Hacker News

made a faster way to type with one finger (STT aside), uses a DTW algo to compute and compare paths. engine written in Rust and compiled to WASM and FFI for mac.

Comments URL: https://news.ycombinator.com/item?id=46941612

Points: 2

# Comments: 0

I built ShapeGuard because shape errors in numerical code are uniquely painful. They're silent (wrong shapes often produce garbage instead of crashing), late (errors surface deep in XLA, not where the bug is), and cryptic (shapes (3,4) and (5,3) not aligned — but why should they match?). ShapeGuard lets you declare shape contracts on functions using symbolic dimensions: from shapeguard import Dim, expects, ensures n, m, k = Dim("n"), Dim("m"), Dim("k") @expects(a=(n, m), b=(m, k)) @ensures(result=(n, k)) def matmul(a, b): return a @ b When shapes don't match, the error traces bindings back to their source: ShapeGuardError: function: matmul argument: b expected: (m, k) actual: (5, 7) reason: dimension 'm' bound to 4 from a.shape[1], but got 5 from b.shape[0] bindings: {n=3 (from a[0]), m=4 (from a[1])} The key idea is unification — the same Dim object used across arguments must resolve to the same integer. ShapeGuard tracks where each binding came from, so conflicts pinpoint the exact source. What it does: - @expects / @ensures / @contract — input and output shape validation - Symbolic Dim with cross-argument unification - Batch() dims and ... ellipsis for ML patterns - broadcast_shape() and explain_broadcast() for debugging broadcasting - Configurable JIT modes (check / warn / skip) for JAX - ML helpers: pre-defined dims (B, T, C, H, W, D), attention_shapes(), conv_output_shape() What it doesn't do: - No dtype checking (jaxtyping does this well) - No named-tensor wrapper (Haliax's approach) - Not a replacement for static type checking It's zero-dependency, drop-in (works with existing code — just add decorators), and the motivation came from analyzing 40 real JAX GitHub issues where users hit cryptic shape errors. PyPI: pip install jax-shapeguard Would love feedback on the API design, error message format, and whether this would actually help your workflow. What shape debugging pain points am I missing?

Comments URL: https://news.ycombinator.com/item?id=46941604

Points: 1

# Comments: 0

Article URL: https://harshanu.space/en/tech/ccc-vs-gcc/

Comments URL: https://news.ycombinator.com/item?id=46941603

Points: 2

# Comments: 0

Article URL: https://www.mayoclinic.org/healthy-lifestyle/nutrition-and-healthy-eating/in-depth/caffeine/art-20049372

Comments URL: https://news.ycombinator.com/item?id=46941537

Points: 1

# Comments: 0

Article URL: https://medicalxpress.com/news/2026-02-brain-mechanism-intuition.html

Comments URL: https://news.ycombinator.com/item?id=46941536

Points: 2

# Comments: 0

Article URL: https://fastretro.app/blog/how-we-improved-rails-response-times-by-87-percent

Comments URL: https://news.ycombinator.com/item?id=46941530

Points: 1

# Comments: 0

Article URL: https://www.reuters.com/world/europe/paris-prosecutors-cybercrime-unit-searches-french-office-musks-x-2026-02-03/

Comments URL: https://news.ycombinator.com/item?id=46941525

Points: 2

# Comments: 0

Article URL: https://cipscorps.io/#

Comments URL: https://news.ycombinator.com/item?id=46941508

Points: 1

# Comments: 1

Article URL: https://www.reuters.com/science/musk-says-spacex-prioritise-building-self-growing-city-moon-2026-02-08/

Comments URL: https://news.ycombinator.com/item?id=46941505

Points: 2

# Comments: 0

Article URL: https://www.bloomberg.com/news/audio/2026-02-03/odd-lots-the-us-and-chinese-internets-similarities-podcast

Comments URL: https://news.ycombinator.com/item?id=46941494

Points: 1

# Comments: 0

Reality Asset Existence Manifesto (RAEM) – v0.1

Author: Janus Pater

Objective: To establish an engineering-based asset existence framework, replacing credit, discounting, and valuation estimates with physical executable capability, defining whether an asset truly exists, can generate value, and can be traded.

Chapter 1: Hard Criteria for Asset Existence

For any asset to be recognized as “real” by the system, it must satisfy four hard conditions:

1. Executability

The asset must be able to output real-world function or value.

Function types include:

Energy (kW, MJ)

Computing power (TFLOPS, GPU-hours)

Space / Capacity (m², tons)

Services (transportation, repair, manufacturing, etc.)

Determination formula:

if Output(t) = 0 → Asset Invalid

2. Measurable Lifecycle

The asset’s lifespan must be quantifiable:

Design lifespan

Consumed lifespan

Remaining lifespan R(t)

Cannot rely on promises, insurance, or future compensation.

3. Observable State

Asset state S(t) must be real-time or periodically observable.

