HackerTrans
トップ新着トレンドコメント過去質問紹介求人

crowbahr

no profile record

コメント

crowbahr
·28 日前·議論
New account made just for this comment is always suspicious.

Look - the correct amount of hate crime is always 0 but using a percentage masks the amount of crime that comes from noise. Hate crime is not evenly distributed - no crime is - and a rise from 10 to 20 per month in a city of 8.5 million is not the cataclysm you're acting like it is.

Again: no hate crime is good and this increase is an unalloyed ill. It's something that deserves attention and NYPD resources, as well as public campaigns.

The last thing I want is anyone to be attacked for their religious beliefs.

But the fact that this talking point is heavily botted is indicative of broader initiatives to make sure it stays a talking point disproportionately to its impact.
crowbahr
·28 日前·議論
Cynically: this is an attempt to quash open source or discount model competition through regulatory capture.
crowbahr
·先月·議論
Not when a datacenter is the size of manhattan
crowbahr
·先月·議論
How many biometrics per minute can every glass in Chicago collect, each one costing 5k? It's not per biometric, it's per violation. Collecting the same face 20x is 20 infractions.
crowbahr
·2 か月前·議論
Google is making money on selling cloud compute. Their margins have gone from 9% to 32%.

They're soaking up the investor bonanza into AI - Gemini ain't making them money.

For context Cloud Compute made 20bn in Q1, Other services made 90bn.
crowbahr
·2 か月前·議論
Your example is cultural.
crowbahr
·2 か月前·議論
And human civilization entirely sprung up during it, all of our nations, our cities, our pastures, our lives are built on the ice age. We need to start cooling the world down and we're doing the opposite
crowbahr
·4 か月前·議論
Brussels have chemically changed significantly over the past 20 years of genetic engineering - they're more nutritious than ever and are nowhere near as bitter!
crowbahr
·5 か月前·議論
If you give your assistant a task and they fall for obvious lies they won't be your assistant long. The point of an assistant is that you can trust them to do things for you.
crowbahr
·5 か月前·議論
Where on earth are you getting these numbers? Why would a SaaS company that is fighting for market dominance withhold 10x performance if they had it? Where are you getting 2.5x?

This is such bizarre magical thinking, borderline conspiratorial.

There is no reason to believe any of the big AI players are serving anything less than the best trade off of stability and speed that they can possibly muster, especially when their cost ratios are so bad.
crowbahr
·5 か月前·議論
I do think xcode is the worst IDE I've ever worked in, bar none. Even Eclipse from 2010 was better than xcode from 2025
crowbahr
·6 か月前·議論
Flutter has been abandoned by all the large companies - Google is throwing their weight in on KMP and has laid off the flutter teams.
crowbahr
·6 か月前·議論
Touching xcode to avoid touching Android is like touching concentrated hydrochloric acid to avoid breathing a fart
crowbahr
·7 か月前·議論
You can get brick sets that are just a bunch of bricks - look up Creative Brick Box or Creative Vehicle Box

Feels just like Grandma's ole box o bricks
crowbahr
·7 か月前·議論
Minecraft: weak magnets
crowbahr
·7 か月前·議論
He didn't say AI helps you ship with fewer bugs, just faster.

It's faster. It's not better, it's not more stable - the only promise is velocity on initial ship.

Oh you want to follow up? Extend it? Reconfigure? Sry can only go faster.
crowbahr
·7 か月前·議論
You do know that HN in in the category of social media right?
crowbahr
·7 か月前·議論
You can't really compare those 2. Agents a re non-deterministic. I can tell Clod to go update my unit test coverage and it will choke itself, burn 200k tokens and then loudly proclaim "Great! I've updated unit test coverage".

I'll kill that terminal, open it again and run the exact same command. 30k tokens, actually adds new tests.

It's hard to "learn" when the feedback cycle can take 30 minutes and result in the agent sitting in the corner touching itself and crooning about what a good boy it is. It's hard to _want_ to learn when you can't trust the damn thing with the same prompt twice.
crowbahr
·7 か月前·議論
What data source are you using? AeroAPI?
crowbahr
·8 か月前·議論
The linked study r values are all minor to no effect. -0.37 as an r value is barely correlated. Talk to me when they're up above -.5

How much of it is just constantly being told that SFVs are bad for you?

The linked article is a meta analysis with basically 0 controls that finds a statistical correlation barely better than background noise. Yawn