• 0 Posts
  • 31 Comments
Joined 1 year ago
cake
Cake day: August 15th, 2023

help-circle

  • I’m not redefining anything, I’m just pointing out that intelligence is not as narrow as most people assume, it’s a broad term that encompasses various gradations. It doesn’t need to be complex or human-like to qualify as intelligence.

    A single if statement arguably isn’t intelligence, sure, but how many if statements is? Because at some point you can write a complex enough sequence of if statements that will exhibit intelligence. As I was saying in my other comments, where do we draw this line in the sand? If we use the definition from the link, which is:

    The highest faculty of the mind, capacity for comprehending general truths.

    Then 99% of animal species would not qualify as intelligent.

    You may rightfully argue that term AI is too broad and that we could narrow it down to mean specifically “human-like” AI, but the truth is, that at this point, in computer science AI already refers to a wide range of systems, from basic decision-making algorithms to complex models like GPTs or neural networks.

    My whole point is less about redefining intelligence and more about recognizing its spectrum, both in nature and in machines. But I don’t expect for everybody to agree, even the expert in the fields don’t.



  • I would put it differently. Sometimes words have two meanings, for example a layman’s understanding of it and a specialist’s understanding of the same word, which might mean something adjacent, but still different. For instance, the word “theory” in everyday language often means a guess or speculation, while in science, a “theory” is a well-substantiated explanation based on evidence.

    Similarly, when a cognitive scientist talks about “intelligence”, they might be referring to something quite different from what a layperson understands by the term.


  • In a way, yes, if you frame it right. To simplify, you’re basically asking “is a calculator intelligent?”, right? While it’s an inanimate object, you could say that, in a way, it acquires knowledge from the buttons user presses and it applies knowledge to provide an output.

    “But that’s not making decisions, it’s just circuits!”, you might say. To which I might reply “Who’s to say that you’re making decisions? For all we know, human brains might also just be very complicated circuits with no agency at all, just like the calculator!”.

    IIRC, in his book The Singularity Is Near, Ray Kurzweil even assigns certain amount of intelligence to inanimate objects, such as rocks. A very low amount of course, and it might be a stretch, but still.

    So yeah, it’s really hard to draw a line for intelligence, which is why there’s no firm definition and no consensus.


  • Of course there are various versions of NPCs, some stand and do nothing, others are more complex, they often “adapt” to certain conditions. For example, if an NPC is following the player it might “decide” to switch to running if the distance to the player reaches a certain threshold, decide how to navigate around other dynamic/moving NPCs, etc. In this example, the NPC “acquires” knowledge by polling the distance to the player and applies that “knowledge” by using its internal model to make a decision to walk or run.

    The term “acquiring knowledge” is pretty much as subjective as “intelligence”. In the case of an ant, for example, it can’t really learn anything, at best it has a tiny short-term memory in which it keeps certain most recent decisions, but it surely gets things done, like building colonies.

    For both cases, it’s just a line in the sand.





  • To be fair, PS2 emulation is still not that great, but I guess it’s due to sheer amount of games for that system. Last summer I decided to check the PS2 emulation after 10 year break and 2 out of 3 games I tested didn’t work properly. Granted, those are kinda niche games (Transformers (2004) and Free Running), but compatibility still needs work. Hardware requirements are decently low for the games that do work, though.




  • Hey, I saw that you added more content to the comment that I responded to, that wasn’t there when I was composing my response. And seeing that content, I think I understand where the confusion is coming from.

    If that screenshot is yours and you think those are the permissions, I don’t think that’s the case. That looks like a screenshot from an app store where it just lists what data the app might be using and not the permission system. It’s just a list of categories of data that may or may not be collected if you use the app, which must be disclosed by the developer. You can’t agree or disagree to those things from the OS side, because that’s all that happens on the developer’s side. In case of FB, you might be able to opt out of those things in their settings, but I wouldn’t bet on it, cause that’s their bread and butter.

