Андрусь | Andrus (@andrus_a) 's Twitter Profile
Андрусь | Andrus

@andrus_a

Open-source & ObjectStyle. Making Java a viable option for data processing / analysis at DFLib.org

Now at B*Sky as bsky.app/profile/andrus…

ID: 85586288

linkhttps://objectstyle.com calendar_today27-10-2009 14:59:40

1,1K Tweet

378 Followers

290 Following

Андрусь | Andrus (@andrus_a) 's Twitter Profile Photo

Experimenting with B*sky. Looks quiet nice. No ads, very few bots, custom feeds and starter packs are cool. Also, now there are 20mln people there so nothing like my Mas*don ghosttown. If any of the Twitter friends are there, let's connect! My handle is in the profile

Андрусь | Andrus (@andrus_a) 's Twitter Profile Photo

I find myself often using Series object from DFLib instead of a Java List. It has so many advantages: * Immutable at the API level * Content search * Direct transform API - no need for streams in many cases * Slice and dice, e.g. "head(-1)" * Conforms to Java Iterable

Андрусь | Andrus (@andrus_a) 's Twitter Profile Photo

Woah, Liquibase is calling home since v4.30.0! There's a way to disable it, but you need to know about it. Totally not cool. We are upgrading LB in Bootique 3.0-M5, and will disable analytics for our users. liquibase.com/blog/product-u…

Bootique Project (@bootiqueproject) 's Twitter Profile Photo

Bootique 3.0-M6 is released. Jobs as lambdas, Jakarta version of Tapestry, new static servlet mapping API and security upgrades of upstream dependencies groups.google.com/g/bootique-use…

Андрусь | Andrus (@andrus_a) 's Twitter Profile Photo

Breaking a streak of Java data processing presentations, I will be talking about Bootique Project at NYJavaSIG. Bootique is a lightweight, fast, simple, CLI-friendly technology to write #Java apps, big and small

Igal Tabachnik (@hmemcpy) 's Twitter Profile Photo

Another local company that used Scala heavily 5 years ago is rewriting to Java because of difficulties with the language. People who think Scala is on the rise are delusional, sorry.

Андрусь | Andrus (@andrus_a) 's Twitter Profile Photo

Don't post here that much anymore, but this is exciting. DFLib 1.0 reaches GA 🎉💪 Don't let the #bigdata world intimidate you, now you can have a #DataFrame in any #Java app to do fun stuff with data (transformations, joins, ETL, charts, etc.)

Андрусь | Andrus (@andrus_a) 's Twitter Profile Photo

And our announcement was just posted on Reddit. Looks like there are folks in the Java community who care about DataFrames reddit.com/r/java/comment…

Андрусь | Andrus (@andrus_a) 's Twitter Profile Photo

On the heels of DFLib GA, 1.1.0 is released with: * web / URL connectors for binary data * primitive float columns (50% memory savings for cases like #1BRC) * chart item styling (incl. dynamic item sizes - essentially another visual dimension) github.com/dflib/dflib/di…

Андрусь | Andrus (@andrus_a) 's Twitter Profile Photo

The first milestone of DFLib (Java DataFrame) v2 is out. Lots of improvements, e.g: * 8x reduction of memory footprint of bool columns * Major new capabilities in "window" expressions * Shortened aggregation syntax in "group" expressions * Heatmap charts

The first milestone of DFLib (Java DataFrame) v2 is out. Lots of improvements, e.g: 
* 8x reduction of memory footprint of bool columns
* Major new capabilities in "window" expressions  
* Shortened aggregation syntax in "group" expressions 
* Heatmap charts
Andriy Burkov (@burkov) 's Twitter Profile Photo

Apple did more for AI than anyone else: they proved through peer-reviewed publications that LLMs are just neural networks and, as such, have all the limitations of other neural networks trained in a supervised way, which I and a few other voices tried to convey, but the noise

Андрусь | Andrus (@andrus_a) 's Twitter Profile Photo

DFLib just got a parsable expression grammar. Those expressions can be thought of as parts of SQL for the purpose of column transformation, filtering and (soon) DataFrame sorting github.com/dflib/dflib/di…

François Chollet (@fchollet) 's Twitter Profile Photo

In a parallel way, it is plausible that the primary effect of AI code assistants is to make you *feel* more productive, to gamify the development experience -- while delivering net negative productivity gains (which would explain the lack of increased output since 2022)