FWIW I think if you are just doing pure analytics and nothing else, Parquet will probably continue to do the job for you just fine, and you don't need to touch your workloads at all.
These new formats I think will find a niche where people aren't just running Spark jobs, but doing lots of systems building over large tables. If you're building a PB-scale data warehouse, you care a lot about the file format b/c it is a big factor in your performance curve, and you're willing to ship new experimental codecs in response to new datatypes you want to support that the system wasn't originally designed for, or you want to use a newly invented compressor.
> Also, my main gripe with parquet (single table per file) is not even addressed, so, also the name is a bit hyped up.
This is really more of an expectation that has been put on file formats by the query engines. Spark/Datafusion/DuckDB wouldn't really know what to do with a multi-table file.
> Parquet is unfortunately very good just by virtue of being first, and so widely supported
IMO that is not how technology works. It is great that Parquet is so good at a lot of things, but that does not mean just because it came first that it deserves to be the only analytic file format forever.
> Its main result seems to be improved random access which, although certainly welcome, is not the point of columnar storage, as columnar storage was invented to exchange random access for something else: fast analytics
Fast analytics, as well as newer ML-shaped workloads, are inherently mix of batch scans and random access.
Some of the authors of F3 previously authored another paper that goes into the details of the shortcomings of Parquet
All of the newer formats that popped up recently (Vortex, Lance, F3 now) have been working on solving the problems outlined in that paper.
Lance has some interesting ideas, Vortex focuses on extensibility and performance by replacing all of Parquet's black-box encoders with fully transparent encodings. This solves the tradeoff between bulk and element decoding, allowing you to have efficient full scans and really fast random access.
Disclaimer: I work on Vortex, so a lot of these questions about "what is the point of building a new format" are things that I have grappled with myself.
> “I never got my driver's license, and I rely on Waymo to commute to an office every day," said Sarah Paige Roland, a Waymo rider in Phoenix. "I get privacy, time back, a safe ride, and I'm not obligated to talk to someone that I don't want to talk to."
I recognize that this is a luxury product but I kind of laughed out loud at this testimonial. The amount of privilege you need to have to grow up and live in *Arizona* without ever learning how to drive is insane.
I think I'm missing the excitement. This is an artist's rendering of a supposed massive orb in the sky? I am more impressed by the actual UAV footage that has been released previously.
In the case of the "Gulf of America" thing there was a clear and open statement by the executive that they wanted the maps changed (note that even in the US the name is still legally "Gulf of Mexico"). Apple and Google both decided to acquiesce to curry favor.
TMK there is no current government order to eliminate large swaths of Lebanon from maps. So the fact that Apple is doing this (seemingly on its own, despite all other mapping services reflecting the original place names) is the thing I'm explicitly calling out as being weird.
Sure, but I'm in the US, which is not a party to the conflict between Israel and Lebanon. To my knowledge the US continues to recognize Lebanon as a sovereign country.
OSM is a foundational data layer for GIS. If you're building a mapping service, you're almost certainly using OSM augmented by satellite imagery and other sources to find population zones that OSM has not found yet.
If you look at the Apple Maps satellite layer, you see thousands of structures spread across the area.
It is a reasonable assumption that these population centers were labeled and Apple (or one of its data partners) has withdrawn the labels.
This reminds me a bit of the Gulf of America fiasco from last year where if you changed your location to outside the US it would go back to showing Gulf of Mexico.
I'm not sure why they would do this for US users unless the US government requested it.
I'm not sure how much of that converts to revenue. If it's free plan users, that's just cost. You can say what you want about "creating a training data moat" but that doesn't seem like it's prevented the other labs from putting out excellent models.
There's something a little off about the projection logic when you drop into the Leaflet view, you'll notice that when you pan around after zooming the planes shift their location.
It's even funnier actually. Yes, Palantir does have free Zyn vending machines in every office, but the Zyn is only for visiting customers. Employees are explicitly prohibited from using the machines.
> - They assume their hard-to-program but faster architecture will get figured out by devs. It won't.
Or it will get figured out in the niche fields where people are willing to figure out really hard stuff to squeeze out max performance (PE, hedge funds, intelligence)
Either way agree, it's hard to get mass adoption without the software ecosystem feeding back in
FWIW I think if you are just doing pure analytics and nothing else, Parquet will probably continue to do the job for you just fine, and you don't need to touch your workloads at all.
These new formats I think will find a niche where people aren't just running Spark jobs, but doing lots of systems building over large tables. If you're building a PB-scale data warehouse, you care a lot about the file format b/c it is a big factor in your performance curve, and you're willing to ship new experimental codecs in response to new datatypes you want to support that the system wasn't originally designed for, or you want to use a newly invented compressor.