nrjames

That will never happen because there are too many other power-hungry people in the GOP who are not going to just let Trump sit in the White House indefinitely, if for no other reason.

Laudspeaker looks very useful but also a bit underbaked. For enterprise settings, you may want to consider its segmentation and journey capabilities in the context of orchestrating a suite of other tools that already are available. In our case, I’d love to use both of those features but would need native Snowflake integration and would want the Laudspeaker tools to integrate with Dagster/Airflow for deployment purposes. We have hundreds of millions of active users and send tens of billions of messages annually, across 17 or so channels, many of which are custom. The docs leave me wondering how I could integrate the tool with our existing infra.

We do overlapping inserts and let ReplacingMergeTree remove the duplicates. You can use the FINAL statement on select queries if you’re concerned about queries returning duplicates before the Clickhouse backend performs the deduplication mutation.

I work in gaming and stream events into a self-hosted Clickhouse db without Kafka. We just use the CH python connector and send records in batches of 100K, using ReplacingMergeTree for backfills, etc. It works very well. Unless you truly need up-to-the-minute analytics, it’s super easy to schedule with Dagster or Airflow or whatever. We process 100M+ events per day this way.

How are you going to feel when a random person you've never seen shows up and pushes deploy on your production code on a Friday afternoon? What is it that makes you think that this is a field that doesn't require training. Ever heard of the Sphere Standards[0]? Why don't you go flip through their handbook[1]? These standards exist because when the US and the USSR stopped their proxy war of providing humanitarian assistance and let everybody and their untrained NGO flood into Somalia 30 years ago, it was an absolute nightmare that caused total chaos of the people in need. It turns out the world has tried the "let everybody and their people come in and help" approach and it's terrible.

[0] https://spherestandards.org [1] https://spherestandards.org/handbook/editions/

Sure, talk to the American Red Cross Society, which is the only nonprofit with a Congressional mandate to provide services during disasters. During Katrina, their Volunteer to Paid Employee ratio was about 64 to 1. It literally is an organization that exists to train volunteers for providing appropriate response during disasters. Is it bureaucratic and bloated? Sure, but its entire job is to get trained people in during these situations. Likewise, the Southern Baptist Convention provides most of the mobile kitchen that the ARCS uses (though they have no mandate to do so).

There are official channels for getting volunteers trained up for these situations. The untrained folks have a tendency to make things worse, which is why the authorities usually don't want them there.

I've spent a significant amount of time working in disaster response. In the US (and probably elsewhere), you have "Disaster Cowboys." These are people who are untrained but fancy they know better how to do things than people who have spent their lives training for these types of scenarios. Picture 1000 people showing up with their pickups and John boats following Hurricane Katrina. Or, imagine the Coast Guard conducting a search grid and all of these people zooming around the grid in every which way.

On one hand, yes, they save some people. On the other hand, they are not on the official search grid rosters, don't tend to take directions well from the authorities, and can cause all sorts of havoc. They can make it harder to find people in need. In fact, when they don't coordinate with the authorities, they end up wasting the experts' time (who may end up covering the same ground) or, when they do coordinate, people in need end up getting missed because the random people who show up to help are not properly trained to conduct searches.

Finally, as this news article shows, these folks have a tendency to go to the media and present themselves as heroes. You very rarely see that from the official search and rescue/recovery teams.

Disaster response is extremely psychologically and emotionally stressful. You may ride a wave of adrenaline while you're out in the field, but most staff end up rotating out (by mandate) after several days to protect them. Even among professionals, there are a lot of people who end up relying on alcohol to ease their minds.

Most importantly, the dignity of the people being rescued/recovered is important. It is all about those people, not about the hero rescuers. They are literally in trauma and can be dangerous and/or in need of urgent help, medical or otherwise. It is desperately unfair to them when somebody untrained, with a "cowboy" attitude, shows up and flips the script. Or, even worse, the cowboy crashes his helicopter and there's one more person who needs to be rescued.

Not all people who randomly show up in their helicopters or John boats are like this, but enough of them are to sow chaos. And look at us all wasting time talking about this guy who feels slighted because his unprofessional help was refuted. It's the same story every time.

Let the professionals do their job and, at the very least, let them coordinate the volunteers who show up to help. It sounds like a great idea for everybody to show up in their helicopter, but trust me, it's not.

There's a documentary called Objectified that may interest you:

https://en.wikipedia.org/wiki/Objectified

I don't find them useful for coding much, but I do find them useful for searching through documentation and making suggestions. With Cursor, you can have it reference the docs for the library/framework you are using and then ask it questions about the docs. While this does not always work well, it can save a lot of time.

I'd like to see them sell a dock and turn the phone into a full blown gaming console that you can pick up and take with you, like the Switch.

My understanding is that it is somewhat common in adults to have a virus (cold, flu, etc) trigger a dairy protein allergy. They’re most often irreversible, or so the MDs told me. Anyhow, it’s been 15 years without dairy so I’m used to living without it. There are far more high quality non-dairy alternatives now than there were even 5 years ago, thankfully!

