Rudism's recent activity

  1. Comment on What is the oldest TV show you *actually* enjoy? in ~tv

    Rudism
    Link
    Excluding shows I grew up with (like the Star Treks) to avoid nostalgia contamination, here are the oldest shows I've watched more recently (say the last 10 years or so) and have enjoyed: Monty...

    Excluding shows I grew up with (like the Star Treks) to avoid nostalgia contamination, here are the oldest shows I've watched more recently (say the last 10 years or so) and have enjoyed:

    Monty Python's Flying Circus (1969) - I'd seen a lot of the movies when I was younger, but it wasn't until recently that I found the old TV series had a blu-ray release, so I bought it and ripped the whole thing to my Jellyfin server and have been enjoying the hell out of it.

    Survivors (1975) - British show that follows a group of people in the post-apocalyptic aftermath of a worldwide pandemic. I had watched this for the first time probably in the early 2000s but re-watched during covid times and liked it a lot more this second time around.

    Sherlock Holmes (1984) - The British series starring Jeremy Brett is bar none the greatest Sherlock Holmes adaptation to ever hit the screen. Still great. This is kind of cheating because I did watch some of this with my mom (who was obsessed with it) when I was a kid, but only really caught a smattering of episodes here and there and Sherlock Holmes wasn't really something I was interested in at the time. It's not until recently I started watching it in earnest all the way through.

    Twin Peaks (1990) - I watched this for the first time a few years ago, followed by the 2017 follow up series, and have to say I have come to regard that as the single greatest television media consumption experience of my life.

    4 votes

  2. Comment on Any Letterboxd users here? in ~movies

  3. Comment on VR headsets, what do I need to know? in ~tech

    Rudism
    Link
    Here's a list of VR headsets I've used and owned and my comments on them: First experience was on a friend's HTC Vive around when they first hit the market. It was pretty amazing, but was never an...

    Here's a list of VR headsets I've used and owned and my comments on them:

    • First experience was on a friend's HTC Vive around when they first hit the market. It was pretty amazing, but was never an option for me because 1) I use Linux, 2) I didn't and don't own a graphics card, and 3) you basically needed to set aside a whole room for VR because of the requirement to set up cameras everywhere for tracking.

    • Second experience was buying the original PSVR. Played it a ton, loved it. It also uses external cameras for tracking, but just one that you can set on top of your TV assuming you have enough room there. As a result, the tracking is quite bad--it loses track of your hands a lot and you can't really move around much or you step out of the camera's visible area. The PS Move controllers are also not great, since they only have buttons and triggers and no joysticks.

    • Next I got a first-gen Quest, and it made me truly realize just how terrible the PSVR tracking was. The Quest is a standalone headset (basically a phone processor stuffed into some goggles) and while it can't reach anywhere near the same graphical quality as the PS4, the resolution is better, and having the tracking cameras embedded into the headset itself means it's far more convenient and the tracking is night and day more reliable. Meta doesn't support the original Quest anymore and newer games and stuff won't even run on it, so it's not worth getting now even if you can find one dirt cheap.

    • Current headset is a Quest 2 which is pretty similar to the first Quest but with beefier specs and a better resolution. I highly recommend it, it's basically the only option for a good, affordable standalone headset right now.

    I haven't tried the PSVR2, but I understand it also has cameras in the headset so I assume the tracking issues of the first PSVR are fixed, and the PS5 can probably push some pretty jaw-dropping graphics to that thing. The main downside I can see is it's still tethered to the PS5 via a cable, but if that doesn't bug you and you have a PS5 already anyways then maybe that's a better option than the Quest 2.

    As for other PCVR headsets, forget about it on Linux. There is basically zero support for VR gaming using a Linux PC--your only viable option is probably to dual boot Windows. The Quest and Quest 2 also have various ways to plug your headset into a computer or even ways to use it for PCVR wirelessly, but none of that works on Linux, at least as of the last time I looked into it.

    2 votes

  4. Comment on Reddit is going to enforce rate-limiting the API's free tier as well as charging for higher rates in ~tech

    Rudism
    Link
    I haven't noticed any indication in the various announcements and posts that creating an app and letting users provide their own API key isn't a viable route to avoiding these massive API bills on...

    I haven't noticed any indication in the various announcements and posts that creating an app and letting users provide their own API key isn't a viable route to avoiding these massive API bills on the part of developers. 100 queries per minute seems like it wouldn't pose a problem for an individual user using an app with their own API key and logging in via Oauth.

    Am I missing something? Is creating API keys going to be locked down and unavailable to the general public?

    6 votes

  5. Comment on ROT13 + base64 on GPT4 = reliable hallucinations in ~tech

    Rudism
    Link Parent
    Kind of true, as far as the base training data goes, but it's fairly easy and not uncommon to feed internet search results into the LLM along with your prompt which would generate output using...

    LLM's currently don't see a live version of the Internet and they won't amplify what they didn't see yet.

    Kind of true, as far as the base training data goes, but it's fairly easy and not uncommon to feed internet search results into the LLM along with your prompt which would generate output using information newer than what it was trained on.

    So it seems like until something big changes, we should probably pay more attention to social media problems than LLM-LLM replication problems.

    I think that it's almost inevitable that the biggest problem social media platforms are going to face in the very near term is the mass proliferation of LLM-generated content for subversive purposes.

    2 votes

  6. Comment on What's your p(doom)? in ~talk

    Rudism
    Link
    I don't think we're in any danger of a true existential crisis that isn't almost entirely human generated any time soon (like within the lifetime of anyone currently alive as I'm posting this, or...

    I don't think we're in any danger of a true existential crisis that isn't almost entirely human generated any time soon (like within the lifetime of anyone currently alive as I'm posting this, or probably even the lifetimes of their grandkids). I do believe that the internet is doomed on a much smaller timescale though--maybe even within the next few years. I think it's almost inevitable that political and corporate interests are going to start pumping out low-effort AI-generated content (either to push ideological views or for ad revenue) at an unprecedented scale to the point where genuine human-generated content is going to be completely obscured and un-discoverable (even moreso than it already is).

    I foresee AI systems set up to automatically register domain names, spin up blogs, wikis, product reviews, all uniquely branded and SEO optimized, and continue generating ad-supported content at a regular pace, all with zero human oversight or interaction beyond an initial prompt giving it a topic plus maybe a few directives or ideas that the owner wants to focus on. I expect an entire ecosystem of garbage AI content-farm products and services will arise where people can start generating these things en-masse. Type in a few topics or let the AI choose for you, pay some monthly fee, start earning ad revenue.

    The law of diminishing returns means this will probably cap out and maybe even start dying down after some time, but by then the damage will have already been done and the "dead internet theory" will have already been realized.

    4 votes

  7. Comment on ROT13 + base64 on GPT4 = reliable hallucinations in ~tech

    Rudism
    Link Parent
    I have no special insight into the models, but if they were trained on general content off the internet I suspect they'd run into enough base64 and rot13 encoded content either preceded or...

    I have no special insight into the models, but if they were trained on general content off the internet I suspect they'd run into enough base64 and rot13 encoded content either preceded or followed by the decoded equivalent that the rules for those specifically are something it could learn as part of its model, without the need for any "intuition" analog.

    The examples in its training data of rot13 and base64 combined to encode content is likely significantly smaller, which I'd guess is why it's so much easier to invoke these hallucinations that way.

    As a mostly unrelated aside, it's interesting to think about taking it a step further down the road: Future LLMs that are training themselves off data like this thread will have examples of rot13 and base64 encoded content alongside the incorrect hallucinated translations, which could possibly lead to more "confusion" in its data model, turning this into a self-perpetuating problem. I'm interested to see what (if anything) future research will come up with to avoid training new LLMs on the output of old LLMs as the volume of AI-generated content slowly grows to outpace and overshadow human-generated content.

    2 votes

  8. Comment on What games have you been playing, and what's your opinion on them? in ~games

    Rudism
    Link Parent
    TotK is all I've been playing since it came out. I'm loving it too, but I'm not getting the same sense of exploration that I did with BotW. For example, getting all of the Shiekah towers was a...

