sberens

My experience with auth solutions/libraries is you invariably want to customize some flow/data/functions, but it's impossible because the library isn't flexible enough.

A better solution might be premade auth primitives (oauth providers, db adapters) that you copy paste into your codebase shadcn-style, and modify when necessary.

I feel like "don't roll your own auth" is less true than it was 5-10 years ago as now there is an abundance of good references and core libraries.

I wish IDEs had more features/tooling around types. For example, something like "expand all types by one level" where

{ foo: Bar } would expand to { foo : { bar1: string, bar2: Baz } } (and you could trigger it again to expand Baz)

(this would be especially nice if it worked with vscode/cursor on-hover type definitions)

Yep you can schedule it to get brighter/dimmer and warmer/cooler from your existing [Matter compatible] smart home system, and with HomeKit adaptive lighting you can have it follow the sun.

Really cool! I'm working on a lamp that gives you daylight levels of light indoors (albeit no raleigh scattering and columnated light). On the bright side (pun intended), it's 50,000 lumens instead of ~4500. https://getbrighter.com/

These are not the only 6 engineers working for DOGE, just the youngest. They have supervision as well.

You get 6x less lumens/watt with a halogen lamp

They claim 101,000 lux, not lumens. Lux is light per square meter (roughly), lumens is total light output. The closer you get to the source, the higher the lux will be, so it's hard to compare lux equally. Based on their 230W power draw and typical COB efficiency, I'd guess it's 20,000-35,000 lumens.

Tldr, we're brighter, fanless, have adjustable color temperature, smart home compatible, and more aesthetic.

I'm working on something like the lamp depicted in the images[0]. It's bright, diffuse, dimmable, flicker free, and high CRI.

[0]https://getbrighter.com

This is really cool! Though for some reason the lat/long coordinates of places are completely off, despite the addresses being correct.

I searched "show me 5 parks in SF", and all 5 were in the wrong spot. For example:

Lafayette Park Lafayette & Gough St, San Francisco, CA 94109 Latitude: 37.7955 Longitutde: -122.4668

From the post:

* Signal Intel is in it for the long term, specifically kill the dividend.

* Be more customer centric/rebuild customer trust, specifically publicly acknowledge the 10nm slip.

Though it seems customer trust in Intel has eroded to the point where only actions will change customers' perception of Intel — I'm not sure what Intel could have communicated to Tofino/Oxide that they hadn't already.

Does anyone have a set of queries that google returns poor results for?

I spent a few minutes looking at my search history (filtering chrome history by "google search"), and the vast of my queries are quite simple (e.g. people's names) that google does well on (in fact I find google search for people better than linkedin sometimes).

I also tried a few complex queries and compared them to Kagi:

"How much bitcoin does microstrategy own" -> Google returns the correct snippet from here[0] while Kagi only linked to articles about how much it acquired in the last few days.

"how to pronounce stratchery" -> Google returns the correct snippet from the Stratechery website[1] while Kagi's first result is a spam entry[2] with the wrong pronunciation (the second result is a tweet with the correct pronunciation).

I'd be curious to see more comparisons!

Edit: I just remembered Dan Luu's post (https://danluu.com/seo-spam/) but after looking through my search history, the queries he uses are not at all representative of my day to day searches.

[0]https://www.forbes.com/sites/digital-assets/2024/11/29/micro...

[1]https://stratechery.com/category/about/#:~:text=UPDATE%3A%20....

[2]https://www.howtopronounce.com/stratechery

I'm curious, did you consider hiring a CEO to run the company? Or is the company too small for that?

How long do you expect LLMs/agents to be unable to do this?

I wouldn't say humans are so different. You could argue we've been trained on about one quadrillion bytes of visual data by the time we're 4 years old: https://x.com/ylecun/status/1750614681209983231

What is an abstract reasoning task that your average 15 year old (who has "general intelligence") can do that you think LLMs can't do?

I'd like to remind people of this single-sentence statement:

> Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war. [0]

Signed by Demis Hassabis, Sam Altman, and Bill Gates among others

[0] https://www.safe.ai/work/statement-on-ai-risk

I tried both Mint and YNAB but they both had syncing issues that infuriated me. My friend recommended Copilot[0] and it's been a joy to use. It's designed so well that it somehow got me to take the time to label my transactions & set my budget, where with Mint/YNAB I would just add my accounts & not do anything else because of the laggy interfaces.

[0]https://copilot.money/

I've heard good things about Bradfield CS[0] and CSPrimer[1] (both run by the same person).

[0] https://bradfieldcs.com/

[1] https://csprimer.com/

it's significantly slower than dedicated vector dbs at >100k vectors

Where does the H200 fit in if the B100 is coming out the same year with 2x the performance? Is the H200 just cheaper than the B100?

> They are ... as likely to be the next MySpace as they are Facebook.

I will happily take a 1:1 bet on them not going the way of MySpace in, say, 5 years!

Who will be the first to wrap this in an API using twilio?

Here is the Processing Foundation's 2022 Funding Update:

https://medium.com/processing-foundation/processing-foundati...

Nice! I'm curious, does Pelican support custom components in your posts?

[Genuinely curious,] what have people moved on to?

I created a prediction market to estimate the success of this manifesto (along with other efforts): https://manifold.markets/SimonBerens/will-terraform-be-mpl-l...

Did you generate the diagrams in the README.md with some tool?