I developed a severe dairy protein allergy in my late 30s. At first, I thought it was just a lactose issue but anything dairy triggers it (with symptoms aligning with vasculitis). While I'm not vegan, it's just easier to eat vegan food when I go to a restaurant. I generally know what's safe from the grocery store but always look for the Contains Milk or Dairy Free labels.

The labeling is extremely important. I find it upsetting that there aren't more rigid requirements for labeling common allergens. Given that a large percentage of people are at least mildly lactose intolerant by the time they hit age 70, you would think this would have stricter labeling requirements.

Unity issues aside, Godot is awesome! I'd love to see an integration.

Daniel Shiffman has a fun neural network section in his (free online) book Nature of Code:

https://natureofcode.com/neural-networks/

Machinations[0] is a cool tool in this category:

[0]https://machinations.io

This is awesome and very useful. Thanks!

I’m not sure if you use Ultimate Guitar, but if so, go look up a simple fingerpicked song, like Helplessly Hoping, by Crosby Stills and Nash. If you look at the chords… well, that’s a straightforward view mostly for people interested in strumming. The chords are listed above the lyrics and the chord shapes are listed at the top. This is all relevant info.

Then look at the Pro tab. The fingerpicking patterns are shown, and which string/fret should be picked, but there’s no chord name listed when it would be helpful for somebody like me to see just a simple chord name above the measure as a cue for my fretboard hand without having to look at the individual plucked strings/frets spread over the measure to determine the chord. I hope that makes sense.

Sometime it might be a challenge to fit it, but there are many instances where it would be a helpful assist for reading the tab.

No, I really just want the chord name: C, F#, G/B, etc.

As somebody learning guitar and focused on fingerpicking... One of the things that bugs me about Ultimate Guitar tabs is that there are many places where it would help me better to learn to read the tabs if the relevant chord were displayed over the measures/notes. I'm not sure where you are going with your project, but I thought I'd just mention this since I spend a lot of time deciphering the tabs just to figure out what chord position my hand should be in.

I always bounce off of Factorio because I want to be able to play with the default settings but I always end up getting killed by bugs/aliens not long after starting the game.

How about hooking a camera up as input to some generative art algorithms in p5.js, Processing, or similar. Kids move/dance in front of the camera and the results are projected on an adjacent screen in realtime. Then you can save the pieces they generate and have a few of them printed!

I saw this p5.js usage of Hobby’s algorithm yesterday:

https://openprocessing.org/sketch/2257597/

I'm not sure how much of the underlying Clickhouse feature set is available, but the Clickhouse documentation about index types is thorough and informative.

https://clickhouse.com/docs/en/optimize/sparse-primary-index...

Varroa mites are a vector for diseases and fungus and most likely the distal cause of most of the dying colonies. If they are not aggressively managed, they eventually lead to the death or absconding of almost every colony.

I'm not sure it's the downturn, necessarily. I used to be enthralled with tech, full of energy to explore new tools, programming languages, etc. I think I slowly awoke to the fact that it's not the tech but the people that make it interesting and rewarding. For me (and I'm not trying to project on anybody else), interactions and activities with people who were doing things that are not tech seemed deeper and more genuine -- almost as if it was easier for me to get to know them and enjoy their company in a non-tech context. As a result, my hobbies and interests have pivoted to activities that don't involve computers much at all. My career remains in tech and I try to put my best efforts into the work that I do, but it's becoming more difficult as I continue to realize that life, for me, is better when most of it is spent away from screens.

I love this and hope that a django-observable package comes along that makes it very easy to integrate and serve these static apps through a larger Django site.

Not sure where you live and whether there's an opportunity nearby, but curling is a ton of fun and a very social sport. Check to see if there's a local curling club and, if so, whether they have an instructional league. I really enjoy the sport but it's the people that I've met while doing it that have been the best part. There's a tradition called "broomstacking" that follows games, where the winning and losing teams all sit around a table and socialize (maybe have a beer, etc.). Being heavily Canadian influenced, the #1 rule is not to disparage anybody else. The rude and mean folks kind of self-select out.

And mangosteen: https://en.wikipedia.org/wiki/Mangosteen

Part of me looks forward to these experiences. A larger part of me already mourns the decline is purely human creativity that they suggest. Perhaps AI models will make the perfect game or music or write the perfect novel. I'm certain they'll be programmed to reproduce the funky jank and humans bring to everything they create. At some level, though, we each have our own story that we want to tell and when there are so many automated voices in the room, it's going to be harder to tell those stories.

Because of the integration with Dagster, the calls are made when my Dagster flows run. I also send unreasonably large data payloads. Anything going into Django gets handed to a Celery batch processing job and just returns the job id (Dagster monitors for failure, too). Data pulled out of Django is paginated. Sometimes it's 1000s of calls per minute, sometimes it's none at all for hours. I realize this isn't a normal use case, though. :)