frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

LLM Embeddings Explained: A Visual and Intuitive Guide

https://huggingface.co/spaces/hesamation/primer-llm-embedding
162•eric-burel•6h ago•24 comments

Debian switches to 64-bit time for everything

https://www.theregister.com/2025/07/25/y2k38_bug_debian/
128•pseudolus•2h ago•61 comments

Show HN: I made a tool to generate photomosaics with your pictures

https://pictiler.com
29•jakemanger•1h ago•9 comments

Enough AI copilots, we need AI HUDs

https://www.geoffreylitt.com/2025/07/27/enough-ai-copilots-we-need-ai-huds
556•walterbell•14h ago•184 comments

SIMD within a register: How I doubled hash table lookup performance

https://maltsev.space/blog/012-simd-within-a-register-how-i-doubled-hash-table-lookup-performance
107•axeluser•7h ago•12 comments

Trying to play an isomorphic piano (2022) [video]

https://www.youtube.com/watch?v=j4itL174C-4
21•surprisetalk•3d ago•11 comments

Performance and telemetry analysis of Trae IDE, ByteDance's VSCode fork

https://github.com/segmentationf4u1t/trae_telemetry_research
880•segfault22•19h ago•312 comments

VPN use surges in UK as new online safety rules kick in

https://www.ft.com/content/356674b0-9f1d-4f95-b1d5-f27570379a9b
176•mmarian•10h ago•174 comments

Dumb Pipe

https://www.dumbpipe.dev/
817•udev4096•22h ago•197 comments

What would an efficient and trustworthy meeting culture look like?

https://abitmighty.com/posts/the-ultimate-meeting-culture
61•todsacerdoti•5h ago•48 comments

Blender: Beyond Mouse and Keyboard

https://code.blender.org/2025/07/beyond-mouse-keyboard/
196•dagmx•3d ago•55 comments

How I fixed my blog's performance issues by writing a new Jekyll plugin

https://arclight.run/how-i-fixed-my-blogs-performance-issues-by-writing-a-new-jekyll-plugin-jekyll-skyhook/
56•arclight_•3d ago•20 comments

I hacked my washing machine

https://nexy.blog/2025/07/27/how-i-hacked-my-washing-machine/
279•JadedBlueEyes•16h ago•125 comments

Software Development at 800 Words per Minute

https://neurrone.com/posts/software-development-at-800-wpm/
110•ClawsOnPaws•3d ago•39 comments

Terminal app can now run full graphical Linux apps in the latest Android Canary

https://www.androidauthority.com/linux-terminal-graphical-apps-3580905/
17•thunderbong•3d ago•8 comments

Samsung Removes Bootloader Unlocking with One UI 8

https://sammyguru.com/breaking-samsung-removes-bootloader-unlocking-with-one-ui-8/
77•1una•11h ago•56 comments

Making Postgres slower

https://byteofdev.com/posts/making-postgres-slow/
298•AsyncBanana•15h ago•32 comments

200k Flemish drivers can turn traffic lights green

https://www.vrt.be/vrtnws/en/2025/07/24/200-000-flemish-drivers-can-turn-traffic-lights-green-but-waze/
25•svenfaw•3d ago•53 comments

Claude Code Router

https://github.com/musistudio/claude-code-router
99•y1n0•12h ago•22 comments

Ask HN: What are you working on? (July 2025)

202•david927•19h ago•616 comments

Why I write recursive descent parsers, despite their issues (2020)

https://utcc.utoronto.ca/~cks/space/blog/programming/WhyRDParsersForMe
97•blobcode•4d ago•45 comments

ZUSE – The Modern IRC Chat for the Terminal Made in Go/Bubbletea

https://github.com/babycommando/zuse
84•babycommando•14h ago•38 comments

Multiplex: Command-Line Process Mutliplexer

https://github.com/sebastien/multiplex
26•todsacerdoti•7h ago•6 comments

Solid protocol restores digital agency

https://www.schneier.com/blog/archives/2025/07/how-solid-protocol-restores-digital-agency.html
53•speckx•3d ago•30 comments

The JJ VCS workshop: A zero-to-hero speedrun

https://github.com/jkoppel/jj-workshop
146•todsacerdoti•1d ago•13 comments

Big agriculture mislead the public about the benefits of biofuels

https://lithub.com/how-big-agriculture-mislead-the-public-about-the-benefits-of-biofuels/
184•littlexsparkee•11h ago•166 comments

EU age verification app to ban any Android system not licensed by Google

https://www.reddit.com/r/degoogle/s/YxmPgFes8a
865•cft•14h ago•500 comments

Mobile BESS Powers Remote Heavy Equipment

https://spectrum.ieee.org/mobile-bess
9•defrost•3d ago•4 comments

Formal specs as sets of behaviors

https://surfingcomplexity.blog/2025/07/26/formal-specs-as-sets-of-behaviors/
39•gm678•1d ago•6 comments

