iiinova — where your digital self keeps evolving

Data Elixir @dataelixir.com · 07/07/2026

Nice opportunity for researchers and applied folks working with text data: Text as Data 2026 is accepting presentation proposals. The focus is on computational analysis of documents across politics, society, culture, social science and the humanities. One-page proposals are due Aug 1. tada2026.org

tada2026.org

Text as Data 2026

New Directions in Analyzing Text as Data (TADA) — Monday, October 5, 2026 at UC Berkeley. One-page submissions due August 1.

031

Data Elixir @dataelixir.com · 20/02/2026

"You need a vector database" is the default advice but complexity means there's new infrastructure, new formats, and data coordination overhead. What if tuning Parquet page sizes and adding lightweight footer metadata gave you vector search without any of that? blog.xiangpeng.systems/posts/vector...

blog.xiangpeng.systems

Vector search using only Parquet and DataFusion – Xiangpeng’s blog

Just tune your existing stack

020

Data Elixir @dataelixir.com · 13/02/2026

Most data bugs aren't algorithm failures, they're assumption violations upstream. Pointblank makes validation a first-class citizen in R/Python workflows instead of an afterthought scattered in assert statements.

posit.co

Fifty Releases of Pointblank: a Year of Building Data Quality Tooling - Posit

The Pointblank Python library's journey has resulted in a premier, data-agnostic toolkit for tabular data validation.

041

Data Elixir @dataelixir.com · 06/02/2026

Here's what they don't tell you about topic modeling: preprocessing determines everything. Corpus homogeneity, document length, and context windows matter more than algorithm choice. BERTopic won't fix bad inputs. No model will.

css-polytechnique.github.io

The General Inquirer in the time of LLMs: a BERTopic tutorial – CSS@IPP — Tutorials and resources

A. Morin

093

Data Elixir @dataelixir.com · 05/02/2026

Indexes speed up reads but slow writes. The real insight: indexes only help when you're returning <15-20% of rows. Beyond that, sequential scans win. Space and memory costs matter too. btree indexes often exceed table size.

dlt.github.io

Introduction to PostgreSQL Indexes

Who’s this for Basics How data is stored in disk How indexes speedup access to data Costs associated with indexes Disk Space Write operations Query planner Memory usage Types of Indexes Btree Hash…

010

Data Elixir @dataelixir.com · 05/02/2026

What if a model’s prediction was literally an eigenvalue? This post kicks off a thoughtful series exploring spectrum-based models as a middle ground between linear models and neural nets, with interpretability and robustness baked in. A fun, slightly off-the-beaten-path ML read.

alexshtf.github.io

Behold the power of the spectrum!

Eigenvalues as neurons: represent nonlinear models as the k-th eigenvalue of a learned symmetric matrix pencil. Explore monotonicity/convexity properties and train simple spectral models.

010

Data Elixir @dataelixir.com · 04/02/2026

Data To Art is a new curated gallery transforming datasets into visual storytelling. Latest addition: Alisa Singer's Environmental Graphiti turns climate science into vibrant data-driven art. The line between science and abstraction is thinner than you think.

data-to-art.com

Data to Art

A curated gallery celebrating data visualization as art. Discover innovative artworks from international artists who transform data into emotionally compelling visual experiences.

020

Data Elixir @dataelixir.com · 10/01/2026

ClickHouse solved Advent of Code 2025 puzzles using single ClickHouse queries. No UDFs, no temp tables, no preprocessing. Just pure SQL doing things SQL probably shouldn't do. Impressive and slightly cursed in the best way.

clickhouse.com

Solving the "Impossible" in ClickHouse: Advent of Code 2025

At ClickHouse, we don't like the word "impossible." We believe that with the right tools, everything is a data problem. To prove it, we decided to complete the 2025 Advent of Code unconventionally:…

000

Data Elixir @dataelixir.com · 09/01/2026

If you're wondering where database tech is really headed in 2026, Andy Pavlo's annual review cuts through the noise. Sharding, Postgres evolution, and hot takes you won't find in vendor blogs.

cs.cmu.edu

Databases in 2025: A Year in Review

The world tried to kill Andy off but he had to stay alive to to talk about what happened with databases in 2025.

