pmcf

I think Google doesn’t turn Gemini loose on docs the same reason Apple doesn’t turn AI loose on your phone. It’s just not reliable enough to let 99.99% of the world use it. Those of us on the bleeding edge have been fine tweaking and working with inconsistencies. If you put a lot of work in, you get a productivity boost. Think of the family member you are “tech support” for. (You know who you are) would you recommend that to them? Yeah. Me neither.

And includes a YouTube subscription. Top that OpenAI!

Oh look. A flying saucer. Quick, get our worst camera.

Line of sight issues are simply a wattage issue. A gigawatt laser is impervious to rain, a bird, a flock of birds, a bird and the tree it sitting in. Probably the entire forest. Let’s just say there are some solutions well in hand.

Not today, regulatory capture. Not today.

"The Navy is a master plan designed by geniuses for execution by idiots."

- Herman Wouk, The Caine Mutiny

This is completely true. 18-20-year-old kids launch and arrest aircraft on a carrier while simultaneously performing an underway replenishment, and it's just another day.

This was right before the gulf war so I may have met him! Assuming he was a gunners mate, that crew had a lot of moments of touching history. Besides mechanical computers, it’s a really dangerous place since they had to handle massive bags of flash powder.

My ship was near the USS Iowa when turret two went up. A sobering experience when you think how much risk the turret crews are in just by doing their jobs.

In 1989 I was a data systems tech on a Destroyer going through some overhaul at the shipyard in Pascagoula Mississippi. Moored right next to us was the battleship Wisconsin. Huge relic from WW2 but still going through modernization. A bunch of us that worked on combat systems got invited for a tour of their fire control systems.

Wow. Just wow. All mechanical computers calculating fire control solutions for the big 16 inch guns. The guys giving the tour were well beyond the age for regular military retirement. Come to find out, they were all reactivated because practical knowledge of the mechanical computers had since left the navy. That was a very cool day.

I read this as “In My Opinion” and really thought this about AI dealing with opinionated people. Nope. HN is still safe. For now…

TBH I was expecting something about plugging humans into a huge bioenergy farm and letting them live out their lives in a simulation.

Not to be confused with Astra DB, Cassandra as a Service.. http://astra.datastax.com

“Space is big. You just won't believe how vastly, hugely, mind-bogglingly big it is. I mean, you may think it's long way down the road to the chemist's, but that's just peanuts to space.”

I did write a book related to the topic, however it may take a bit longer to get through.

https://www.oreilly.com/library/view/managing-cloud-native/9...

Python for analyzing data and crunching numbers? Awesome!

Python for serving HTTP? This could get expensive—the why, what, and choices teams have in deploying AI applications.

Beyond the VR thing. I keep thinking this could have more immediate impact on manufacturing and logistics. Imagine a delivery truck with this on the floor that could optimally organize packages dynamically.

What’s the airspeed velocity of an unladen swallow in a hurricane?

I do the same thing to my kids to change their performance.

It was in research, but now the FDA has approved it as a therapy. So I guess it takes 13 years to go from paper to clinic?

That speed thing is a nice bit of marketing from Scylla but not really true in the real world.

Fast writes and slow reads was true in 2012. The project has been busy since.

Yes. You still have to read the docs before randomly setting “auto_bootstrap” to false.

Or just use Astra and not worry about scaling your own cluster.

Everyone complains about how expensive Splunk is but the amount of compute and storage consumed by processing logs is ridiculous.

I feel like we should be talking about the sad state of logging where we think it’s perfectly ok to dump millions of 10k stack trace dumps and think that should be cheap.

You have to be ducking kidding. Works great for me.

Same argument applies when I hear people say “we can just do an ETL job.” And why storage is so cheap in the cloud. Once you get a lot, it’s hard and expensive to move. And definitely why things like AWS Snowmobile exist.

This link is being copied to slack channels everywhere with “I told you so” statements and something about “You kids and your GPTs” and “Back in my day”

Meanwhile copilot is becoming a superpower to those who figure it out.

And it hasn’t even been a year! I’m going to need a lot more popcorn.

Survey results from Java developers about the rise of Gen AI.

The Cassandra project recently[1] added vector search. Relative to a lot of other features, it was fairly simple. A new 'vector' type and an extension of the existing indexing system using the Lucene HNSW library. Now we'll be finding ways to optimize and improve performance with better algorithms and query schemes.

What we won't be doing is figuring out how to scale to petabytes of data distributed across multiple data centers in a massive active-active cluster. We've spent the last 14 years perfecting that, and still have work to do. With the benefit of hindsight, if you have a database that is less than 10 years old, all I have to say is good luck. You have some challenging days ahead.

1. https://cwiki.apache.org/confluence/display/CASSANDRA/CEP-30...

Finally. I was wondering if Java was going to show up to this circus. Python has definitely had it's viral moment with GenAI.

OP here! When asked about LLMs and Apache Cassandra, ChatGPT started going on about this amazing OSS library from OpenAI called CassIO. Here's the story about making a hallucination into reality. This was a fun project but did the AI just coerce us into something?

Sure I can answer both. The OS scheduler CFQ is generally bad for high volume disk applications. Used deadline in this case. Amy's Guide is still a treasure trove of info and a solid recommended read: https://tobert.github.io/pages/als-cassandra-21-tuning-guide...

If you are using Spark in a bare metal cluster with spark-submit, the above advice applies. In Kubernetes, never use the default pod scheduler with analytic workloads. Great choices are Volcano(https://volcano.sh/en/) and Yunikorn(https://yunikorn.apache.org/). Also great and evolving projects to support and contribute if you can.