IanCal

32 ms·

571.
by IanCal 1y ago
This is a very valid point and importantly one of the more detailed issues. So it's a good one to start with when arguing against the OSA - you say harm but what does that mean? What must sites assume it could mean? And examples of h
572.
by IanCal 1y ago
Age checks are not fundamentally required by the OSA. It's really, really important that if you want to argue against it you argue against what's actually in it.
573.
by IanCal 1y ago
Again that's a second order thing as I said, and those are choices by reddit - not mandated by the OSA. > When you have to prove you're a child you have to prove you're and adult. Again, not required by the OSA.
574.
by IanCal 1y ago
> You are covering for a bad law Read the post again. I am not. I am trying to show those against it that they are failing to argue effectively against it.
575.
by IanCal 1y ago
> but it wasn't their responsibility to prevent children from looking at their magazines They weren't made to guarantee no child could peek at them, no, but they do have age restrictions that are followed (a child who picks
576.
by IanCal 1y ago
I'm not a fan of the OSA but proponents of it will *keep winning* if you *keep misrepresenting it*. You can, and should, argue about the effects but the core of the OSA and how it can be sold is this, at several different levels: One,
577.
by IanCal 1y ago
Not sure the majority of projects do anything valuable at all.
578.
by IanCal 1y ago
I’m not saying they’re hard I’m saying they’re common problems that don’t need solving each time. I don’t re-solve title casing every time I need it. > Because you have control over the abstractions. And depending on what you’re using
579.
by IanCal 1y ago
This has the side benefit of likely being easier to navigate for humans too. The less I need to keep in my head to figure something out the better.
580.
by IanCal 1y ago
How do you pull out the call? Parse the response? Deal with invalid calls? Encode and tie results to the original call? Deal with error states? Is it custom work to bring in each new api or do you have common pieces dealing with, say, rest
581.
by IanCal 1y ago
I disagree. I think it’s worthwhile trying these things out to see what’s working and what isn’t. We can sit and think and discuss all we want but sometimes you need to build something and try.
582.
by IanCal 1y ago
The common but lazy joke. This is about connecting agents which have a specific set of requirements about the apis they can call and existing APIs.
583.
by IanCal 1y ago
> All your data is stored locally on your device, and your audio goes directly from your machine to your chosen cloud provider (Groq, OpenAI, ElevenLabs, etc.) or local provider (Speaches, owhisper, etc.) Their point is they aren’t a mid
584.
by IanCal 1y ago
Aside from noting that reviews are not perfect and increased attacks is a risk anyway - the other major risk is running code on your dev machine. You may think to review this more for an unknown pr than an llm suggestion.
585.
by IanCal 1y ago
But fundamentally it requires that it can actually see the thing it’s trying to fix. Lots of these models can essentially barely see.
586.
by IanCal 1y ago
Lots of kids will get this wrong, I don’t know what age you’re thinking of here. They need years of direct coaching to get to words, what stars are, how to hold and move a pen, how to count… Comparing physical drawing to these models is fra
587.
by IanCal 1y ago
Imagine being asked to draw what the op said, but you couldn’t see what you’d drawn - only a description that said “a man and a woman in a Honda “ Asked to draw a new picture with the history of : Draw a picture of a man in the driver seat
588.
by IanCal 1y ago
The problem here is there is a big bit of legislation and people are focussed on one outcome of it, which is what you will have heard. It’s a bill about safety online. The onus is moved to the provider to mitigate harms or decide they don’t
589.
by IanCal 1y ago
The image encodings often don’t have positional information in them very well.
590.
by IanCal 1y ago
Or they simply don’t have that information. OpenAI models have done badly traditionally on placement because the encoding of the image doesn’t include the information very well. Gemini is better as it seems to be passed pre segmented images
591.
by IanCal 1y ago
This isn’t about presenting a document to enter a forum.
592.
by IanCal 1y ago
People dedicate their lives to making realistic paintings despite being able to buy a far more accurate camera for a few hours of work. I’m not hugely convinced that we should worry about work to stay alive and sheltered.
593.
by IanCal 1y ago
That’s not what it’s primarily about, but is the more visible aspect. Lots of the rest is ensuring moderation and the like is supported, scanning for csam, etc. where the risks are higher.
594.
by IanCal 1y ago
You can argue the intentions here but this isn’t different from what’s been said by them before - the public line is that it’s about safety, and while children add additional things to do, its not a child safety bill and it’s focussed on la
595.
by IanCal 1y ago
Is it easier to spoof GPS transmissions such that the phone will calculate it's at a specific coordinate? Rather than just check where they are reporting?
596.
by IanCal 1y ago
As someone who loves the show this is very fun to see. I'm impressed at the number of correct answers most get. Helpfully as well there is a current season on, and while the total questions wouldn't be enough to fully validate res
597.
by IanCal 1y ago
I’ve not experimented with the recent models for this but older Gemini models were awful for this - they’d lie about what I’d said or what was in their system prompt even with short conversations.
598.
by IanCal 1y ago
Density doesn’t mean specifically per unit area though. In fact the default meaning without context to me would be volumetric.
599.
by IanCal 1y ago
But vitally they are not explicitly shown the letters individually and so “count the letters” is a much harder problem to solve than it is for us.
600.
by IanCal 1y ago
The tokenisation means they don’t see the letters at all. They see something like this - to convert just some tokens to words How many 538 do you see in 423, 4144, 9890?

More ›