010

Data Elixir @dataelixir.com · 06/01/2026

MarkItDown (Microsoft) converts a variety file types to Markdown optimized for LLMs. Handles structure, not just text. Pairs well with Kreuzberg for extraction (supports 50+ file formats!). If you're building RAG systems, these belong in your stack.

tryolabs.com

Top Python libraries of 2025

Explore our 11th annual Top Python Libraries roundup, featuring two curated Top 10 lists for General Use and AI / ML / Data tools that matter today.

030

Data Elixir @dataelixir.com · 03/01/2026

jax-js compiles NumPy-style array code into WebAssembly and WebGPU kernels that run entirely client-side. No server, no dependencies, just JAX's programming model in your browser. This changes what's possible for interactive ML demos.

ss.ekzhang.com

jax-js: an ML library for the web

JAX in pure JavaScript, as a flexible machine learning library and compiler.

000

Data Elixir @dataelixir.com · 02/01/2026

Jan Van Haaren's 2025 soccer analytics review is a solid reference if you're doing applied ML in sports or want to see how domain experts handle sequential data. Covers spatio-temporal models, graph methods, Bayesian forecasting, and tracking data metrics. janvanhaaren.be/posts/soccer...

janvanhaaren.be

Soccer Analytics 2025 Review – Jan Van Haaren

Collection of the soccer analytics content that I liked the most in 2025!

010

Data Elixir @dataelixir.com · 20/12/2025

ML progress isn't driven by elegant theory. It's benchmarks, leaderboards, and engineering culture. In this post, Ben Recht explores why empirical testing beats clean math in practice and why that tension defines the field.

argmin.net

Benchmark Studies

It is impossible to disentangle technical innovation from technical debt

020

Data Elixir @dataelixir.com · 19/12/2025

The paradox: show a complex graph and execs check their phones. Show a simple one and they demand endless breakdowns. The solution isn't more or less data. It's recentering discussions on the actual decision at hand. methodmatters.github.io/true-stories...

methodmatters.github.io

True Stories from the (Data) Battlefield – Part 1: Communicating About Data

A blog about data science, statistics, and data analysis with open-source software.

000

Data Elixir @dataelixir.com · 18/12/2025

Your browser can run Python (Pyodide), execute OCR on PDFs, crop videos, and call LLM APIs—all without uploading anything to a server. The localStorage + CORS pattern makes surprisingly powerful tools possible with zero backend infrastructure.

simonwillison.net

Useful patterns for building HTML tools

I’ve started using the term HTML tools to refer to HTML applications that I’ve been building which combine HTML, JavaScript, and CSS in a single file and use them to …

000

Data Elixir @dataelixir.com · 17/12/2025

Tired of charts that hide the story? Density plots + percentile intervals reveal what averages can't: full variability, extremes, historical context. Nine examples comparing Spain's temperature data show how geometry choice changes what readers understand.

dominicroye.github.io

Broken Chart: discover 9 visualization alternatives

Researcher in climate science at MBG-CSIC

010

Data Elixir @dataelixir.com · 16/12/2025

Fisher arbitrarily chose p<0.05 a century ago and we've just... kept it. The problem: calling it "arbitrary" only works if you can suggest something less arbitrary. No one has, so here we are circling p-values close to 0.05 like it means something. vilgot-huhn.github.io/mywebsite/po...

000

Data Elixir @dataelixir.com · 15/12/2025

Haskell for data science? dataHaskell adds dataframes, NSE-style column operations, and compiler optimizations that turn chained operations into single-pass computations. Immutability + strong types + functional composition might be the combo we've been missing. jcarroll.com.au/2025/12/05/h...

jcarroll.com.au

Haskell IS a Great Language for Data Science

I’ve been learning Haskell for a few years now and I am really liking a lot of the features, not least the strong typing and functional approach. I thought it was lacking some of the things I missed…

060

Data Elixir @dataelixir.com · 05/12/2025

Learning SQL is like learning a foreign language: you need to read more variations than you'll actually write. Learn disciplined canonical syntax for your own queries, but understand the messy dialects others use.

