• 17 Posts
  • 306 Comments
Joined 1 year ago
cake
Cake day: June 4th, 2023

help-circle




  • Ehh. That’s like accident billboards. I maintain that most people don’t know they can block ads, and a large part of the masses who have heard of it think it’s complicated or too hard for them.

    With ad blocking I have a small tension that if I know a sort of thing exists, I presumably will find it when I search for it. So I don’t want another vacuum ad.

    If I don’t know something exists then I have to stumble on it somehow.

    The bigger problem would be if they didn’t block their own ads. I honestly didn’t even know they did ads so my blocking, of which they’re a part, apparently is working.



  • Yes definitely. Many of my fellow NLP researchers would disagree with those researchers and philosophers (not sure why we should care about the latter’s opinions on LLMs).

    I’m not sure what you’re saying here - do you mean you do or don’t think LLMs are “stochastic parrot”s?

    In any case, the reason I would care about philosophers opinions on LLMs is mostly because LLMs are already making “the masses” think they’re potentially sentient, and or would deserve personhood. What’s more concerning is that the academics that sort of define what thinking even is seem confused by LLMs if you take the “stochastic parrot” POV. This eventually has real world affects - it might take a decade or two, but these things spread.

    I think this is a crazy idea right now, but I also think that going into the future eventually we’ll need to have something like a TNG “Measure of a Man” trial about some AI, and I’d want to get that sort of thing right.



  • I think it’s very clear that this “stochastic parrot” idea is less and less accepted by researchers and philosophers, maybe only in the podcasts I listen to…

    It’s not capable of knowledge in the sense that humans are. All it does is probabilistically predict which sequence of words might best respond to a prompt

    I think we need to be careful thinking we understand what human knowledge is and our understanding of the connotations if the word “sense” there. If you mean GPT4 doesn’t have knowledge like humans have like a car doesn’t have motion like a human does then I think we agree. But if you mean that GPT4 cannot reason and access and present information - that’s just false on the face of just using the tool IMO.

    It’s also untrue that it’s predicting words, it’s using tokens, which are more like concepts than words, so I’d argue already closer to humans. To the extent it is just predicting stuff, it really calls into question the value of most of the school essays it writes so well now…


  • Well, LLMs can and do provide feedback about confidence intervals in colloquial terms. I would think one thing we could do is have some idea of how good the training data is in a given situation - LLMs already seem to know they aren’t up to date and only know stuff to a certain date. I don’t see why this could not be expanded so they’d say something much like many humans would - i.e. I think bla bla but I only know very little about this topic. Or I haven’t actually heard about this topic, my hunch would be bla bla.

    Presumably like it was said, other models with different data might have a stronger sense of certainty if their data covers the topic better, and the multi cycle would be useful there.


  • You don’t have to. Thousand of people who know what they’re doing does. But why would I trust any of them? I’m pointing out you have to choose who you trust, and from the history with the makers of Vivaldi, I trust them. Same as I don’t trust Google given their history.

    Of course, I’m screwed anyway because there’s not reasonable competition in the phone space, and I have to use Microsoft products for work, and… {insert a dozen more things here}. Given all that, I’d like the browser that works better for me.


  • Although you could take into account what the makers are telling you. You have to trust someone, and at least to my knowledge, Google fails and it’s all over the news, Vivaldi has not. It’s not like I can validate the Firefox source either, I’m just trusting the website I download it from, or more likely my distro packaging. And people do look at call outs browsers make etc.


  • My problem is - last time I looked, which was a while ago to be fair - there weren’t good tab management plugins that also supported tab title search, a list of tabs to easily close ones I didn’t need anymore with ctrl+click or shift+click, no session management, problems with cross window tab viewing/searching, no tab stacks, and now workspaces are kind of awesome for me too.

    I’m not saying there aren’t extensions for each thing, I’m saying I could NOT get them all to work together, and have a fast performant browser without weird hangs, and the UI was kind of all over the place and hard to remember cause none of the extensions were designed to work together from what I could tell.

    What I don’t get is why Vivaldi didn’t code on top of Firefox, but I think it’s because there are sites that work in chromium and don’t in Firefox, and fail silently - and just like in IE6 days, they’re sites like my parents retirement site, they can’t NOT use them.



  • Mostly because the browsing experience IMO is much much worse with Firefox. I tried extensions to get functionality back, it made it worse - slower, buggy, extensions would stop being developed etc. I wish Firefox was better, I really do. But IME it’s frozen functionality like it’s 2010 or so. Like, they have tabs, who hoo. I really find save/restore, multi window control, tab stacks, sessions, workspaces, and easy UI config pretty important in day to day use. That said, I also think ads are a deal breaker, but I really wonder if this won’t bring back some of the ad-blocking proxies you run locally or something.

    Or, someone forks chromium to keep Manifest v2 or whatever.


  • Not so much chrome, but many browsers (like my favorite Vivaldi) are chromium based. I wish they’d just keep uBlock going in the chromium rebuilds, but IDK if that’s possible. Seems like it should be to me though.

    Also, we switched at work from Firefox because somehow they broke system level updates a few years ago, and nothing I could do was able to figure out why their installer stopped working without first having someone run the uninstall graphically to update to the new version. It would just say Firefox wasn’t a valid windows exe till I manually removed it. And even the Mozilla Enterprise list seemed flummoxed. Honestly, I think they should have reverted the installer change, or even just use a standard installer that doesn’t have this problem, but hey.