How big can I print my image?

https://maurycyz.com/misc/printing/
34•LorenDB•3d ago•6 comments
Open in hackernews

Digitising CDs (a.k.a. using your phone as an image scanner)

https://www.hadess.net/2025/07/digitising-cds-aka-using-your-phone-as.html
22•JNRowe•8h ago

Comments

dvh•4h ago
I recently used ffmpeg to undo perspective from the image you just provide 4 corners coordinates and it produced straightened image:

    ffmpeg -i input.jpg -vf "perspective=x0=784:y0=396:x1=2396:y1=397:x2=684:y2=2479:x3=2610:y3=2467" output.jpg
voidUpdate•4h ago
Is there anything FFMPEG cant do?
cadamsdotcom•58m ago
I like the irony that LLMs, full of hidden capability, are so helpful for exposing the hidden capabilities of ffmpeg.
WesolyKubeczek•24m ago
Hmm, I don’t even know where ImageMagick ends and ffmpeg begins anymore.
sandbach•4h ago
I thought this was going to be about recovering data from a CD from just a photo of the shiny side. Could that be possible?
schoen•4h ago
I don't think so.

The total data on a CD-ROM including the error correcting redundancy exceeds 800 MB, or 6.4 Gb.

If you could imagine getting 1 bit from the optical disc per pixel (which is way too optimistic physically), you would need a 6 gigapixel camera focused super-precisely at the disc surface.

Looking at the problem from a different angle, Wikipedia says the features that store the data on the disc surface are about 800 nm (or about a micrometer) long. So to photograph them, you'd want to have pixels ideally smaller than a micrometer on each side. It's easy to check that an ordinary camera isn't achieving that kind of resolution without adding on external magnifying equipment.

jackweirdy•4h ago
Does it have to be one photo? If you reproduced a spinning drive but with the camera positioned to see half of the spinning disc, I wonder if it could capture the "stream" of pixels in one arc of the spinning disc
schoen•3h ago
You would still need some significant magnification. And there might also be a measurement latency issue if the disc is in motion (the camera CCD might not be fast enough to capture the image before it rotates away).

The optics of a CD-ROM drive are optimized for something pretty different than the optics of a camera. But if you made enough tweaks and adaptations, sure, the data is ultimately there and can be captured by a different kind of sensor than the one it was designed for. It would be a cool project.

I'm mostly just pointing out that adapting your camera to successfully capture billions of sub-micrometer features isn't that trivial.

makeitdouble•4h ago
CD data pitch would be 700nm, and it seems that camera sensors have a pixel pitch at the same order of magnitude.

Ignoring the lens resolution you'd need for a near 1 to 1 rendering at that size, any hand movement or misalignment would also be catastrophic.

That sounds like a crazy dream until we get to sensor in the peta pixel range ?

https://www.researchgate.net/figure/CD-DVD-Blu-ray-disc-data...

https://letmaik.github.io/pixelpitch/

easyThrowaway•3h ago
I can't find the link, but there was a Stackoverflow question about ripping a cdrom by using a flatbed scanner, the conclusion was that the scanner would've required 2-3X the DPI currently available on a commercial device to correctly parse the gaps, given that a cdrom laser size is roughly 800nm.

On the other hand I'm still puzzled why there are no homebrew projects for a dumb dumping device that simply reads all the data on a cdrom, error correction, subchannels completely raw. There are a lot of CDs with a very weird data structure (Console Games, Early Copy Protected pc games, AKAI sample discs, some Hybrid Macintosh discs, in other words anything not using the iso9660 standard) which are at risk of Disc Rot[1], and simply storing them as Iso or Bin/Cue files (including proprietary variants like Alcohol 120%/Daemon Tools mdf files) is basically useless both for archival and real world usage purposes.

[1]https://en.wikipedia.org/wiki/Disc_rot

ralferoo•2m ago
> On the other hand I'm still puzzled why there are no homebrew projects for a dumb dumping device that simply reads all the data on a cdrom, error correction, subchannels completely raw.

I think you'd have to write a custom firmware for one specific drive for that. That information is simply not exposed outside the device itself on standard drives, which is what makes it effective for copy protection.

Igrom•1h ago
There is photo scanning/camera software out there on phones that detects edges of documents/regions, then crops and deskews the image. My Xiaomi Redmi Turbo 3's camera app has a "document" mode that does that. I also know of other software (for example, Paperless Mobile — though it is not its primary feature) that does that.

I think this would be an improvement to the OP's process: Print a thick black rectangle sized such that the contour's inner edge is slightly larger than the CD. Use the phone to take a picture and deskew it - the scanner should "catch" on the inner contour. Repeat with all other CDs. Finally, load the images onto the computer and run batch processing on them using your raster image editor to trim whitespace. This way, you'd keep manual labour to a minimum.