วีดีโอ

You are a genius! Fantastic video! Thanks!

amazing video!!

great explanation, thank you 👍🙏

Greaaaat video, thanks!!

This looks really promising!

Love it!

What is so funny?

Hi Matt - I think what you've done here is absolutely brilliant!!!! I think you solved the one major limitation of GraphQL that's been preventing it from taking over the data world. Have you thought about open-sourcing this?

Awesome presentation. Thanks for sharing this

This was really helpful, quite a full and easy to understand introduction, thanks!

Tom, I am not clear what specific data (from nflverse) you choose to work with. In your code what exactly is "data" ??

A brief discussion of what the parquet file format is and what the advantages are over regular flat files.

An excellent case for Arrow and DuckDB

Thank you for the session. Is there a detailed documentation for JS support. The apache arrow site points to a blank page

🎉🎉🎉😊

Hi, is it possible to share link to the slides used the presentation? Thanks

Great video - thanks Tom!

Incredible useful thank you so much!…

So to summarize it, using SQL is getting old very fast! So would you suggest that a beginner should learn the very basics in SQL and focus more on ibis or now ponder for example? I don’t see why you would need SQL on a very high level when the playground is changing so rapidly nowadays. I get that many users with 7+ years of SQL knowledge are very frustrated that you basically could just use something like ibis but it is what it is

Thank you! I enjoyed your presentation and learned a few important “missing pieces” in my understanding of how these tools can work together.

Great talk and great initiative. Could you point out how and where to find out when the arrow timeseries compute functions come online?

super! thanks for this very useful tutorial.

great video, thank you for such a detailed comparison

I don't know much about arrow but if it is lazy the first timing can be wrong because arrow may not do the calculation until you need to use the output

Huge thanks for this talk! And for uploading it :)

Thank you Danielle Navarro !

Super tutorial Danielle. Thank you.

Very insightful talk, thank you!

That looks fantastic! But what if I do not have my date as cleanly organzied in many "smaller" files, but rather one giant csv. Does arrow still have benefits?:)

Can we replace Arrow Compute Engine (Acero) with Velox ?

Thanks Tianyu, we are focused on a similar scaled down approach regarding transactional Arrow for production workloads and your talk and reference papers put it all together. Thank you again.

thanks for sharing this video.😁

If you have any questions, please ask me on Twitter! twitter.com/ktou (English is OK!)