DEV Community: Yoana Popova

The Numbers Behind World Cup 2026: Carbon, Prize Money & Ticket Prices

Yoana Popova — Wed, 10 Jun 2026 10:06:36 +0000

This article was originally published on Datopian's blog.

The 2026 World Cup is on track to be the most carbon-intensive sporting event ever staged. It is also the most commercially inflated — record prize money, record ticket prices, and a formal legal subpoena over affordability filed just weeks before kickoff. Three datasets we built and published on DataHub tell the story in full.

The Most Polluting Tournament Ever Projected

The 2026 World Cup carbon footprint is projected at 9.0 million tonnes of CO₂e — roughly equivalent to the annual emissions of 2 million cars. Against a recalculated 2010–2022 average of 4.71 Mt on the same accounting basis, that is a +92% increase. Nearly all of it — 86%, or 7.72 Mt — comes from air travel. A three-nation tournament spread across the United States, Canada, and Mexico runs on flights.

This number requires context. Earlier per-edition figures (South Africa 2.75 Mt, Brazil 2.27 Mt, Russia 2.16 Mt, Qatar 3.63 Mt) are FIFA self-reports on a narrower accounting boundary that systematically undercounts aviation. A direct comparison to those figures would be misleading. The honest comparison is the recalculated 4.71 Mt average on a consistent fuller scope — which makes the +92% increase clear and defensible.

Qatar 2022 drew significant criticism for its carbon accounting. 2026 is projected to exceed it substantially once aviation is counted consistently.

📊 Dataset: worldcup-carbon-footprint on DataHub · 📖 Read the full story

Record Prize Money — and a Gender Gap That Is Closing

The 2026 prize money pool totals $655M — the largest in World Cup history. Winners take home $50M. Teams eliminated in the group stage still earn $9M for qualifying and showing up.

Zoom out across editions and two trends are stark. The winner's prize has climbed at every edition: $30M (2010), $35M (2014), $38M (2018), $42M (2022), $50M (2026). Meanwhile, the gender gap remains significant: the 2022 men's pool of $440M was four times the $110M awarded at the 2023 Women's World Cup. That gap was infinite until 2007 — women's prize money was simply zero before then. FIFA has committed to equal prize money by 2027.

A note on data integrity: the 32 individual team payouts for 2022 in our dataset sum exactly to the published $440M pool. Pre-2014 men's figures use a broader "total contribution" scope rather than pure prize money — flagged clearly in the dataset so comparisons are not made across incompatible definitions.

📊 Dataset: worldcup-prize-money on DataHub · 📖 Read the full story

A $10,990 Ticket, a Subpoena, and the Widest Affordability Gap in Tournament History

On May 27, 2026, the attorneys general of New York and New Jersey subpoenaed FIFA over 2026 World Cup ticket pricing, citing price gouging and consumer protection concerns. It is the first time a major tournament host's law enforcement has taken formal legal action over World Cup ticket prices. Our data shows exactly why.

The most expensive face-value ticket to the final is $10,990 — a 6.8× increase on the $1,607 top price in 2022. Resale prices are running near $33,000 for the final. The cheapest entry point is $60 for a Supporter Entry tier. The spread from floor to ceiling is one of the widest in the history of major sporting events. At $10,990, a single final ticket represents more than three months of median US household income.

The affordability gap between the cheapest and most expensive tickets has widened at every edition since 2010. 2026 is not an anomaly — it is the continuation of a trend, now large enough to attract prosecutors.

Two data-quality notes: all figures are nominal, not inflation-adjusted. We excluded a widely-cited 2014 figure that appears to reflect resale rather than face value — flagged explicitly rather than silently dropped. Even adjusted for inflation, the 2026 jump holds.

📊 Dataset: worldcup-ticket-prices on DataHub · 📖 Read the full story

What the Data Says

Taken together, these three datasets tell a consistent story: the World Cup is growing in every direction except affordability and environmental accountability. Bigger tournament, more matches, more money — and the costs, financial and ecological, are increasingly concentrated on fans and the atmosphere rather than distributed across the institutions that profit most.

Explore the Data

All three datasets are live, open, and downloadable as CSV with full provenance and methodology notes. They join our 92-years-of-results World Cup dataset published last month.

DataHub.io is a platform for high-quality open datasets — free to use, fully documented, and built for reuse. Built by Datopian · PortalJS · Flowershow

Multi-Agent AI Is Ready. Your Workflow Infrastructure Isn't.

Yoana Popova — Wed, 03 Jun 2026 20:35:57 +0000

A conversation between Rufus Pollock, founder of Datopian, and Anuar Ustayev, CTO — from the AI Learned Today minicast.

Disclosure: The authors work at Datopian. This post is based on our own experience running multi-agent AI workflows in production — no sponsors, no affiliate links.

Two months ago, Anuar Ustayev — Datopian's CTO — adopted Gas Town as his primary tool for orchestrating multiple AI coding agents. It changed how he worked. Today, if he were starting from scratch, he might not choose it.

