What's cool is not only this new feature itself, but that I didn't built it: You can now download @taleshape.bsky.social tables directly as CSV. Last week Shaper got attention on HN and now people start to actively contributing and making it better for everyone. That's why Shaper is open source 💙

@taleshape.bsky.social now has "watermarks" for chart download to make it easy for your customers to credit you. One of the companies I work with integrated Shaper into their product. And now their customers use various dashboards regularly as part of their daily work. 1/2

Got curious why there is a sudden spike in github stars and turns out someone posted @taleshape.bsky.social on HN: news.ycombinator.com/item?id=4705...

I couldn't imagine a better fit than DuckDB for @taleshape.bsky.social: Shaper is all about using SQL to build data visualizations and sharing them. There are many databases, but DuckDB and its powerful extension system allow Shaper to query many different data sources, all from a single binary.

I sometimes forget that @taleshape.bsky.social supports dark mode:

@taleshape.bsky.social is open source for one reason: I am building the tool I wish I had. I don't want to keep user data in-house and not bother with regulations. And I never want to worry about integrating analytics too deeply into our product because the vendor might some day raise prices.

Have you seen @taleshape.bsky.social's file-based workflow? Use your favorite text editor to build dashboards all in SQL! Easier to show than explain. Here is short demo video: www.youtube.com/watch?v=HHYv... (also: my first youtube video! will keep practicing 🙏)

I am launching @taleshape.bsky.social on @peerlist.bsky.social this week. Appreciate any upvotes and comments 🙏 peerlist.io/jorin/projec... #buildinpublic

I will be speaking at FOSSASIA this year to show how to quickly visualize data entirely in SQL using Shaper and DuckDB! I had a great time as participant last year: Great energy and many interesting conversations about what everyone is building in open source - in Asia and around the globe.

Bar and Line Charts, Tables, Text, Gauges, Boxplots, Filters, PDF, CSV, Excel, Pie and Donut Charts - @taleshape.bsky.social starts to have a lot of features! I just created a demo board that shows how to use them all: (link in comment)

New Year, New Features! 🍰🍩 @taleshape.bsky.social now has Pie and Donut Charts for when your dashboard needs some more colors. And big Thank You to James for contributing this feature: github.com/taleshape-co... It's beautiful seeing the magic of open source collaboration🙏

The BI tooling market is changing rapidly: Lots of consolidations around data platforms + AI tools replacing traditional BI dashboards. So why bother building @taleshape.bsky.social? Shaper is open-source, SQL-driven, focused on embedded analytics + data sharing: 1/2 github.com/taleshape-co...

@taleshape.bsky.social is essentially a DuckDB wrapper to visualize + share data. And I am more excited than ever about @duckdb.org. I just want to do everything in SQL: Data ingest + transform, data viz + automation. The ease of using DuckDB. The extension system. AI agents for writing SQL.

Together with Shaper's new File-based Workflow we introduced the shaper-bin PyPi. Live-previewing dashboards right from your own editor is as simple as: $ pipx run shaper-bin dev

Shaper now deploys dashboards directly from SQL files and live-previews changes. $ shaper dev Build analytics dashboards: - using your favorite editor + LLM - version controlled in Git - with pull requests + code reviews - integrated into your CI/CD pipeline taleshape.com/blog/analyti...

Added "Copy Markdown" buttons to the @taleshape.bsky.social docs. More AI support coming very soon 🙏

Teaser: File-based Workflow for @taleshape.bsky.social coming soon ✨ Edit SQL file in your favorite editor and live-preview dashboard changes. Once this is ready you can track dashboards in Git, and use your own editor with all its AI features to build dashboards.

🟢🔴🟢 is rare to see on this screen

Who would have thought Box Plots are this popular! Less than 24h after release I already saw people using them for: - Visualizing medical scores - Debugging audio translation quality - Monitoring urban mobility data

My favorite part about @taleshape.bsky.social Box Plots: You can define what what to show when hovering an outlier data point to make it directly actionable.

Box Plots just shipped in @taleshape.bsky.social: Quickly visualize data distributions with a single line of SQL. Thanks to @duckdb.org macros you just type "BOXPLOT(my_value)" to calculate all statistics - including outliers. Give it a try: github.com/taleshape-co...

Is it just me or anyone else has this strange workflow? I basically spent 80% of my time in a single "todo.md" file. It's where I plan out work, write messages and so on. And now with LLM prompts I spent even less time in other apps.

Boxplots are coming to @taleshape.bsky.social! And I just created a DuckDB helper function to do the math for you. Pretty neat that that's possible as a simple macro. Makes using boxplots straight forward. And please tell me if I got the math wrong somehow 🙏

Who would have thought text-based tools would have such a comeback! Looking back it seems obvious that language is the right media to communicate knowledge. And I can't imagine letting an LLM touch anything that is not versioned in Git.

