CoderKat @ CoderKat @lemm.ee

I write bugs and sometimes features! I'm also @CoderKat@kbin.social.

---

Read more

Posts

7

Comments

728

Joined

2 yr. ago

I've tried a few apps (Liftoff, Wifwif or whatever it was called, and now Sync). I've not stuck with any of them, despite the fact I used RiF and later Boost both for years.

They just feel too early stage. Eg, Wifwif was super sluggish and I accidentally cancelled multiple long comments I typed up by trying to scroll, which is a big oof. With Sync, I noticed the comment hierarchy was messed up and when I went to post about that, I couldn't do it from within the app because posting isn't supported yet (feels like a big thing to be missing).

So for now, I'm still just using the website. And not with the PWA, because support for multiple tabs is really great and avoids weirdness with scrolling when going back.

Side note: why do literally no apps support tabs????? Number zero feature request I've ever had, for every single app. I even noticed that Sync has a long press to open posts in background, but it's only for posts and it is rather convoluted cause it's more like opening a second instance of the app. The back button goes to my Android home screen (not back to Sync) and it is hard to navigate to what you just opened.

Mobile Chrome's tab groups feature is absolutely amazing. I hate using "normal" tabs anymore, cause the tab groups having the tabs accessible at the bottom of the screen is unbeatable.

It's a shame that most game companies can't manage to treat both their fans and staff well. It's rare to treat even one of those well, let alone both. I wonder what the developer of BG3 is like for employees? Crunch time is usually so ubiquitous in game dev that I assume everyone does it until proven otherwise.

Spam callers have basically ruined telephone as a medium. For many, a phone call is more likely to be fake and spam than it is to be legitimate. And even if the call claims to come from a source you might trust, good odds it's spoofed and thus cannot in fact be trusted.

A shame on telecoms for not being willing to tackle the problem.

Oh, I would love that. I hate having to wait for the server to bring me the bill so I can leave. When I eat alone, I often ask the server to bring me the bill with my food or so, in hopes of paying for it before I'm done eating. For whatever reason, the norm is for the bill to not come until well after you're done eating. That can be fine when you're with a group and want to spend time chatting after you finish, but if you're in a hurry, you'll be waiting a long time.

If they design the online menu well, I'm all for it. In theory, an online only menu should improve the quality of the menu, by making it much easier to keep it up to date (eg, there should never be an "ask your server what the soup of the day is" in an online menu -- update that you lazy schmucks) and it's also easier for them to have pictures.

The fact that not every food item has pictures drives me crazy. Show me what the dang food looks like. Unless it's something you never sell, it's so dang easy to get a picture when someone orders it. I don't want a fake picture anyway; I want the pictures to be representative of how it will really look. With physical menus, it's understandable because printing is so expensive and pictures would make the menu extra long. Online menus don't have that issue or excuse.

But yeah, if their online menu is gonna be a mediocre PDF copy of their former print menus clearly designed for 8x11 or whatever, it's a downright worse experience using a digital menu. That's the case for a shocking number of restaurants I've been at. You'd think mobile devices are some new fangled fad or something, cause they sure don't think mobile is something someone will view their menu on.

Yes. But the fact that the law even exists, that the guy got charged, and that the charges didn't get dismissed before this point is extremely concerning. There were multiple points of failure before we even got to this point.

Lawmakers never should have created such a shitty law. Police never had to issue tickets. Prosecutors never had to charge him. The judge could have dismissed the charges as utterly ridiculous. But no, all went through and we had to depend a jury to stop this bullshit.

Also, the article has the quote:

The city of Houston said it will continue to “vigorously pursue violations of its ordinance relating to feeding of the homeless,” according to a statement released to news outlets.

So despite all of this, the city of Houston is determined to still be evil little fucks. People tell me Houston is supposed to be progressive, but I'm not seeing it. Houston folks, what the hell?

Especially since PornHub has an excellent point. Even though they theoretically could do an ID check, the sketchier porn sites simply wouldn't. All these laws would do is push minors to use more dangerous porn sites. They're not going to not watch porn just because the big, law abiding site checks IDs.

What kinda adult would care about your ISP knowing you watch porn?

VPNs are often over blown. Their ads will make it seem like they're some critical privacy thing, but these days virtually every website is encrypted (HTTPS). Your ISP can know the IP address you're visiting (and thus typically the site), but pretty much everything else, including what page you're viewing on the site, is known only to you and the site in question. VPNs had a lot more merit before HTTPS was ubiquitous.

They're still useful if you're visiting illegal sites or using peer to peer technology (most particularly illegal torrents), but that's not the case for many and certainly not for just viewing normal porn.

Senate passed the bill by unanimous consent, although several senators stated later that they would have objected if they had known that the bill could pass. No iteration of the bill has passed the House.

