pjot

10 year old me was terrified of the FBI warning message at the beginning of every VHS tape.

Love your page. Especially your careers page - never seen one posted as a git repo before.

You really know your audience.

I’m curious who your customers are; being tech savvy enough to use an api, but not enough to scrape the web confuses me. Mind explaining how you make money?

Cosine similarity is not affected by the size of the vector but only by the angle between them. This means that vectors with large or small values will have the same cosine similarity as long as they point in the same direction.

In semantic search the magnitude of vectors isn’t relevant - if you needed a measure that took magnitude into account than Euclidean would make sense. E.g. image embeddings based on pixel intensity.

I actually just finished a POC using DuckDB that does similarity search for HN comments.

https://github.com/patricktrainer/hackernews-comment-search

You can define a cache key in your .gitlab-ci.yml

https://docs.gitlab.com/ee/ci/yaml/index.html#cache

The A320s do seem bigger now that you mention it. On the other hand, I do really enjoy the blue ambient light in the 737 - can’t really explain why, but it’s comforting.

What else would you suggest? Maybe “new” and “toil”

What are your thoughts around the various vector dbs (pinecone, etc)? DuckDB or pgVector or with all of the work going forward in Arrow being able to already support vectors/arrays, it seems that the specific “vector” class of db is hype/marketing.

The point is more so in aim of creating interoperability between systems and making them in turn composable.

When there’s a common intermediate representation you can pass around those compute instructions and execute. And when there’s shared memory formats data can pass from storage to engine without serialization/deserialization.

So it wouldn’t matter if data is here or there, in this or that format, because the instructions are the same the specific interface (snowflake, MySQL, a local parquet file, etc) is irrelevant mitigating the need for glue code.

Apache Arrow and Substrait have been working towards making this a reality. I see a future where executing a query can/will send plans to many different engines distributed across the cloud, but also locally on your on machine.

“Street fighting mathematics” is a fantastic textbook.

https://books.google.com/books/about/Street_Fighting_Mathema...

I enjoyed this post on pull/push based query engines - its examples were wonderfully simple.

https://justinjaffray.com/query-engines-push-vs.-pull/

Do you not think rodents are in your garden … where food is left around?

Using Apple’s translate function I was able to read many of the posts - very interesting to see the differences between American and Turkish social media.

There were many posts about cats and their livelihoods and protection. Love that

Comrade 1: What are one potato say other potato?

Comrade 2: Premise ridiculous! Who have two potato?

That’s funny! So strange it thinks it’s GL

I wasn’t sure what Perplexity was so I asked it “What is this”. This was its response

  Google Lens is an application that allows users to search and identify objects, translate text, and perform various tasks using just a camera or a photo. It is available on both Android and iOS devices. Some key features of Google Lens include:
  
  Using the camera to identify objects, such as artwork, plants, or everyday items.

  Translating text in real-time from over 100 languages.

  Copying paragraphs, serial numbers, and more from an image and pasting it on your phone or computer.

  Searching for songs by humming.

  The application is designed to make everyday life easier and more convenient by providing users with quick and accurate information about their surroundings.

This is what Substrait and Apache Arrow seek to solve; but by serializing query plans that can be produced/consumed by different engines. It’s like an IR for sql.

https://substrait.io/

In the Philippines McDonald’s sells spaghetti!

  Voyager 1 is so far away that it takes 22.5 hours for commands sent from Earth to reach the spacecraft. Additionally, the team must wait 45 hours to receive a response.

I’m guessing “hotfix” commits don’t exist in this domain

killall -9 vim

What are you writing the games in?

And the drywall guy is his brother!

The ATT guy just drilled a hole in the side of my wall to pass the fiber through. I rent, so I don’t really care, but if the plumber did that I’d be pissed.

The Blackwing 602 is my favorite pencil. One very nifty feature is its swappable eraser - it’s also rectangular.

https://blackwing602.com/products/blackwing-602-set-of-12

Swiggity-shwa!

“Courage, the cowardly dog” was horrifying.

“I said consummate v's! Consummate!”

“First draw an S, okay now a more different S”