https://blog.google/products-and-platforms/products/maps/gni...
The U.S. Geographic Names Information System (GNIS) has formally changed the name for "Lake Ontario" to "Lake America" in the United States. Since we update Google Maps to reflect name changes in official government sources, which is GNIS for the U.S., people using Maps in the U.S. will see "Lake America," those in Canada will continue to see "Lake Ontario," and those outside of the U.S. and Canada will see both names. These updates follow our long-standing policy for bodies of water with names that vary from country to country, and are starting to roll out now.
Comment · 2 months ago
I'm getting a 503 on the server, but the Wayback has an abstract.
https://web.archive.org/web/20260815004033/https://studios.d...
Vectorization of line art and sketches is an important problem in computer graphics and remains an active area of research. In this paper, we leverage several recent advances in the field to introduce a new state-of-the-art method for line art vectorization. First, we show that it is possible to go beyond heuristics to extract a meaningful set of strokes from a raster image. We leverage both depth prediction and semantic feature extraction models to propose a more principled approach to split the skeleton graph of the sketch into subgraphs to initialize a set of strokes, producing results that better align with artistic intent. Second, we model strokes as Bézier curves with both geometric and appearance components, and employ 2D Gaussian splatting for fast, di erentiable rendering. This formulation enables e cient tting of strokes to the input while jointly optimizing control points and brush texture.We further demonstrate promising results on video by incorporating temporal tracking and adaptive keyframe introduction. Our evaluation shows that we achieve state-of-the-art results for image reconstruction, while being fast and producing strokes of high quality. More importantly, the tting is highly compatible with user input, through the correction of the splines and the selection of optimization parameters.
Comment · 2 months ago
> You can't have 24hour sunlight AND low altitude, low-latency orbits.
The key issue is "low-latency". The DCs in space are about getting ~10x the energy/day compared to ground DCs. Moving from 300km to 1400km doesn't increase latency that much since those DCs aren't meant to have realtime connections to users. I think the criticism is confusing something like Starlink which gets low latency by having a ton of routers that are only ~300km away from the user with these DCs which will by design have to transmit from polar orbit to wherever the ground control is. 300km or 1400km won't be a big deal compared to the tramission time from polar orbit to the urban or tech hubs.
Gemini's math about median RTT from these DCs to a random spot on Earth suggests the altitude change only adds 15ms of latency:
The median surface distance between any two arbitrary points on Earth is 90deg along the globe, or 10,008km. Scaled up to the respective orbital radii (6,671 km vs. 7,971km), the median one-way vacuum ISL path length is ~10,480km at 300 km altitude and ~12,520km at 1,600 km altitude...When factoring in realistic ground-to-space uplinks/downlinks, atmospheric delay, and optical switching overhead, the median global RTT increases to approximately 85-100ms for a 300 km constellation compared to 100-115ms for a 1,600 km constellation.Comment · 3 months ago
I'm not going to share because https://www.antipope.org/charlie/blog-static/fiction/acceler... has a Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License. But I've paid for my copy of the book so I can do what I want with it.
If you want to build your own, you can feed this DESIGN.md to your favorite clankie: https://pastebin.com/huRkGnba
Comment · 3 months ago
The protagonist who believes that information should be free and that scarcity is mostly artificial due to bad economic models. And that it would be possible to implement proper pro-human capitalism or communism if we had enough compute to solve standard CS problems like filling knapsacks or maximizing flow through a graph.
Comment · 3 months ago
> I'd quite like a tool (running entirely locally on my own hardware, with no cloud service and no copyright-thieving grifters making bank on it via subscription fees) that digests a manuscript and derives a scene-by-scene timeline, that I could then query interactively and use to plan my next round of edits. Being able to map out where and when each protagonist and minor character shows up, and see a frequency distribution heat map of names in the manuscript, would be useful.
Interestingly enough, I do this with various books I'm reading. E.g, I'm currently rereading Accelerando and had Claude generate a wiki-like timeline of key events, characters, and salient plot points. That makes it easier to jump around when I want to re-read a section and grok a plot thread that is scattered across chapters. Ironically, it also exposes inconsistencies (or "hallucinations" as some might call them) in the text because the author didn't have an AI proofread the text.
Comment · 3 months ago
Yeah. Ironically, if he used an LLM to proofread his work, it would have told him that GAN's are generative not "generalized". That LLMs are primarily built on Transformers rather than GANs. And that they're more famous for image generation rather than image recognition.