WTF? Why the heck would those people have wanted it to fail and why would they vote for it if they did?

And since it got unanimous consent in the Senate, why the heck didn't it go to the house? It's the rare case of a bill that I think everyone likes.

Yeah, I think those go hand in hand. The kind of leadership that would push RTO is the kind that frequently would also do other bad things (or let their managers).

Woah, too far. They shouldn't have to die for being dumb. Just have them row to the edge and once they find it, they can come back.

I think for cooking, it's important to have habits. I don't turn off the alarm until I'm physically getting up. I try to avoid cooking when gaming, too. Or at least try to avoid any gaming situation where I couldn't quickly leave to cook. It's actually something I wish PC games could copy from consoles. Every modern console can pause anywhere, which isn't possible in some PC games and makes it easier for me to take breaks.

Going in blind is the best. Unfortunately, it's probably pretty difficult to do, since spoilers are extremely rampant. I don't even know how I managed to do it. But it really is magical if you go in blind.

Since you didn't mention how they usually work, for OP, they literally just plug in to any power outlet. That's it. They cost like $50 CAD. They're just about the size of a smoke detector. They'll display a number and beep if CO levels are dangerous (upon which you should immediately leave the house and call the fire department).

Honestly 2 days is entirely understandable to forget.

When I change such an important password (which is very, very rarely), I keep a hidden hard copy for a short while, until I'm confident I have it memorized (usually a couple of weeks). While this has some risks, obviously so does losing access to your data. It depends on your living situation and threat model.

Within Lemmy, I think active is most used as it's the current default for most instances. I'm not sure how it's implemented, but based off my observations, I suspect it has some massive drop off after 2 days, because I regularly see posts up to 2 days old and never any older.

Though OP mentions just one day and I don't have a good explanation for that. Personally, I don't like to comment on anything older than a day and often self censor accordingly. I consider "within the last day" to be the sweet spot for engagement. Beyond that, it often doesn't feel worth my time to comment because fewer people will see it and be able to respond.

I'm impatiently awaiting a "best" algorithm and will switch to it as soon as it's available. I dislike literally all the sorting algorithm choices. I just dislike active the least because comments are what I'm here for. Sure would be nice to see smaller subs too, though.

ITT: people happy that Canadians won't see local news as often, just because they perceive this as hurting Facebook.

You can hate Facebook and understand that this is a shitty law. To be fair, if you aren't in the loop on what the law is, it's easy to accidentally think it's a reasonable law. When I first heard of it, I assumed it was just preventing companies from stealing articles and keeping news sites from getting clicks. After all, that's a real problem and a totally reasonable solution.

But nope, the law actually prohibits linking to a news site without having an agreement to pay them. Yes, linking. I really hope it's obvious how dumb such a law is. They wrote the law to only apply to super big corporations. Otherwise this post would have had to pay Global because you linked to it. Similarly, copying any part of the article in any amount also needs to pay them. Even the headline and tagline that Lemmy shows. It's hypocritical that anyone cheers this on when this very post wouldn't comply with the law.

If it's stable diffusion img2img, then totally, this is a misunderstanding of how that works. It usually only looks at things like the borders or depth. The text based prompt that the user provides is otherwise everything.

That said, these kinds of AI are absolutely still biased. If you tell the AI to generate a photo of a professor, it will likely generate an old white dude 90% of the time. The models are very biased by their training data, which often reflects society's biases (though really more a subset of society that created whatever training data the model used).

Some AI actually does try to counter bias a bit by injecting details to your prompt if you don't mention them. Eg, if you just say "photo of a professor", it might randomly change your prompt to "photo of a female professor" or "photo of a black professor", which I think is a great way to tackle this bias. I'm not sure how widespread this approach is or how effective this prompt manipulation is.

Yeah, no news (as in, from the moment Musk acquired it) would have made me continue to casually use Twitter. But the bad news that came out of Twitter quickly made me delete my account and eventually even make an active effort not to even click a twitter link (and encourage others to do the same). No way "all publicity is good publicity" applies to Twitter.

Agreed. Personally, all I wanna do is take the vast majority of their money and redistribute it to society. They can keep a few million to fuck off with. I don't really care about them beyond that. They largely only have power imbalance because of their money. I don't think anyone should be a billionaire (or frankly more than about $10M USD or so -- which is currently enough to comfortably live your life without having to work, yet isn't utterly crazy).

Once they no longer are so rich, why would I care about them anymore? I'm not one to try and get revenge or anything, and I think that's a harmful way of thinking. I just want society to suck less and fixing the massive wealth imbalance is a big part of that.

If we were in a position where "eat the rich" could be taken literally, then "seize their current and future wealth" is just as achievable.