Hacker Newsnew | past | comments | ask | show | jobs | submit | drited's commentslogin

Yes the circumstances are well known. Li Liu convinced Munger and Munger talked to Buffett.

Cahm


Like the "cam" in "camera"?


I've been thinking about this for a minute, and I think if an American were to say "why", and take only the most open vowel sound from that word and put it between "k" and "m", you get a pretty decent Australian pronunciation. I am an Australian so I could be entirely wrong about how one pronounces "why".


No, with a long vowel sound. Caaahm. The L is blended into the M so much that it's almost silent.

Unless you're specifically enunciating it. The common usage lacks the L sound, but it is acceptable to intentionally add it back in for disambiguation


Would you have any standard prompts you could share which ask it to make a draft with you'd want (eg unit tests etc)?


    C++, Linux: write an audio processing loop for ALSA    
    reading audio input, processing it, and then outputting
    audio on ALSA devices. Include code to open and close
    the ALSA devices. Wrap the code up in a class. Use 
    Camelcase naming for C++ methods.
    Skip the explanations.
``` Run it through grok:

    https://grok.com/ 
When I ACTUALLY wrote that code the first time, it took me about two weeks to get it right. (horrifying documentation set, with inadequate sample code).

Typically, I'll edit code like this from top to bottom in order to get it to conform to my preferred coding idioms. And I will, of course, submit the code to the same sort of review that I would give my own first-cut code. And the way initialization parameters are passed in needs work. (A follow-on prompt would probably fix that). This is not a fire and forget sort of activity. Hard to say whether that code is right or not; but even if it's not, it would have saved me at least 12 days of effort.

Why did I choose that prompt? Because I have learned through use that AIs do will well with these sorts of coding tasks. I'm still learning, and making new discoveries every day. Today's discovery: it is SO easy to implement SQLLite database in C++ using an AI when you go at it the right way!


That rely heavily on your mental model of ALSA to write a prompt like that. For example, I believe macOS audio stack is node based like pipewire. For someone who is knowledgeable about the domain, it's easy enough to get some base output to review and iterate upon. Especially if there was enough training data or you constrain the output with the context. So there's no actual time saving because you have to take in account the time you spent learning about the domain.

That is why some people don't find AI that essential, if you have the knowledge, you already know how to find a specific part in the documentation to refresh your semantics and the time saved is minuscule.


Fer goodness sake. Eyeroll.

   Write an audio processing loop for pipewire. Wrap the code up in a 
   C++ class. Read audio data, process it and output through an output 
   port. Skip the explanations. Use CamelCase names for methods.
   Bundle all the configuration options up into a single
   structure.
Run it through grok. I'd actually use VSCode Copilot Claude Sonnet 4. Grok is being used so that people who do not have access to a coding AI can see what they would get if they did.

I'd use that code as a starting point despite having zero knowledge of pipewire. And probably fill in other bits using AI as the need arises. "Read the audio data, process it, output it" is hardly deep domain knowledge.


Results with gemini

https://pastebin.com/6b4yhfYw

A 5 second search on DDG ("easyeffects") and a 10 second navigation on github.

https://github.com/wwmm/easyeffects/blob/master/src/plugin_b...

But that is GPL 3.0 and a lot of people want to use the license laundering LLM machine.

N.B. I already know about easyeffects from when I was seeking for a software equalizer

EDIT

Another 30 seconds exploration ("pipewire" on DDG, finding the main site, then goes on the documentation page, and the tutorial section).

https://docs.pipewire.org/audio-dsp-filter_8c-example.html

There's a lot of way to find truthful information without playing Russian roulette with an LLM.


It seems to have a clock so it doesn't turn on at night?


Correct.


Can you setup entra authentication with PgAdmin? I'm more of a MS Sql person so I don't know, but if not the security improvement from this would be a huge improvement


I would be curious about context window size that would be expected when generating ballpark 20 to 20 tokens per second using Deepseek-R1 Q4 on this hardware?


The evidence is mixed, but some studies (e.g. Jaeggi) did find transfer effects from n back training to fluid intelligence.

It only takes 40 mins a day for 8 weeks to test it out. Much less time than the commitment to learn a new language.

Having tried it, I wouldn't be surprised if the mixed results were due to improper adherence and misunderstanding of how n back works by some study participants. In other words, I think it's possible that results would be less mixed for someone who is already starting from a point of solid intelligence and who is driven enough to put in the hard,focused work to get to higher n back levels.


I see in another comment thread you mentioned downloading the VM iso, presumably from a central source. Your comment in this thread didn't mention that so perhaps this answer (incorrectly) assumes the VM you are talking about was locally maintained/created?


Cool but offence is only half the story! Any plans to add defensive stats or stuff like turnovers / steals etc?

Also any plans to make it work from camera footage e.g. A 360 camera that can capture the whole court at once instead of syncing?


Yeah! We are planning to add blocks, steals, and rebounds in the coming months. I think fouls is still a reach goal right now (hard even for humans) but maybe one day.

Re: capturing the whole court, I think we are considering supporting panning of the camera (in addition to syncing). We focus pretty heavily on making sure things work with just a normal phone so it's more accessible to everyone. We are definitely starting bottom up but maybe one day we will get to more high tech hardware.


Re:fouls.

At my pickups courts we call our own fouls by saying Foul. Perhaps the mic can pickup key words matched the play stopping.


Oh neat, yeah we could some speech to text thing or maybe even some like visual signal thing to call fouls.


Very cool of you to take on board feedback.

In relation to user experience on first try...I just wanted to test it out for 30 second to see if it's worth keeping. I haven't tested now, will have to wait until I get an hour free later because there are a few roadblocks. - You require email and password registration and to click a verification link. Not too bad - When first launched the app asks you to allow mic permissions by clicking settings. However in Hooper settings on Android there is no mic permissions option. Bad. - You require 5gb free to use the product at all. Understandable given storage is required for video but how about reducing that so users can do like a 30-second test for first use? That way they don't have to spend an hour going through their phone's media to see what they want to keep and what they can delete.

You'd be amazed at how many people (like me lol) have almost full storage on their device most of the time.

Edit: oh....it's not a setting for mic permissions, it's a general permissions setting and if you touch that you can add additional permissions. Based on the text of the popup I hadn't guessed that was the right place to look and I hadn't realised it was possible to add permissions by touching that area of the settings menu.


Wow that's terrible. I'm curious as to whether your contract with them allows for meaningful compensation in an event like this or is it just limited to the price of the software?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: