naasking

27 ms·

571.
by naasking 1y ago
In principle, serverless should be very efficient and cheap if all code is running the same VM. The heterogeneous architecture must make it expensive. Maybe serverless using WASM will one day make it cheap.
572.
by naasking 1y ago
> only maintain their outdated kernel versions, and never maintain them in the sense of updating their driver code into newer kernel releases Just an aside, but this is one of the major downsides of monolithic kernels, and a case where m
573.
by naasking 1y ago
Most people have no interest in redistributing software, they only want to use it.
574.
by naasking 1y ago
Also, libreboot supports quite a bit of hardware. What binary blobs are left of you have one of those systems?
575.
by naasking 1y ago
Deleting comments even when instructed not to do so is another failure mode. They definitely require more fine-tuning in these cases.
576.
by naasking 1y ago
> I feel that free software sometimes obsesses over the 1% when the 99% of their objective is achieved. Yes, that's what it means to have ideals.
577.
by naasking 1y ago
> I'll never doubt the ability of people like yourself to consistently mischaracterize human capabilities Ditto for your mischaracterizations of LLMs. > There are still so many obvious errors (noticeable by just using Claude or C
578.
by naasking 1y ago
I think more charitably, they meant either that 1. There is often more than one way to arrive at any given answer, or 2. Many questions are ambiguous and so may have many different answers.
579.
by naasking 1y ago
> The problem with LLMs is that they don't have a process to guarantee that a solution is correct Neither do we. > They will give a solution that seems correct under their heuristic reasoning, but they arrived at that result in a
580.
by naasking 1y ago
I dispute the infinity of stupidity! It can only be infinite on an infinite timeline, as there are a finite number of people at any given time! But is time infinite?
581.
by naasking 1y ago
Finite resources are a common problem though.
582.
by naasking 1y ago
Hey, guess what else is processed at runtime: all of the dynamically typed programming languages that run half of the internet. I wonder how they can still get reliability despite that?
583.
by naasking 1y ago
Requiring specialized hardware seems overly strict now that we can handle such things at a higher level via something like the fault tolerant lambda calculus.
584.
by naasking 1y ago
There are definitely some assumptions that hold for all environments. For instance, "all resources are finite".
585.
by naasking 1y ago
Yes, which presumably would have actually been tested by someone before it was uploaded.
586.
by naasking 1y ago
Everything can be abstracted as a file, it just may not be most efficient interface.
587.
by naasking 1y ago
Brushing your teeth is not central to your life, recalling facts correctly is, and a 30 fold improvement in the latter very well could change your life. I'll leave it to you to figure out which is a better analogy to RAG.
588.
by naasking 1y ago
It is true sometimes, but other times it saves hours. We're all still in the learning stage of how best to use these new tools, and their capabilities are growing constantly.
589.
by naasking 1y ago
A 30 fold improvement seems a tad more than incremental.
590.
by naasking 1y ago
UNIX has a bunch of standard tools actually, that's what made it UNIX. How is this any different?
591.
by naasking 1y ago
> the core insight here is actually: if embeddings are generated by layers within the LLM, it makes no sense to convert them back to natural language, just for another LLM to compress those tokens back to embeddings. Doesn't this ti
592.
by naasking 1y ago
"You don't need plugins to do useful work with Datastar", and "You don't need these plugins for most work" are both perfectly reasonable interpretations.
593.
by naasking 1y ago
> By adding a Pro subscription, what's the incentive to work on core? The article literally says that they consider core complete after numerous rewrites and optimizations. The whole point of a plugin architecture is that extensions
594.
by naasking 1y ago
They're not charging for the core, just for some plugins that many will not even use. I really don't see the problem.
595.
by naasking 1y ago
> I am getting really mixed signals here. First the Pro features used to be free (libre), but now they no longer are. That's not my understanding. From what I read, the open source "features" were incomplete, the Pro versi
596.
by naasking 1y ago
I recently installed it on a 13 year old old MacBook, haven't had any issues.
597.
by naasking 1y ago
People can develop open source equivalents you know, you're not required to use the pro version to get a certain feature. From my understanding, datastar was designed to be entirely modular and extensible.
598.
by naasking 1y ago
> The server deciding what to replace reminds me of some old (dangerous, I think) patterns like returning actual JS from the server which the client then executes. Basically every single web page on the modern web has the server returnin
599.
by naasking 1y ago
I thought alpine ajax did OOB on any ids returned in a response. Edit: right, as long as the element has x-sync on it, it will receive any OOB updates from any response.
600.
by naasking 1y ago
Let's have scattered bits of JS inject HTML instead, that'll fix it!

More ›