kb.databasedesignbook.com

A modern guide to SQL JOINs

There are many SQL JOINs guides and tutorials, but this one takes a different approach. We try to avoid misleading wording and imagery, and we structure the material in a different way. The goal of…

110

Data Elixir @dataelixir.com · 01/12/2025

Side projects still open more doors than traditional applications in data roles. The trick: build small, interesting things that signal skills and make your work discoverable. Especially relevant given the current job market.

presentofcoding.substack.com

Make Things, Tell People

On side projects and finding work

083

Data Elixir @dataelixir.com · 29/11/2025

Most time-to-event metrics are broken. Amazon thought customer support wait times were under 1 min. Bezos called in: 10+ minutes. Why? They only measured customers who stayed on hold long enough to be served. The ones who hung up? Never counted. www.counting-stuff.com/why-you-are-...

counting-stuff.com

Why You Are (Probably) Measuring Time Wrong: Why do we need to use Survival Analysis more

Author: Michał Chorowski [Hey everyone! A guest post from Michał this week! This newsletter is always willing to host/share data-related content, so if you've created anything you'd like to share,…

030

Data Elixir @dataelixir.com · 29/11/2025

Hot take: Your analysts doing "some basic data engineering" is killing your analytics function. The MTA hired 5 dedicated data engineers and it unlocked everything else. Stop asking data scientists to maintain pipelines. www.mta.info/article/less...

mta.info

Lessons learned in starting a central data team

Learn how the MTA succeeded in setting up a central data team and a general purpose, cloud-based platform for data analytics.

073

Data Elixir @dataelixir.com · 24/11/2025

Sometimes the best polars pattern is knowing when to exit the DataFrame. partitionby() splits data into a dict of frames, letting you process with list comprehensions. Cleaner than forcing everything through mapgroups() when further wrangling isn't needed.

emilyriederer.com

Python Rgonomics: User-defined functions in polars | Emily Riederer

Polars provides a consistent API for conducting transformations against a DataFrame. But what do you do when you need to apply a user-defined function beyond the native API? This post surveys the…

010

Data Elixir @dataelixir.com · 23/11/2025

Most devs treat AI coding agents like infinite context machines. Reality: a 200k token window fills fast. The /compact feature is a trap. Better approach: /clear + document state in markdown, then resume. Treat context like disk space. You need a cleanup strategy.

blog.sshh.io

How I Use Every Claude Code Feature

A brain dump of all the ways I've been using Claude Code.

110

Data Elixir @dataelixir.com · 22/11/2025

Most modern dimensionality reduction (t-SNE, UMAP, Isomap) shares a pattern: represent data as a graph capturing local similarity, then embed to preserve that structure. It's graphs all the way down.

alechelbling.com

A Visual Introduction to Dimensionality Reduction with Isomap

"To deal with hyper-planes in a 14-dimensional space, visualize a 3D space and say 'fourteen' to yourself very loudly. Everyone does it." - Geoffrey Hinton

010

Reposted by Data Elixir

Maria Antoniak @mariaa.bsky.social · 14/11/2025

I curated some readings for class on "data tensions" and the list felt worth sharing. Come on a tour of datasets, books, the web, and AI with me... We'll start with this piece on the Google Books project: the hopes, dreams, disasters, and aftermath of building a public library on the internet. 1/n

theatlantic.com

Torching the Modern-Day Library of Alexandria

“Somewhere at Google there is a database containing 25 million books and nobody is allowed to read them.”

612042

Data Elixir @dataelixir.com · 14/11/2025

Everyone's rushing to pgvector for "simple" vector search in Postgres. This reality check shows what actually happens at scale: indexing nightmares and performance walls. Simple isn't always sustainable in production.

alex-jacobs.com

The Case Against pgvector | Alex Jacobs

What happens when you try to run pgvector in production and discover all the things the blog posts conveniently forgot to mention

000

Data Elixir @dataelixir.com · 13/11/2025

When healthcare becomes algorithmic, what gets optimized out? This Guardian essay asks the hard question about AI spreading through diagnostics and therapy: are we trading care quality for efficiency without realizing the cost?