Data sources:

Sensors

Logs

Manual verification protocols

Update frequency ≥ transaction frequency

4. Deliverability

Usage rights must be transferable

Independent of asset owner credit

No third-party guarantees required

Chapter 2: Core Asset Data Structure

Each asset corresponds to a state vector S(t):

S(t) = { R(t): Remaining lifespan O(t): Current output capability M(t): Maintenance cost per time unit F(t): Failure probability function E(t): External environment dependency factor }

Time is consumed, not discounted

Value is based solely on current state + remaining lifespan + output capability

Chapter 3: Value Calculation (V(t))

(

) =

(

) ×

(

) − ∫

(

)

−

(

(

) ) V(t)=O(t)×R(t)−∫M(t)dt−Risk(F(t))

Explanation:

Maximum potential output = O(t) × R(t)

M(t) = Maintenance and operational cost

Risk(F(t)) = Deduction due to failure probability

Future promises, discounting, or credit are excluded

Chapter 4: Trading Rules Trading Objects

Only remaining life usage rights ΔR are traded

Ownership, promises, or future income are not traded

Usage Contract Structure Contract { Asset_ID Usage_Type ΔR State_Snapshot S(t₀) Settlement_Rule }

Settlement Principles

Instant delivery

State-driven dynamic pricing

Expired or failed assets → contract value = 0

No leverage, no collateral, no reliance on credit

Chapter 5: Minimal Executable Architecture (MVP)

Asset State Acquisition Module

Updates S(t)

Ensures data is real and timely

Lifecycle Engine

Input: S(t)

Output: R(t), V(t)

Usage Rights Market Matching Module

ΔR ↔ Instant demand

Real-time transaction settlement

Clearing Module

Life consumption → automatically deducted

Failed assets → contract cleared

Chapter 6: Engineering Examples Off-Grid Compute Unit

O(t) = available compute power in TFLOPS

R(t) = remaining stable operational hours

M(t) = maintenance + energy cost

Trade: 100 hours compute usage, instant delivery, independent of future revenue

Computable Building

O(t) = usable space / compute / energy output

R(t) = remaining operational lifespan

Trade: usage rights transfer + instant settlement

Chapter 7: Core Principles

Assets must generate value immediately

Value derives only from current state and remaining lifespan

Time is consumed, not discounted

Credit and promises are invalid

Failure → cleared; no bailout from the system

Core concept: Real existence ≡ Executable ≡ Tradable

Chapter 8: Practical Significance

Prevents “virtual asset scams”

Survives during macro credit contraction cycles

Transparent, real-time asset value

Simple, trustless trading model

Engineering implementation of Digital Materialization → Lifecycle-Anchored Asset execution model.

Comments URL: https://news.ycombinator.com/item?id=46941486

Points: 1

# Comments: 0

Article URL: https://www.techdirt.com/2026/02/06/the-cia-erased-the-world-factbook-with-no-warning-and-told-everyone-to-stay-curious/

Comments URL: https://news.ycombinator.com/item?id=46941460

Points: 2

# Comments: 0

Article URL: https://www.youtube.com/watch?v=aCN9iCXNJqQ

Comments URL: https://news.ycombinator.com/item?id=46941454

Points: 1

# Comments: 0

Article URL: https://www.youtube.com/watch?v=TiwADS600Jc

Comments URL: https://news.ycombinator.com/item?id=46941451

Points: 1

# Comments: 0

Article URL: https://www.mklogo.sh/

Comments URL: https://news.ycombinator.com/item?id=46941426

Points: 1

# Comments: 0

Applying to X26 (deadline is tomorrow).

I built this because I got rejected from W25 (AI tutor idea) and wanted to sanity check my future apps based on what empirical indicators I could find.

Here is how it works: Vector Search: Checks your idea against 5,000+ past YC companies (successes and failures). Rubric Analysis: Evaluates the startup mechanics based on criteria extracted from YC Startup School and Paul Graham's essays.

Tested it on a few friends building stuff. Correctly identified a VS Code docs extension as a nice-to-have / low urgency (Fail). Gave a pass to a campus sublet marketplace, but correctly flagged the chicken and egg supply risk.

Comments URL: https://news.ycombinator.com/item?id=46941413

Points: 1

# Comments: 0

Article URL: https://www.nytimes.com/2026/02/07/us/politics/doj-prosecutors-recruiting-trump.html

Comments URL: https://news.ycombinator.com/item?id=46941405

Points: 2

# Comments: 1

Article URL: https://publicdomainreview.org/collection/japanese-snowflake-book/

Comments URL: https://news.ycombinator.com/item?id=46941383

Points: 2

# Comments: 0

Article URL: https://smallcultfollowing.com/babysteps/blog/2026/02/08/fun-with-dada/

Comments URL: https://news.ycombinator.com/item?id=46941366

Points: 1

# Comments: 0

WriteMore helps writers write more. Built by writers, it turns daily prompts into a shared creative practice—write, post, and explore what others are creating every day.

Comments URL: https://news.ycombinator.com/item?id=46941358

Points: 1

# Comments: 0