I have a backlog for @taleshape.bsky.social with 200+ feature ideas. But I won't built any of them. Unless an actual user requests it. I am more afraid of making Shaper feel messy and complex than having too few features.

Learn Bash and SQL. I have used so many programming languages and even more frameworks over the years. But basically every project used Postgres and Linux. Learning shell scripting and SQL is probably the best tech investment I made.

Are you worried about AI taking your job? This is probably the best article I read about how to deal with this and in general how to act in the face of uncertainty:

What made me switch to light mode: I want to be able to code in the airport without scaring people.

Did you know that you can download @taleshape.bsky.social dashboards as PDF? Thanks to chromedp we can generate PDFs server-side, in the same Docker image. Thanks @mvdan.cc! Means you get consistent PDF files - even on mobile or using an unusual browser.

Wish I would have started using Postgres Row-Level Security years ago! I am helping a client designing a new system and we decided to implement their permission system directly in Postgres. And I am amazed how solid it feels. 1/2

Feels strange when a handful of companies own the world's cloud infrastructure. But we can change that. Build your software on open standards. When you aren't stuck with a specific platform, providers have to compete to get you as a customer.

Every company is a data company, but most don't know where to start. I just published a new blog post on how to get started building data capabilities at your company, one step at a time. Curious to hear your thoughts! taleshape.com/blog/getting...

The most underrated way to support open source is through words. Yes, money is crucial and writing code too. But giving feedback and spreading the word is something so easy to do with a big impact down the line.

Fivetran buys DBT. Just after they bought SQLMesh. What now? Ignore what they say. Judge them by their actions: There is only one way to look at it when you buy the two leading open source data transform tools - A move to establish control over a critical layer of the data stack.

Do you prefer light or dark theme for your dashboards? Shaper supports both: taleshape.com/shaper/docs/

Why choosing the right data stack matters more than any other technology decision you make: Simple - Because data outlives code. Your product gets replaced. The whole company is sold. The data stays relevant. That's why I am always be betting on open source and open standards for my data stack.

At Taleshape we challenge how you think about data dashboards: Dashboards are not for telling stories. Dashboards are not for exploration. Dashboards get the right data to the right person at the right time. github.com/taleshape-co...

With @taleshape.bsky.social chart annotations are just another SQL query: You can load calendar events or goal lines from any data source and display them in your charts. And not only chart annotations - any text, label or dropdown in Shaper can be generated dynamically.

How am I able to build a product all on my own? Because of all the things I am not doing: No pitching to investors. No meetings. No hiring. No sales pitches. No enterpise sales processes. No marketing compains. No SaaS platform.

Open Source won. At least that’s what where my bet is. Open source is resilient. Open source is adaptable. I myself wouldn’t bet my company on a 3rd-party vendor when making fundamental technology choices. So of course my own product also has to be open source: github.com/taleshape-co...

The most underrated DuckDB extension: duckdb.org/community_ex... I am already using it to: - Fetch data from APIs - Send emails - Trigger Slack notifications Thank you for this @qxip.bsky.social 🙏

This podcast changed my understanding of ETL and how DuckDB Airport has the potential to fundamentally change what data pipelines look like. What if instead of needing a separate system to load data you query any external data source like just another SQL table? www.youtube.com/watch?v=qngR...

Today seems like a good day to emphasize that @taleshape.bsky.social is open source and runs on any hosting provider. Code-driven data dashboards and embedded analytics without vendor lock-in or platform risk.

Updated the @taleshape.bsky.social website: Before: - Overview tried to cover everything. Now: - Overview: Demo + Features (what) - About: Background (who & why) - System Architecture: Tech (how)

CREATE TEMP TABLE and getvariable() are my favorite DuckDB features. Thanks to this DuckDB is a natural fit for interactive data dashboards. Temporary tables allow you to define datasets once and reuse them across a dashboard. Variables make defining dynamic data filters intuitive.

I would probably never have built a data visualisation tool if I knew before how much math it takes. Somehow I imagined charting libraries handle it all. But here I am coming up with algorithms and magic numbers to properly format chart legends and labels no matter what data you throw at Shaper.

Why did I decide to build an embedded analytics product? I was setting up a data platform for a client. The scale and complexity of their data was trivial. But they needed to be able to embed data dashboards into their application. And the solution had to run on premise. 1/2

I am all-in on @duckdb.org. With Shaper you can build data dashboards all in SQL. Powered by DuckDB. Why DuckDB? Because of how much flexibility its simple design enables. My customers use DuckDB extensions to query Postgres, Google Sheets and even trigger webhooks. And we are just getting started.

To scale a business you have to eventually work with other people. Yes, AI can help each person to be more efficient. Someone still needs to take responsiblity for the results. That's what you pay people for. Also means functional org structures reflect who is responsible.

My problem with React is not JSX. I like it. But the way data updates is so complex to reason about and it often feels super inefficient too. Doesn't seem like the solution we should settle with to build all our UIs.