theguardian.com

What we lose when we surrender care to algorithms | Eric Reinhart

A dangerous faith in AI is sweeping American healthcare – with consequences for the basis of society itself

010

Data Elixir @dataelixir.com · 08/11/2025

Thinking Machines Lab solved a problem everyone accepted as unsolvable: LLM nondeterminism at temperature 0. Same prompt, same model, 1000 runs → 80 different outputs. With batch-invariant kernels? Bitwise identical every time. Open sourced. www.distributedthoughts.org/will-i-make-...

distributedthoughts.org

Will I Make It To The Restaurant Before The Soup Dumplings Get Cold? (And Other Problems In Machine Learning)

I'm chronically late. Not because I want to be rude - I feel terrible about it every single time - but because I'm catastrophically bad at predicting how long it takes to get anywhere. Turns out…

010

Data Elixir @dataelixir.com · 06/11/2025

Most marketplaces have SKUs. Etsy has 100M+ unique items with no standard attributes. How do you build filters when one listing is a "porcelain sculpture that looks like a t-shirt" and dimensions live in random photo text? www.etsy.com/codeascraft/...

etsy.com

020

Data Elixir @dataelixir.com · 31/10/2025

GeoUtil converts between GeoJSON, TopoJSON, Shapefile, KML, WKT, and CSV without touching a server. TopoJSON compression alone cuts file sizes 80%+ while preserving topology. All free, all browser-based. geoutil.com

geoutil.com

GeoUtil — Free Online Map & Geography Tools

All-in-one online geography toolkit. Measure distance & area, convert GeoJSON, TopoJSON, JSON, merge or minify files, and more — fast, free, and browser-based.

000

Data Elixir @dataelixir.com · 30/10/2025

Debugging constraint problems is backwards: remove constraints until something works, then figure out what broke. No stack traces, just an "unsatisfiable." Forces you to think differently about what you're actually asking the system to solve. www.righto.com/2025/10/solv...

righto.com

Solving the NYTimes Pips puzzle with a constraint solver

The New York Times recently introduced a new daily puzzle called Pips . You place a set of dominoes on a grid, satisfying various condition...

000

Data Elixir @dataelixir.com · 29/10/2025

Most interpretable models sacrifice accuracy. Most accurate models are black boxes. TRUST breaks this trade-off by combining parametric and non-parametric approaches and offers full prediction explanations without losing performance. Published at PRICAI 2025.

pypi.org

trust-free

Transparent, Robust & Ultra-Sparse Trees (TRUST™) - Free Version

000

Data Elixir @dataelixir.com · 29/10/2025

Data products shouldn't live forever by default. Netflix treats outdated metrics like deprecated software and actively sunset them. The cost of maintaining zombie datasets? Lost trust and accumulated technical debt that blocks innovation.

netflixtechblog.medium.com

Data as a Product: Applying a Product Mindset to Data at Netflix

Introduction: What if we treated data with the same care and intentionality as a consumer-facing product? Adopting a “data as a product”…

000

Reposted by Data Elixir

Cadence @cadence-gis.bsky.social · 24/10/2025

What will you be using this #30DayMapChallenge? Cadence is offering its users £2,500 worth of prizes this #30daymapchallenge. And every user (new or existing) gets a free Professional upgrade for November! Learn more: cadence.cityscience.com/blog/30-day-...

cadence.cityscience.com

30 Day Map Challenge

The 30 Day Map Challenge with Cadence – November 2025 This November, Cadence is proud to support the 30 Day Map Challenge – a global celebration of maps, creativity and storytelling. Whether …

011

Data Elixir @dataelixir.com · 24/10/2025

Why does neural network training almost never fail? Pure combinatorics. A 6K parameter network contains 10^1089 possible sparse subnetworks. That's 10^900 solutions per atom in the universe. We're not smart, we're just brute forcing.

embedding-space.github.io

Sparse Networks and Lottery Winners

Embedding Space is a blog about machine learning and artificial intelligence.

040

Data Elixir @dataelixir.com · 23/10/2025

The best part about #30DayMapChallenge is that it's tool-agnostic. Whether you're using QGIS, Python's geopandas, R's rayshader, or even Blender for 3D visualizations, the focus is on creativity over tech stack. No programming required.