    TotK is all I've been playing since it came out. I'm loving it too, but I'm not getting the same sense of exploration that I did with BotW. For example, getting all of the Shiekah towers was a pretty big multi-day undertaking for me and felt extremely satisfying to finally complete in BotW; in comparison, the sky towers in TotK I had done within the first few hours and it felt like the whole surface map was pretty much open to me at that point.

    I think the depths might have scratched that itch a bit more, if only they weren't so darn scary.

    9 votes

  9. Comment on Neeva.com is shutting down in ~tech

    Rudism
    Link Parent
    I wouldn't guess that upload speed would have that big an effect on latency (I don't imagine much data is being sent to the servers per query), but I suppose it's possible. Depending on what you...

    I wouldn't guess that upload speed would have that big an effect on latency (I don't imagine much data is being sent to the servers per query), but I suppose it's possible. Depending on what you mean by indirection that sounds like the more likely culprit to me. Also it's a Python app so not the most efficient, running on a slower machine like a Raspberry Pi could also be a speed bottleneck.

    1 vote

  10. Comment on Neeva.com is shutting down in ~tech

    Rudism
    Link Parent
    Latency hasn't really been a problem for me. I haven't actually timed it but I don't perceive searches as taking any longer than second or two at most to show me results, pulling data from a...

    Latency hasn't really been a problem for me. I haven't actually timed it but I don't perceive searches as taking any longer than second or two at most to show me results, pulling data from a couple main engines (google and bing) plus a few instant answer sources (wikipedia, wikidata, dictzone).

    2 votes

  11. Comment on Neeva.com is shutting down in ~tech

    Rudism
    Link Parent
    I'm not the original commenter but I also self-host a SearXNG instance. Mine's running on a VPS, and I put it behind a webauthn proxy so it's not publicly accessible. Kagi isn't currently one of...

    I'm not the original commenter but I also self-host a SearXNG instance. Mine's running on a VPS, and I put it behind a webauthn proxy so it's not publicly accessible. Kagi isn't currently one of the supported engines but there's an open issue to add it so it might be supported at some point in the future, assuming Kagi has an API that you can use with an account there.

    The main customization point in SearXNG is you can pick which combination of the supported engines to source results from. However, since a lot of it involves scraping, some engines don't work consistently (DuckDuckGo in particular seems to actively work against scrapers and is constantly breaking). It does sort of have something like Kagi's "lenses," but it's a pre-baked list (web, images, news, videos, stuff like that) and each one has a specific list of supported engines that you can enable or disable. There's nothing similar to Kagi's ability to block or boost specific domains in results, you just get what your selected engines return without any ability to really refine beyond that.

    4 votes

  12. Comment on What games have you been playing, and what's your opinion on them? in ~games

    Rudism
    Link Parent
    I put a few hours in over the weekend and am loving it so far. The sense of exploration and discovery was what really drew me into BotW and this one doubles down on that with the addition of the...

    I put a few hours in over the weekend and am loving it so far. The sense of exploration and discovery was what really drew me into BotW and this one doubles down on that with the addition of the sky islands and chasms.

    So far the caves on the main world map don't seem all that interesting to me, but maybe I just haven't figured out what they're for yet. Likewise the whole vehicle building system feels a little gimmicky and chore-like to me so when I find a puzzle that was clearly designed to be solved by building vehicles I feel compelled to skip it if there's no obvious non-vehicle way to solve it.

    2 votes

  13. Comment on Comic fans GOTG 3 thoughts? in ~movies

    Rudism
    Link
    I liked the first two movies quite a bit (I'd say they're my top 2 MCU films, having seen all of them up until Black Widow or thereabouts before I stopped being a completionist). Due to extreme...

    I liked the first two movies quite a bit (I'd say they're my top 2 MCU films, having seen all of them up until Black Widow or thereabouts before I stopped being a completionist).

    Due to extreme superhero fatigue I wasn't excited about this one, but my youngest kid was all aboard the hype train so I ended up seeing it anyway. I was pretty underwhelmed. It's not that it was bad, just wasn't outstanding enough to overcome the general sense of ennui that Disney franchises stir up in me now.

    During the 2010s I couldn't have imagined anything more tedious than the zombie apocalypse craze that all the studios latched onto, turning something I once loved into something I couldn't care less about. But the vigor with which Disney is milking Marvel and Star Wars has made me realize how naive I truly was back then.

    2 votes

  14. Comment on What are you reading these days? in ~books

    Rudism
    Link
    I read Piranesi by Susanna Clarke. Twice actually--I liked it so much after reading the ebook that I got the audiobook too and listened all over again. I don't think it's one I would recommend to...

    I read Piranesi by Susanna Clarke. Twice actually--I liked it so much after reading the ebook that I got the audiobook too and listened all over again. I don't think it's one I would recommend to many people I know, since it's intentionally pretty slow and repetitive, but I found it fascinating in the same way I find YouTube videos based on the lore of The Backrooms fascinating.

    3 votes

  15. Comment on What programming/technical projects have you been working on? in ~comp

    Rudism
    Link
    Back in the 90s one of the local AM radio stations where I lived had a late night programming block where they played old radio shows from the 1930s to 1950s, stuff like Escape, The Whistler, The...

    Back in the 90s one of the local AM radio stations where I lived had a late night programming block where they played old radio shows from the 1930s to 1950s, stuff like Escape, The Whistler, The Shadow, Jack Benny, Our Miss Brooks, and so on. Every night for years while I was in high school and college I'd go to bed listening to that (followed by Coast to Coast AM with Art Bell).

    Several years ago I discovered that all those same shows were available on archive.org, and due to loopholes in the US copyright system may not be protected by copyright at all, so I spent way too much time downloading, tagging, and cataloging every old time radio show I could find and built an Android app that could stream them on demand. There were already a lot of other apps that offered OTR (old time radio) streaming, but I found each one lacking in some way (either their catalog was lacking, background play was buggy, or it was riddled with ads). I built my app primarily for my own use, but I also added the ability to subscribe for a few bucks a month figuring maybe it could cover its own hosting costs if a few other people were sick of the free garbage apps and were willing to pay for a more premium experience.

    It was chugging along for quite a while, earning a couple hundred bucks a month in subscriptions at its peak which did cover hosting costs plus a bit extra. The problem was that maybe once or twice a year I'd get emails from Google saying I'm using some old API for media playback or billing that was going to be deprecated soon and I needed to switch or else be delisted. Because the app was otherwise so low-maintenance, every time that happened I basically had to re-learn my entire code base and figure out how to swap in the new library and test everything to make sure it was working, which was quite a burden--especially when it came to the billing APIs which I felt were way more complicated than they needed to be. In the end, I decided to shut down the Android app as a result. I refunded all the current subscribers' most recent payments, cancelled all the subscriptions, and de-listed the app from the Play Store--though I left the server running so that anyone who still had the app downloaded could keep using it for free until it actually stopped working.

    I let it sit that way for a while, but recently (due in part to a lot of sad emails I got from people who used the old app) I've decided to spin it back up as a web application, which is now running at radiostasis.com. Professionally I've been a fully backend-only engineer for a long time, so UI/UX is definitely one of my weak points. It's been a fun challenge and interesting learning experience researching the different front-end frameworks that are out now. The last time I did any kind of front end web app work the only real option was JQuery, which I actually liked quite a bit, so seeing all the super-heavy and complex SPA frameworks that people use now is very intimidating. I settled on htmx because it appealed to me the most in its simplicity and lends itself well to server-side rendering.

    One thing the Android app had been missing that I wanted to tackle this time around was brief descriptions of each of the OTR series in my library. I had them tagged with genres, but beyond that there was no real information about the shows. To solve that, I created some scripts that did a few things:

    1. Pulled results from a Google Search API for each show name, presenting me with a list of results, and allowing me to select the most promising of those as "source" links. Mostly these came from archive.org, Wikipedia, and other Old Time Radio websites.
    2. Fetched the content of each "source" link and cleaned it up a bit to remove page boilerplate and reduce the content as much as possible down to details about the show.
    3. Sent the content to the ChatGPT API and asked for a summary of the show (I went through several prompt engineering iterations to get something clean and consistent).

    With over 270 shows, doing the research and writing summaries myself could easily have taken months or years to complete, but thanks to ChatGPT I got descriptions of all the shows in a couple days. I used the newer ChatGPT-4 model so it ended up costing me around $50 in API calls, but I'm super pleased with the results.

    Anyway, that's my primary technical side project right now. I'm much happier writing a web application (currently without any accounts or billing to make things annoying, which I do plan on adding eventually but kind of dreading it) compared to a native Android app, and installed as a PWA it works pretty well on phones too.

    9 votes

  16. Comment on GM killed the Chevy Bolt — and the dream of a small, affordable EV in ~tech

    Rudism
    Link
    My 2003 Hyundai Elantra with around 125k miles on it has been flashing engine lights at me lately (along with a host of other minor-nuisance old vehicle problems), which prompted me to start...

    My 2003 Hyundai Elantra with around 125k miles on it has been flashing engine lights at me lately (along with a host of other minor-nuisance old vehicle problems), which prompted me to start researching cheap, smaller EVs that I could potentially replace it with. Very disappointed by how few options there are, even before learning of this news--the Bolt was one of the leading contenders.

    I don't understand why so many people want to drive tank-sized vehicles. When I have to drive my wife's hybrid Santa Fe I feel like I'm practically driving blind with how high up I am and how poor visibility is in all directions. I guess I can kind of understand wanting more storage space with the back seats folded down if you actually have a regular need for it, but I don't have a need for it--especially not when we already have one SUV.

    Been eying the Mini Cooper SE, mostly out of the desire to show support for smaller EVs and because I don't drive often or far enough that the shorter range would be a problem, but I also might just stick with the old Elantra a while longer.

    3 votes

  17. Comment on What is your most essential pessimistic belief? Conversely, what is your most essential optimistic belief? in ~talk

    Rudism
    Link
    Pessimistically I believe that individual human nature is fundamentally at odds with reaching our full potential as a global community. I honestly can't think of any optimistic beliefs that reach...

    Pessimistically I believe that individual human nature is fundamentally at odds with reaching our full potential as a global community.

    I honestly can't think of any optimistic beliefs that reach anywhere near the same scale. Best I got is I believe the combination of VR and generative AI may give birth to something akin to holodecks that I could experience within my lifetime.

    5 votes

  18. Comment on Reddit API Changes in ~tech

    Rudism
    Link
    I'm hoping these changes don't break self-hosted Libreddit instances. Redirecting all Reddit links to Libreddit is the only thing that keeps the site even remotely usable for me.

    I'm hoping these changes don't break self-hosted Libreddit instances. Redirecting all Reddit links to Libreddit is the only thing that keeps the site even remotely usable for me.

    3 votes

  19. Comment on 2FA not working? in ~tildes

    Rudism
    Link Parent
    The Bitwarden desktop app on my PC and mobile app on my phone were both generating the same codes, so I don't think it could have been time related on my end. At any rate I gave it about half an...

    The Bitwarden desktop app on my PC and mobile app on my phone were both generating the same codes, so I don't think it could have been time related on my end. At any rate I gave it about half an hour and tried again and it worked for me right away. The seed it was giving me on the enable-2fa page didn't even change from before, so whatever the problem was it either fixed itself or I managed to avoid whatever user error was causing it to fail the first time around.

    2 votes

  20. Comment on 2FA not working? in ~tildes

    Rudism
    Link Parent
    I don't think that was it, I was giving it plenty of space on either end when trying the codes (I've run into that problem before though). In any case I just tried again and it's working as...

    I don't think that was it, I was giving it plenty of space on either end when trying the codes (I've run into that problem before though).

    In any case I just tried again and it's working as expected now. Maybe it was just a weird blip or user error on my end.

    4 votes