    For these things it doesn’t matter if you, for example, gave them direct access to the mic or uploaded the audio file, they will process the audio file and gather as much useful information as they can.

    In fact, if you don’t give it any hardware permissions, they will still be able to gather some information, for instance from the Personal Info category (email address, sexual orientation, and home address, etc.) because you enter that info on registration or they infer it from your usage. The OS can’t do anything about that. As long as you use the app and interact with it you give them information, what you clicked, which posts you liked, what you commented and so on.

    When it comes to OS, you can individually (separately) give permission for mic, camera, location data, file storage, contacts info, etc. Most of the things listed in the “Data collected” panel doesn’t even come from your phone hardware.

    Let me know if I have now understood you correctly.


  • When I said “well almost” I meant the impossible case in the second point. Otherwise, everything is implemented as a I listed. What kind of Android do you use that you haven’t seen these features? This granular permission system has been the standard since Android 11.

    In iOS it’s implemented in a very similar manner, but I don’t use it as often to describe it in as much detail as with Android.

    The OS can create the file and then hand it off to the app.

    That is also implemented, but is a separate API, storage access. You’re free to upload any file you like if the app requests it. You can create the file with any voice recorder of your choosing. Although I can’t imagine a scenario where Facebook would request a voice clip. When it’s requesting the mic it’s usually for live audio, like calls.

    How is a user to know if something was captured when the screen was off?

    It’s true, if you gave the app permission to use mic whenever the app is running, it can in theory quietly use mic in the background. If you start a call and lock the screen, the call will continue in the background. Not sure if there are any safety measures implemented for that. But if the case was of a routine sneaky mic spying, it will become obvious fast, due to battery drain and network usage.

    still don’t have basic controls like restricting network access

    There are some network controls, like restricting background data usage (depending on Android version/implementation). But yes, there’s still no granular network permission system, you have to manually go into setting to turn on restrictions. Thought to fair, there isn’t a consumer OS out there that lets you easily restrict network access to a certain app, even on desktop (correct me if I’m wrong). And I can see why, it would be counterproductive for vast majority of users to manually give network access to each app they install, when the whole point if the device is to have apps that have network access.


  • All of those things are implemented in modern Android. Well, almost.

    • Whenever the app wants to use microphone an OS popup asks you if you want to give the app permission to use the feature. The options are “when using app”, “only this time” (it will give the app one-time-use access to the mic) and “never”. If you click the 1st or 3rd options, you wouldn’t see the popup again and you’ll have to change the permission from settings. If you choose the 2nd option, you can manually choose to give permission each time it’s requested.
    • This is impossible? The OS can either let the app use the mic or not, it can’t tell what the app is doing with the mic. Unless you mean give a one-time permission this time, but not in the future, then we covered that in previous point.
    • Android always shows a green indicator on screen (upper right corner) when any app is using the microphone or camera API. Well, almost always, some system apps might not trigger it. But if you want to see which app is using mic/camera you can tap the indicator.

  • “E Ink” is a company, producing displays with a very specific proprietary technology. I think you mean to say more “e-paper”, which is a generic term for “paper-like” displays. And unfortunately, right now the only real competition is RLCD (reflective LCD), which is arguably not paper-like enough to qualify. Yes, it’s reflective, but other than that, it’s just a higher density Game Boy screen. Which is great and all, but it can’t compete with E Ink in a lot of aspects. It doesn’t have retention, you gotta power the screen, so no signage and stuff. It has LCD-grade bad viewing angles.

    RLCDs are cool for certain uses though. For example, I sometimes choose to play on my AGS-001 over my Analogue Pocket if I’m outside in bright daylight.


  • What truly blows my mind is the amount of requests the 1st party Reddit app sends home. Back when I was using Sync I still had the app installed, but then I set up AdguardHome and saw that my phone was spamming requests. Checked the logs and found out that the 1st party app, which I wasn’t even using for months, was “phoning home” literally every 10 seconds! Besides privacy concerns, that can’t be good for battery life. Nuked the app then and there. I’ll take the nagging, thank you.