30daymapchallenge.com

30DayMapChallenge

Daily mapping challenge happening every November!

110

Data Elixir @dataelixir.com · 23/10/2025

Stop fitting separate binomial models to compositional data! Your predictions that 130% of respondents chose option A reveal fundamental model misspecification. Dirichlet regression with Gaussian processes respects constraints.

ecogambler.netlify.app

Compositional modeling of plant communities with Dirichlet regression | GAMbler

Compositional data appears everywhere in scientific research, yet many analysts fall back on problematic approaches that ignore fundamental mathematical constraints. I demonstrate how Dirichlet…

010

Data Elixir @dataelixir.com · 17/10/2025

AI systems are now generating, testing, and validating their own hypotheses. DeepMind's Co-Scientist and Stanford's Virtual Lab represent something new: AI as actual scientific collaborator, not just a fancy search engine.

stateof.ai

State of AI Report 2025

The State of AI Report analyses the most interesting developments in AI. Read and download here.

000

Data Elixir @dataelixir.com · 17/10/2025

Black and white, hand-drawn data viz countering visual noise at one of NYC's busiest hubs. Smart choice. Sometimes the most effective data art isn't about adding more complexity but finding clarity in the chaos through intentional restraint.

pentagram.com

‘A Data Love Letter to the Subway’

A data-driven animation for Fulton Center commissioned by MTA Arts & Design for its 40th anniversary.

020

Data Elixir @dataelixir.com · 15/10/2025

The metrics that matter most are the hardest to measure. Bookings take weeks to materialize, but clicks happen instantly. The trap: optimizing for clicks can actually decrease bookings. Correlation isn't causation, especially in A/B tests.

booking.ai

How to estimate correlation between metrics from past A/B tests

Authors: Miha Gazvoda, Christina Katsimerou

000

Data Elixir @dataelixir.com · 15/10/2025

Psychology's dirty secret: "data available upon request" usually means data not available at all. Researchers found systematic patterns in why data disappears over time. Open-washing is real and it's undermining reproducibility.

open.lnu.se

LnuOpen | Meta-Psychology

Many journals now require data sharing and require articles to include a Data Availability Statement. However, several studies over the past two decades have shown that promissory notes about data…

011

Reposted by Data Elixir

Carl Zimmer @carlzimmer.com · 09/10/2025

Today my @nytimes.com colleagues and I are launching a new series called Lost Science. We interview US scientists who can no longer discover something new about our world, thanks to this year‘s cuts. Here is my first interview with a scientist who studied bees and fires. Gift link: nyti.ms/3IWXbiE

nyti.ms

14147041812

Data Elixir @dataelixir.com · 13/10/2025

Parquet is showing its age. CMU researchers built F3 - a columnar format that embeds WebAssembly decoders directly in files. Universal compatibility without the usual compatibility hell. Smart approach for modern ML workloads. db.cs.cmu.edu/papers/2025/...

db.cs.cmu.edu

021

Data Elixir @dataelixir.com · 10/10/2025

Mathematicians feared nuclear winter would freeze Earth, but it turns out CO2 might do it instead. The math behind climate tipping points is fascinating and terrifyingly unpredictable. Sometimes the thing you're not watching is the real threat.

quantamagazine.org

The Math of Climate Change Tipping Points | Quanta Magazine

Tipping points in our climate predictions are both wildly dramatic and wildly uncertain. Can mathematicians make them useful?

000

Reposted by Data Elixir

Crystal Lewis @cghlewis.bsky.social · 08/10/2025

Data dictionary template: osf.io/ynqcu Project summary template: osf.io/q6g8d Dataset level README template: osf.io/tk4cb

0274

Data Elixir @dataelixir.com · 08/10/2025

Your ggplot2 charts work fine, but are they memorable? Real color engineering: brightness first (strongest differentiator), then hue, finally saturation. Most people get this backwards and wonder why their viz falls flat. www.chartography.net/p/color-engi...

chartography.net

Color Engineering

The tool that snaps mercurial design into mechanical focus.

0189

Data Elixir @dataelixir.com · 30/09/2025

bsky.app/profile/libb...

000