> This is a poor implementation not a framework problem lol.
I've seen literally pixel-identical error messages taking over the entire screen when there is a JS error, in exactly that way, across countless of websites. If it's not a framework, it's a library in wide use, because tons of websites have exactly the same issue.
Read the context first. Initially, we had HTML + JS pages where one function can have one error, and the page keeps on working, because the error didn't block the entire screen.
Now we have some popular framework/library that instead seems to surface these errors that aren't sitting in the top-level scope of the JS application, yet the errors ends up blocking the entire application.
That means it's bad, yes. And it's not any language, we're specifically talking about JS in browsers here.
Modern JS frameworks can absolutely catch and isolate errors.
In those cases you’d only see the error in the specific component that had the error.
Or if they’re a good dev you as a user won’t even know! You’ll see fallback content.
An error blocking the full screen is a failure on the developer not the tools.
> An error blocking the full screen is a failure on the developer not the tools.
I don't see how this is possible, given that the default experience with HTML and JavaScript isn't "Tear down the entire page to display an error", but something that's seemingly becoming more and more common to actually come across. Guess I need to chase down what library/framework this is coming from, since seemingly I'm not able to describe the issue properly.
You don't need the "modern JS framerwork" to "catch and isolate errors" because that's a feature JavaScript comes with by default in browsers! You literally have to add code to make it worse, which seems to be happening in some framework/library.
If you're manually adding logic for doing "error in component doesn't break entire application" then you're already working against what the environment gives you by default for free.
This will change too man. Maybe I am in a bubble but with how fast things are changing, it won’t be too long before the bubble becomes reality.
Either way we should be doing experiments on the actual capabilities of AI not about the stupidest possible way to use AI because it helps validate your own negative bias against AI.
Additionally as software engineers using agentic AI… which HN basically is… this experiment is not at all relevant in the context of where it is posted. We ALL use agentic ai and we all have the agent use surgical tools for editing. Don’t you find it strange that despite the fact we all do this, HN is full of rabid engineers gobbling this paper up as validation despite complete lack of relevance?
> This will change too man. Maybe I am in a bubble but with how fast things are changing, it won’t be too long before the bubble becomes reality.
You can’t get mad at an experiment for not happening in the future.
> Either way we should be doing experiments on the actual capabilities of AI
They simulated common end user behavior
>because it helps validate your own negative bias against AI.
We’ve gone from “this study is flawed because language models don’t do that” to “this study is flawed because while language models do do that, I don’t think that they will in the future” to “data that could support a bias other than my own is bad”
> You can’t get mad at an experiment for not happening in the future.
I’m more getting mad at this sentence not making any sense. I’m disappointed at this experiment for not testing the actual capabilities of an LLM. Comprende?
> They simulated common end user behavior
Not the way you use it. And not the way it will be used.
You love it because you want it to stay this way so you can forever believe AI will never be better than you.
Bro the reality is unfolding as you speak. It’s like humanity just discovered guns but hasn’t discovered the bullets and your saying guns are useless because most of humanity hasn’t figured out bullets yet.
> We’ve gone from “this study is flawed because language models don’t do that” to “this study is flawed because while language models do do that, I don’t think that they will in the future” to “data that could support a bias other than my own is bad”
This is a flat out lie. Models DO do that. The only fucking argument you have is that non technical and average laymen people edit documents the wrong way while all people who use agentic AI as adepts use it the correct way. Like are you fucking kidding me?
The only change I acknowledge is your grandma copies and pastes essays into ChatGPT while YOU don’t. You go pretend you live in that reality where the bullets will never appear.
and I haven’t challenged that this doesn’t sit comfortably with your opinions about the future. I believe that you feel that way, nobody is arguing that you don’t
First off, It’s good to study all kinds of things isn’t it? Even if it’s not strictly practical.
Second, and more importantly these AI tools are EVERYWHERE right now. The effects of people using them for work can be seen throughout many industries and workplaces.
So I think studying how these models perform in the vast majority of use cases is not only a good idea, but it’s actually really important.
Even if you’re strictly pro-AI and believe it is the future, a study like this can help you explain to laymen why they need the harnesses you’re so in support of.
> First off, It’s good to study all kinds of things isn’t it? Even if it’s not strictly practical.
Course it is. But the conclusion everyone is coming to is that LLMs are garbage and can’t be used because of 25 percent degradation which is not in line with reality.
> Second, and more importantly these AI tools are EVERYWHERE right now. The effects of people using them for work can be seen throughout many industries and workplaces.
At 25 percent degradation these tools would not be everywhere. They are everywhere because it’s not actually used that way.
> So I think studying how these models perform in the vast majority of use cases is not only a good idea, but it’s actually really important.
I have less of a problem with this study and more about the interpretation of this study.
> Even if you’re strictly pro-AI and believe it is the future, a study like this can help you explain to laymen why they need the harnesses you’re so in support of.
I’m not pro-AI. I’m anti AI. I fucking hate fucking AI.
What I’m angry at is this delusional denial of reality. This experiment is very obviously not accurate yet people are using this study as a headliner to promote an anti AI agenda.
I don’t like AI but that’s different for lying to myself or trying to say AI sucks at something when it is in fact superior then us in this respect.
But actually creating music or playing an instrument is much more rewarding. The time commitment is part of it, the journey is the destination and all that.
They can’t even deliver their own flagship products without bugs, and terrible UX. So I’m doubtful of their abilities.
These are also the same companies allowing their AI to make decisions in war, have no qualms about the mental issues they’re causing in people, and have abused workers in 3rd world countries for years.
But you think they’re holding out on “destroying the software industry” out of the goodness of their hearts? Come on
I think his reasoning was pretty clearly presented as not the goodness of their heart but rather the medium to long term predicted outcome on their bottom line. Ultimately failing or getting tangled up with regulators any more than necessary is to be avoided. If you move too early and it chases people away from your platform which undermines your ability to keep innovating then a competitor who held back will ultimately eat your lunch.
But then there is no safe way for them to "mortally wound" the software industry. The full argument is moot.
I would add there are more reasons why this wouldn't work: costs due to OOM more usage, adoption/AI backlash, adversarial environment, players with big head starts (Google).
Yes, I believe the original commenter made that exact point.
You don't need to personally win in order to mortally wound someone. It can be informative to speculate about whether or not something is possible regardless of it being strategically advisable in the current context.
My hope would be that since openclaw is communicating with me to my personal device, where I have all noise filtered, it would be a bit better.
I also know it can integrate with TickTick, which has been a huge change for me with task management. Then again - in my experience whatever tool I use to keep track of stuff only works for as long as it's a novelty, but 3 months is a record anyway.
The thing is - when I receive a message and I'm not in the headspace to answer, I close the notification and forget about it. My expectation would be openclaw reminding me that I still haven't replied to this person about that thing.
Obviously, there's a million ways to do it that don't require openclaw. Obviously there's a million things that I won't be able to grant openclaw access to (e.g. company jira or slack). And obviously, I don't want it evaluating every single of my personal messages. But I think there is a reasonable middle ground where it can work well. But I don't yet know how to reach it
It's a funny way of imposing a very large fine. Make the service only available during predefined "visitation hours", prevent updated learning except from resources available in the prison, restrict speech and actions according to prison rules etc.
I mean yeah, obviously. You can’t trust a word out of either of their mouths.