That's not a criticism of Gas Town. It's a signal of how fast the tooling landscape is moving — and how immature the infrastructure around multi-agent AI workflows still is.

This post is a distillation of two candid conversations about what multi-agent AI development actually looks like in practice: the workflow shifts, the unsolved problems, and the honest answers to questions most AI content avoids.

The Bottleneck Has Moved

A year ago, the constraint in software development was execution capacity. You had more work than time to implement it. AI coding assistants helped at the margins: autocomplete, boilerplate, quick fixes.

That constraint is gone.

"Previously it was about having a lot of work to do and then finding time to implement it. Now the bottleneck is around just creating the work — defining it — because you might have ideas, but you need to spec them out."
— Anuar Ustayev

With tools like Claude Code, Codex, and orchestration layers like Gas Town, a single developer can now run multiple agents in parallel — each working on a separate task, simultaneously. The implementation layer has been largely commoditised.

The new constraint is upstream: work definition. Can you write a specification clear and complete enough that an AI agent will implement the right thing?

This is not a minor adjustment. It is a fundamental restructuring of where skilled human effort needs to go.

What Gas Town Actually Does

Gas Town is an open-source tool for orchestrating multiple AI agents from a single terminal interface. Key concepts:

Rigs — project workspaces, each with its own directory, context, and agents. You can have dozens running simultaneously.

The Mayor agent — a central orchestrator that manages all rigs. You interact with it via natural language: "What's the status of project X? Dispatch an agent to pick up the next item in the backlog."

Beads — an AI-native issue tracker that lives as a local database inside each project. Unlike GitHub Issues or Linear, Beads is designed to be read and written by AI agents directly.

Anuar's reported time split after several months: 80% planning and supervising, 20% debugging. He hasn't written code himself in a long time.

🎧 We walked through this live in episode 6 of AI Learned Today — including a real terminal demo of the Mayor agent dispatching work.

The Setup: Claude Code with Subagents

Gas Town was genuinely novel when it launched in early 2025. Since then, the major AI platforms have caught up. Claude Code now supports subagent workflows natively. Codex Desktop has similar capabilities.

Anuar's current recommendation for someone starting today:

"Whatever LLM you're already subscribed to — start there. You don't need dozens of agents right away. Start with a few."

The practical workflow:

Start a planning session with Superwhisper or similar for spec writing assistance
Use plan mode to generate a structured spec from a rough idea
Break the spec into discrete tasks (Beads, GitHub Issues, or local markdown)
Dispatch agents to execute tasks in parallel
Review, merge, iterate

The Spec Quality Problem

The promise of multi-agent AI is compelling: define work once, let agents execute in parallel, review output. In practice, the quality of what agents produce is highly sensitive to the quality of the input spec.

"I find this with UI issues or others that are quite painful to get to a clear spec for the AI to solve. Right now I might spend 20 minutes trying to spec something to the AI that a human developer would understand in 30 seconds from a screenshot."
— Rufus Pollock

The current best approaches:

Interactive spec refinement — tools like Superwhisper and Claude's plan mode ask follow-up questions and force specification decisions before implementation begins.

Screenshot-to-spec — for UI issues, take a screenshot, describe the problem in one sentence, let the AI generate the requirements document.

Iterative shaping — borrow from Basecamp's Shape Up methodology: rough idea → shaped spec → ready to build. Don't hand work to an agent until it's shaped.

⚠️ The uncomfortable truth: most developers skip the shaping step because it feels like overhead. With AI agents, skipping it is expensive. A poorly specified task produces something that looks done but isn't — and unpicking it costs more than starting over.

The Fragmentation Problem Nobody Is Solving

A realistic day of AI-assisted work looks like this: research thread on ChatGPT on your phone. Continue in Claude on your laptop. Spin up Codex Desktop to implement. Ask a quick question in Gemini because Claude is rate-limited.

Each session exists in isolation. No unified history. No search across sessions.

"I want all of my chat sessions I ever did to be archived somewhere I can search — but not getting in my way. The way a good issue tracker has the past, but keeps me focused on what's right now."
— Rufus Pollock

Current workarounds:

Stick to one tool — the discipline of single-tool consistency is more valuable than marginal features of tool-switching
Checkpoint documents — markdown files that capture decisions, dead ends, and current state; portable across tools and sessions
Local-first storage — Beads' local database model is the right architecture: source of truth on your disk, not inside a vendor's SaaS

The tool that builds a unified, searchable, AI-session-aware knowledge layer across Claude, ChatGPT, Codex, and Gemini will be genuinely valuable. It does not yet exist.

The Token Cost Reality

Multi-agent workflows are token-hungry. Anuar uses Claude, OpenAI, and Gemini subscriptions concurrently — and regularly hits rate limits on all three simultaneously.

Mitigation strategies:

Task-appropriate model routing — use cheaper models (Claude Haiku, GPT-4o Mini, Gemini Flash) for mechanical tasks. Reserve expensive models for reasoning-intensive work.

