• 9 Posts
  • 101 Comments
Joined 1 year ago
cake
Cake day: June 9th, 2023

help-circle
  • Everything Wordpress is heavily infested with that. However you don’t have to let it impact you – it kind of looks to me like they pressure commercial vendors to put their stuff under the GPL if they’re wanting to offer a free version, so there’s a robust ecosystem of actually-FOSS tooling for it. My experience has been that it’s always worked pretty well in practice; you just have to keep your nope-I’m-not-paying-for-your-paid-version goggles firmly affixed. (Also, side note, GPT does an excellent job of writing little functions.php snippets for you to enable particular custom functionality for your Wordpress install when you need it.)


  • Wordpress 1,000% (probably coupled with WooCommerce but there are probably some other options)

    I honestly don’t even know off the top of my head why you would use anything else (aside from some vague elitism connected to the large ecosystem of commercial crap which has tainted by association the open source core of it) – it combines FOSS + easy + powerful + popular. You will have to tiptoe around some amount of crapware in order to keep it pure OSS though.




  • Aaah, got it. Right, depending on the version of Windows, keeping it off the internet permanently might make it less prone to doing random stuff. To me it’s highly unlikely that keeping it off the internet will increase security; there will be machines that are way more attractive targets for evildoers (because of the people on them doing stuff) and evil deeds that are way more dangerous than changing the signage. But if you just want to keep its configuration simple so the signs don’t mess up then that kind of makes sense to me.

    I think I misunderstood your setup a little; I thought the signs were their own hardware with their own IP addresses. If they’re just screens of the windows machines it’s a little simpler. You actually can set it up so that machine can see the LAN but never the internet, just by setting it with a static IP configuration with no gateway set. Then you have to go VPN to some other machine on its network, and then hop from there to the local-LAN-only machine that’s hooked to the signs.

    I actually forgot until just right now, but one time when I had this problem, I set up a Chromebox which was set to display a web page in full-screen mode, and used an extension which auto refreshed the page every few minutes, and then set up the web page to look exactly how I wanted it to (I used a Google Docs static export of a spreadsheet page, so I could automatically gather the right data from formulas and then futz around with the spacing and sizing and etc until it looked the way I wanted it to). You could do a similar thing, with a page that was served from some other place on your LAN. Then, the signage machine itself would never need to be accessed remotely or have any access to the internet; you could just unset the gateway, and periodically or automatically update the page that was getting served from the other machine without needing to touch the signage machine.

    Anyway good luck, hope all this gives you some ideas. Probably your IT department will have specific ideas how they want to set it up anyway, but going into it with things a little thought through might help yes.


  • Why do you want to keep it off the internet, though? That’s going to make things more complex both in the setup and in the day-to-day operation. The example you listed of being difficult to upload files is one example. The only reason I can think of to do it that way would be for security but I’m not sure how much actual security benefit it would carry.

    How about this? You could do the two-interface solution like I described, but have the internet-facing interface disabled most of the time – could be disabled in Windows settings, so someone has to have physical access to the machine in order to reenable it when you want to update the sign. Or, it could be disabled at the switch / router level: Just disable the port for that machine, and reenable it temporarily any time you need remote access to the machine to do something, but in the steady state leave it on its own little disconnected network with only the machine and the signs, and no internet access anywhere.


  • What I would do in this scenario is give the Windows machine two network interfaces, and have the second interface connected to a little static network with just the signs and the Windows machine on it (i.e. no internet access). Then, you can access the Windows machine through TeamViewer or whatever. It’ll have access to the internet but the signs won’t be directly visible from the internet. And if someone from the internet is accessing your internal network to tamper with the signs via the Windows machine then you have bigger problems than them tampering with the signs.


  • I am not “arguing.” I’m saying I think the kind of performative anti racism described in the OP article is silly, especially when it involves so badly stretching the definition of “racism.” Maybe, though, it’s overall not the worst thing in the world and I actually tried to partially retract some of my criticism of it as being overly harsh. But I still think it’s silly and can actually be counterproductive.

    If you think different, that’s fine. I think I’ve explained myself at this point. I am in no way shape or form interested in having an exchange with you where we try to determine which of our viewpoints “wins”.




  • Haha oh yeah, I wasn’t talking about you. Just I’ve noticed that certain viewpoints tend to attract a lot of downvotes here. I suspect that a lot of people like to do performative antiracism more than they do genuine antiracism, because it’s a lot less work, and that extends to giving out vigorous downvotes to the “wrong” point of view.

    But yeah, I can see the argument too. Everyone’s going to draw the line of what’s okay and not okay to say in different places, and at the end of the day I do think there’s something to be said for trying to make the world a better place even in some kind of trivial way.


  • Hold on, lemme put on my downvote boots.

    To me the defense is, if people are going around and saying that calling it an “Inca Dove” is racist or misogynistic and we all have to spend time and money and effort changing it around to something else, then it’s going to hinder genuine efforts to resolve racism or misogyny because some people are going to start putting it alongside the “Inca Dove” thing into a category of “stupid stuff that doesn’t matter.” Changing “Oldsquaw” sounds great because that’s actually racist. Changing the confederate name thing, eh, it seems weird to me but I can see it. “Inca Dove,” alright now you’re just making up stuff to get upset about and asking everyone else to play along with it and if they don’t want to, they’re some kind of bad person.

    Just my opinion.





  • This has some explanation. TL;DR get ready to be underwhelmed. This was based on some earlier efforts e.g. one in Sweden that changed bird names containing “neger” (negro), “kaffer” (a racial slur), or “zigenarfågel” (gypsy bird), but the stuff they’ve been able to find in North America is, well:

    • Oldsquaw (a slur)
    • Inca Dove (historically inaccurate, no overlap with Incas)
    • McCown’s Longspur (McCown was a confederate)

    Maybe there were more they didn’t mention but my guess is that there’s a reason they’re writing the story while dancing around what names are actually being changed.



  • I think an LLM+Chroma is probably as good as it gets, and who knows, it might work. Just I’d be very careful of getting screwed by the process. As I’m sure you know LLMs are right at that inflection point where they’re good enough to seem trustworthy but they can still completely malfunction (and they tend to do so in ways that are actually really difficult to spot because they seem perfectly plausible.)

    Yeah the Legal Eagle video was hilarious. The guy used GPT-4 to make his legal briefs, then when it hallucinated cases he lied to the judge and said he’d researched them and the cases existed, then when faced with the clearly obvious fact that they didn’t, he finally came clean but still sort of tried to weasel out of responsibility for the whole thing and the judge quite rightly tore him a new one. And, I have some vague memory of it being discovered that GPT had basically tried to tell him it wasn’t qualified to make his legal briefs and he insisted to it that it needed to do it anyway. It was just an absolute casserole from start to finish.


    1. I would be very hesitant to put my legal reasoning in the hands of an LLM. They’re not AIs, they just come up with plausible text completions. There have actually been cases by now of lawyers who’ve gotten fucked by using AI to try to save themselves effort and then it not being good enough for what they were expecting from it.
    2. If you’re convinced you want to do this, there are basic tutorials on Youtube - I’m not 100% sure but I think that instead of “fine-tuning” in the same way you would do to fit an LLM to a problem space, you want to import the legal documents into something like Chroma, then use something like Llama as hooked up to the Chroma DB. But again, I wouldn’t. For messing around with some things it’s fine, but for legal documents you really want a sentient intelligence involved in the process.

  • Yeah, sure. I know if there’s one thing I hear Tucker Carlson say all the time, it’s stuff like:

    This all seems like a pretty cogent argument. Actually, I’d take it a step further than what you’re saying – I’d say that if a fully functioning adult wants to get a knee replacement or a gender transition, it’s nobody’s business but theirs whether or not they should do it. It’s relevant to talk about the long term outcomes and etc, but at the end of the day it’s up to them.

    Also, in what sense is this verifiable bullshit? Can you point to a specific claim and why you say it’s bullshit?

    Do you have sources indicating clear positive outcomes for adolescents? Like I say, I’m genuinely interested in learning.

    Out of all the people who engaged with me on this, one person sent me any kind of links for that last question, which is basically the factual core of the matter that I was looking for. I’m asking that question so I can learn. I took a bunch of time to read over them and responded in detail to each one. Most of them I pretty much agreed with, and I said so.

    If you can’t tell the difference between someone who’s attacking you, and someone who actually is genuinely asking for information although they don’t fully agree with you, you’re going to wind up “counterattacking” a lot of people for more or less no reason, and making enemies of them.


  • Yeah I get that. The initial reasonable assumption I 100% understand. But I also think a vulnerable community that makes a habit of lashing out and accusing of crypto-democide anyone who’s showing good faith, but just not willing to 100% agree with them on everything without any discussion permitted, is going to find itself more vulnerable and demonized as a result, not less.

    The thing of “you have to agree with me or else you’re the enemy” isn’t a good way to go, whether you’re in the majority or the minority. Again I get the reasons why people arrived there. I’m just saying it’s not a good place to be.