qxip

that's a true horror story 👹

AWS deleted my 10-year account and all data without warning After 10 years as an AWS customer and open-source contributor, they deleted my account and all data with zero warning. Here's how AWS's 'verification' process became a digital execution, and why you s...

Absolutely brutal AWS / cloud story. www.seuros.com/blog/aws-del...

a man in a military uniform points at the camera with the words " you 're awesome good luck today " ALT: a man in a military uniform points at the camera with the words " you 're awesome good luck today "

nooooooo

Interesting to see @openalternative.co like this post. Aren't you guys gatekeeping Open Alternative indexing and asking opensource projects to pay to get listed? How about removing that altogether and helping #opensource while doing your paid promotions. You could make a difference if you wanted to.

I really wish there was a way to make #opensource again without money and greed poisoning everything. We're just as strong as the weakest link and this washed out culture of calling everything open for marketing puts our society at great risk. Humans must fight to save opensource from extinction.

What's the story?

nah, the part you missed is this is NOT our project nor one we use for anything. its an experiment against open-core purely for the fun and the sacrilege. Enjoy the show!

hahaha that's more or less it, although to me its more putting an autocomplete in charge of autocompleting missing parts of a text someone removed - a very SPECIFIC use-case 😉 its not developing, its guessing what was removed. Since the removers left all tests and code, its not task involving trust.

heheh yeah, it would - but it cannot. This is big part of the experiment - the foollm has no write access to the "legacy" upstream code and the tests cannot be manipulated. Sometimes it tries but the road is short. Ultimately makes the prompt look childish and its fun to see it miserably fail :)

More or less. Github actions actually driving the bus and the "megautocomplete" LLM is in charge of merging upstream changes and attempting PRs for requests from issues using the test suite as safety net. The results are of course slightly supervised (and no push rights) = 💯 horror

GitHub - metrico/influxdb3-unlocked: Scalable datastore for metrics, events, and real-time analytics - InfluxDB3 Unlocked w/ Enterprise Features Scalable datastore for metrics, events, and real-time analytics - InfluxDB3 Unlocked w/ Enterprise Features - metrico/influxdb3-unlocked

As Mr. T used to say, I pity the fool - and I've been one too for trusting rug-pulling vendors... if you're curious and/or need the enterprise features, here's a link to the unlocked fork

github.com/metrico/infl...

We're putting this concept to the test by letting Cursor maintain a supervised Apache2/MIT fork of InfluxDB3-core in charge of unlocking features exclusively based on public documentation and test cases - and the results are quite tangible... 🧵

The open-core as "enterprise" gateway drug trick abusing #opensource labels for VC gains is officially about to be over. What used to take years to write now takes hours and filling the gap between "open core" and "enterprise" features using LLMs is very possible as long as the license allows it ..🧵

Scalar Functions · Issue #6 · duckdb/extension-template-rs Thanks for this amazing 🔥 firestarter @samansmink & team 🦆 The example registers a table function - is there a scalar function analogy? error[E0599]: no method named `duckdb_register_scalar_functio...

There's so much left to be desired in the DuckDB Rust extension template to match what's possible in cpp extensions that anything at all would be nice but here's an oldie example: github.com/duckdb/exten...

> Pro comes with at least $20 of included usage per month

The "at least" part is what seems to matter the most

Awesome! Are you also working on the duckdb rust extension template by any chance?

The only fix for any opensource developer is more opensource! Have you tried working on something completely different for a change and build a couple useless things just for fun? Keep on keeping on!

Trending app can turn a $3000 machine into a $1 scale

Clickhouse adds experimental support for Iceberg writing after everyone else did and their "glory shot" confirms they no longer care about performance in their opencore subproduct

How to get clicks in the summer

It's still just an autocomplete after all, with the features of a young politician

Have ChatGPT (or Copilot) respond it to it with an endless review asking it more and more and more 🍿

#GigAPI Assistants are in open beta! Designed to figure out data and timeseries ingested in GigAPI w/ BYO models, resources, compute local or remote. From zero to dashboard widget in < 30s and we're only just beginning!

@gigapipe.bsky.social

#timeseries #duckdb #gigapi #gigapipe #olap #llm #agent

I guess the function name itself wasn't enough :)

GitHub - metrico/influxdb3-community: Community InfluxDB 3.0 "IOx" static builds + containers + Examples for Developers & Integrators. Experiment with low-cost storage, unlimited cardinality and Fligh... Community InfluxDB 3.0 &quot;IOx&quot; static builds + containers + Examples for Developers &amp; Integrators. Experiment with low-cost storage, unlimited cardinality and FlightSQL APIs - metrico/i...

InfluxDB3 Community Builds are resumed for the open source community to leverage! The unlocked community releases (binary + docker) drop most of the "open core" limitations: github.com/metrico/infl...

#datafusion #arrow #flightsql #influxdb3 #iox

Release v1.3.2 · duckdb/duckdb-rs What's Changed Fix Parquet example to run from workspace root by @mlafeldt in #522 Update and clean up dependencies by @mlafeldt in #520 Optimize dependencies by @paolobarbolini in #513 Update Rus...

🐤 duckdb-rs v1.3.2 is here!

Highlights: DuckDB v1.3.2 core update, improved Arrow integration, enhanced Inserter trait, interrupt handling support, and lots of internal tooling updates.

github.com/duckdb/duckd...

Goodie waiting if you ever get to quack: query bsky from SQL in DuckDB using webmacros

parquet should be well supported and will save you ~17 dog years. can be compressed, queried, indexed. convert csv to parquet with DuckDB for extra points (or any other method)

In terms of DuckDB Community Extensions statistics, it seem 400-500 downloads/month is basically the zero line (builders, CI, etc?) above which you see real users. There's a big difference between extensions used by a few dozens and few thousands but unfortunately the stats don't tell us where from