Locally hosted models — Llama 4, Mistral, via Ollama or Cloudflare Workers AI at near-zero marginal cost.

Batching — background agents on non-urgent tasks can be rate-limited deliberately, spreading token consumption across time.

What This Means in Practice

The shift from implementation-focused work to spec-and-supervise work is not coming. For developers actively using these tools, it is already here.

Invest in spec quality. Ten minutes of careful shaping can prevent two hours of agent work in the wrong direction.

Build a checkpoint habit. Write distillation documents at meaningful points in every AI session — not for the AI, but for you.

Choose one orchestration tool and go deep. Self-inflicted fragmentation makes the fragmentation problem worse.

Track the shaping pipeline. Know which ideas are raw, which are shaped, and which are ready to ship to an agent.

Plan for token costs. Model routing and local hosting are budget management, not optional optimisations.

What's Next

The tools are evolving faster than the workflows. The problems described here — fragmented session history, manual model routing, immature spec tooling — will be partially addressed by the major platforms.

What won't be solved by the platforms: the cognitive discipline required to work well with AI agents at scale. The developers building these practices now will have a durable advantage.

The question worth sitting with: where is the real bottleneck in your workflow? If it's still implementation, the tools in this post will help immediately. If it's already moved to definition, the investment is in a different place entirely.

Based on two episodes of AI Learned Today, a minicast from Datopian. Hosted by Rufus Pollock and Anuar Ustayev.

Watch on YouTube:

Datopian builds open data infrastructure for governments, international organisations, and enterprises. We've been building with and for data since we created CKAN in 2006. We've built PortalJS, Datahub.io, Flowershow and many others.

Create a dataset from scratch and publish it with Datahub Cloud

Yoana Popova — Fri, 09 Aug 2024 14:30:08 +0000

In our previous article, we talked about DataDatahub Cloud: your stupidly simple and fast tool for turning your data stories or datasets on GitHub into a published, shareable site. It converts raw data and Markdown files into beautifully presented, interactive sites.

Today, we're going to tell you how to publish a dataset (multiple data files or a single data file) with DataHub Cloud.

As an example we're going to use an example dataset with an analysis of the top 1000 global universities:
https://www.kaggle.com/datasets/zahrayazdani81/univercitiesranking?resource=download

What You'll Need

GitHub account and basic knowledge of GitHub UI (especially editing and adding files)
A DataHub Cloud account.

Step 1: Create a GitHub repository with the data files and README.md file

Any DataHub Cloud site is built off of a GitHub repository. This is where you'd put all your dataset file(s) and any related markdown content that you want to publish. For the sake of simplicity, in this tutorial, we're only going to use a single README.md file. It's going to serve as a landing page for our site.

Tip
Any README.md or index.md file, either in a root of the repository or in a subfolder, will be treated as a "landing" page (of the whole site or a given folder) by the DataHub Cloud.

Go to your GitHub account and create a new repository. Note, you can check "Add a README file" checkbox. This will make GitHub automatically add an empty README.md file to our repository.

Hint
If you're new to GitHub, here are simple instructions on creating a repository: https://docs.github.com/en/repositories/creating-and-managing-repositories/quickstart-for-repositories

Now, let's continue this tutorial in our docs: https://datahub.io/docs/Create+a+dataset+from+scratch+and+publish+it+with+Datahub+Cloud

You can check out how a published dataset page looks like here: https://datahub.io/core/co2-ppm
You can check out how a published data story page looks like here:
https://datahub.io/@cheredia19/us-cities-population

Datahub Cloud: a simple and fast tool to turn your data stories or datasets on GitHub into a published, shareable site

Yoana Popova — Wed, 07 Aug 2024 13:37:21 +0000

Datahub Cloud is your stupidly simple and fast tool for turning your data stories or datasets on GitHub into a published, shareable site. It converts raw data and Markdown files into beautifully presented, interactive sites.

NOTE:
The current version of Datahub Cloud runs only off Github.

How to deploy your first site within seconds

The quickest way to start publishing with Datahub Cloud is to publish our template and customize it to fit your needs. You can do that in 5 simple steps:

Click on the selected template from the ones listed below and click "Use this template" at the top right to create a new repository

TEMPLATES:

Data-rich story template that renders like this

Dataset template that renders like this

PKM template that renders like this

NOTE
If you don't have any data files or notes, you can also start from scratch (go to Start from scratch)

Go to the app and create a new site by selecting the repository you just created (leave the "Root Dir" field empty)

Done, it is now published! Just hit the green "Visit" button at the top right to see what it looks like.

❓ What to do after you publish your first site

After publishing our template and getting a feel for how it works, you have several options:

👷‍♂️ Customize your template
📈 Add visuals
🏖️ Publish another repo of your own
🆕 Create a new repo
🖋️ Publish your Obsidian notes

You can also take the template to the next level: