Not trying to defend them, and I haven’t looked at the feature set of the new Discord integration, but I can imagine a reason for most of those - assuming, of course, that you can basically chat in through WoW on discord.

Afaik it links channels, but to know if your profile is in the server the channel is on, or join you there, it would need to manage and read servers, no? Of course, there are probably better ways how to do that, that respect your privacy a little bit more.

The friendlist I’m having a hard time justifying, can you send DMs from WoW?

But I do honestly wonder how would a MVP permission set looked like if you tried implementing the same integration they are doing, whatever it is.

But yeah, Microslop hoarding data is probably the reason for most of those.

Kind of have an urge now to post an issue that the nstall skill has installed a malware on my machine.

It’s pretty plausible that could happen, and good luck debugging that.

I can imagine that works for most cases, but when you try to include some lesser known and obscure library, that is heavily typosquatted, you might run into issues.

At this point I’m seriously considering just running something like openclaw with a basic model to just keep clicking on random stuff and browsing the internet, changing personalities each day at random, to flood my IPs data.

I’m glad we have options here in Europe. Not many, but there are at least 3 mobile and internet providers at my location, so whenever one tries to pull something like this, I just call them that I’m switching providers. I’ve netted a permanent 50% discount and even managed to convince them to keep the discount at my old flat (which was a student’s shared flat where my friends still live) and also transfer it to my new home I have with my partner.

As far as I know, they can give you massive discounts, but you have to be willing to go through hell and run into a good call operator.

But also, that was like 10 years ago and I didn’t need to do it again. I can imagine that in the age of clankers, things may be a lot different (also with the transition to late-stage capitalism and enshittification).

This will be a goldmine for malware groups that are poisoning AIs that use web search to get them to use their malware-ridden version of software and libraries, and I’m here for it with popcorn at the ready.

I kinda hope this will get more common, because the only people affected kind of deserve it, plus it’s a great way how to filter sloprojects at a glance.

As someone who had his relatively fine and deterministic scripts replaced by skill.md bullshit by coworkers – no. This is exactly the things AI is the worst fit for.

We’ve already ran into it halucinating complete result data-set, because it ran into an error but the skill mentioned that it should give the results data to the user. So it just made it up, since it couldn’t get it by following the correct process.

It has also ran wrong builds or input parameters, because it confused their order.

I’ve also seen it skip a step.

I’m pretty confident that the AI running this script will eventually, for someone, run into an issue with getting one of the dependencies, and go on a wild goose chase in trying to satisfy it, eventually running into one of the typosquatting malware repositories or clones. Not to mention that there are groups that are actively poisoning AIs that search the web to feed them their malware-ridden typosquatting libraries instead.

If you suck so much you can’t write a simple script like this (or just can’t be bothered), you should at least just ask the AI to write it for you, so you have a deterministic set of steps that you are confident will always behave the same.

This kind of does something similar :D https://codeberg.org/TheMikina/git-llmfs

While I’m not a fan of most AI usages, this is the thing that infuriates me the most.

I like writing scripts to automate parts of my job. I’ve had a few for things like build performance testing comparison and the like.

All of it was replaced by skill.md, that does exactly the same, but burns like 3$ per run in tokens, and has also at least once generated hallucinated results, because it ran into an error, ran wrong builds, or in general fucked up in a way that was not easy to detect (and we did in fact not detect it until much later).

But hey, at least my colleagues now don’t have to open the filthy commandline and write py perf-test.py main feature/branch to run the test, and can just talk about it to a clanker.

Ngl, I should update my personality rule.md to talk to me like Duo the Owl would when I’m missing my dailies.

I currently have SHODAN from System Shock, and it has been kind of fun, a little bit of harmless flavor that burns tokens I don’t have to pay for (or have a reason to keep low), since I’m forced to use it anyway. Duo the Owl also sound like fun.

Yes.

This works for me too.

But, there is one exception that worked even better. Find someone who doesn’t struggle with maintaining his momentum and motivation, and work on the project with him. At least for me, not wanting to dissapoint and force myself to keep up with my roommate, who was kind of a workoholic and with whom we worked on a game, is what forced me to keep on working on it for 6 years, before we finally released it.

I would’ve given up after a year, once we handed it in as our semester work, but his team management that he managed to keep up for 5 more years is what made me keep on working.

So, yeah. Either don’t tell anyone, or be the second follower for someone is what works for me.

But, discipline is also what works wonders. When I was at college and used to regularly having to work on schoolwork/my own projects, I didn’t struggle aa much with starting or doing anything.

Once I fell out of this rythm and procrastrinated for few years, it’s now super hard to do anything.

Try to force yourself to work on it regularly, even if it’s just “every two days, work for an hour, without exceptions”. After few months of this, it will become way more easier, once it’s a habit.

We had someone at our office who took a large dump right in the middle of a closed toilet lid, and just left it there for someone else to deal with. It’s not even a shared or public access building, we’re the only company there, like 40 people on that floor at max.

Three times in a few months.

I left before they managed to catch him, because it took that long. Don’t know how why he did it, but I still can’t wrap my head over it. Because from what I’ve heard, it was not an intentional “I want to get fired, fuck all of you”, but he claimed accidents?

It is a tremendously stupid idea, which is why I found it funny (also, it’s one of those things where name came first and inspired the rest).

But, to take it seriously for a moment (and I’m not trying to defend it, not a fan of LLMs as most of the people here), is it necessarily lossy? I mean, you basically have to include the 5Gb large model for it to work, and you just move the data from the file to hoping the summary can trigger a correct combination of parameters.

I didn’t run any larger tests, and I assume that if you managed to keep the API/function names and behavior, the summary would be actually longer than the actual implementation in most cases anyway, so it’s probably not even a compression (especially if you include the model).

It’s just a food for thought, it’s definitely a bad idea to do something like this, to the point where I’m pretty sure you could get millions from investors if you made a startup working on something like this (and that one already exists), but I do honestly wonder if the fact that you kind of have the data in the model would still count as lossy.

If you want the cheapest, go with Cloudfare. They guarantee to ask for the wholesale price, which is the price the registrar pays to the top level domain owner, so they can’t go lower without footing the bill.

What most of registrars do is foot the difference for the first year, so you get a domain for super cheap, then add 50%+ cut on top, so you pay i.e 5$ for the first year for a TLD that has wholesale cost of 10$, while they loose 5$ on that sale, and then you pay 25$ for the second, so they now gain 15$ on holding your domain hostage.

Cloudfare guarantees that they will sell you that domain for 10$, and only raise it when the TLD owner raises the price.

I’m not aware of any other registrar that guarantees wholesale prices, but LMK if anyone knows any.

If you want to get the best deal, buy your first year (or maybe 10 years, if they let you buy 10 for the sale price) with the scammy registrar, i.e get the 5$ sale on Namecheap, and before it expires transfer it to Cloudfare so you don’t pay extra for the second year and can continue with the (much lower now) wholesale price.

Each TLD has a different wholesale price, but every registrar pays the same to them for selling the domain to you. The differences you see is exactly in how much are they willing to foot the bill at first (most have a massive sale on first year, then huge markup on renewals), some just add a flat fee and have markup from the start.

Cloudfare just states “you will pay wholesale”, and don’t do any sheninegans. At least that’s how it was last time I checked.

This made me look into Copilot skills marketplace, if I could make a skill that just runs https://github.com/zoicware/RemoveWindowsAI in a way that works even if Copilot dies in the middle.

I was ready to make a Microsoft dev account to publish that skill to the official marketplace.

Turned out it’s not really possible with how Copilot works, plus it’s also against a lot of the rules you need to pass when sending anything to the Marketplace.

Shame, I still think it’d be hillarious.

I was just about to post this, the fact that it means that there is actually a precedent of astronauts using Teams on a mission, so we may be bringing Teams to the Mars for real, makes it a lot more funnier.

Yeah, what I’ve been trying to do, on a rare ocassion where I chime in with the whole “just use Linux” (because it’s usually not necessary to be 27th person with the same comment), is to at least add more relevant info, or mention misconceptions that I know were reasons why I didn’t switch, or what I found cool.

I.e for me it was gaming, which for a long time used to be considered bad on Linux, especially if you had Nvidia. Which is no longer true in the sligjtest, and I was pretty surprised how smooth the transition was, so I just share a bit of my experience.

But yeah, it’s a balancing act. And as always, if you can’t add anything to the discussion that already wasn’t said (i.e just use linux), it’s better not to add anything :D

It’s what convinced me to make a switch to Linux, two years ago by now.

I’ve seen it mentioned for years at that point. Did it feel annoying? A bit, especially when I was trying to solve something. But it was one post that pushed me over the edge. I’ve probably seen a lot posts/replies like that before.

I can’t say why, I guess the stars aligned, but I’m glad it happened. I probably wouldn’t do it, if it weren’t for years of constant reminders that Linux is in fact an option.

Oooh, so that’s why some clothes could shrink?

I remember hearing that shrinking clothes could be a problem, but I never actually encountered it in my adult life.