🎉 thanks to the developers and everyone who helped!

one bug i noticed after the upgrade: my notifications page shows unread notifications for (what i guess is) every reply i’ve ever received which was later deleted. the count in the bell icon only reflected the actual new unread notifications I had received since I last looked, but when i click to view my unread notifications then all of these old ones about deleted messages appear to be unread now.

🧲 (according to many web pages you’ll find if you search for that hash, anyway… i haven’t tried it)

The mistake is in deploying these things in the first place.

One of the many big problems with LLMs which won’t be solved any time soon is that they are confidently wrong.

For this reason, they are worse than useless; they’re actually harmful. These systems are bad and the people behind them should feel bad about deploying them on a naive public who often doesn’t realize how bad they actually are.

I hadn’t heard of YouChat before so I just spent a few minutes playing with it to demonstrate this:

I predict that the SOTA LLMs in five years from now will either outright refuse to even try to solve math problems, or they will still fail (by providing confidently correct-sounding incorrect answers) simple tests like the above. And of course, if they do just make them refuse to try to do math, they’ll still be confidently wrong about whatever questions they do answer.

how will the fix work?

Would you update the community description to say that predictions should have specific dates associated with them? Falsifiable predictions are much more fun :)

also, as @graphito@beehaw.org said in the xpost:

auto redirect link from any instance:

/c/predictions@lemmy.ca

My mnemonic is that the higher-level more user-friendly tool (adduser), which is usually the one I want to use, is the one with the more grammatically correct name.

Things like that make me increasingly feel that there’s no difference between political administrations in so-called “western” countries like the US and those in China or Iran

That’s really not fair, there are lots of differences… and this is just one of them. (I’m not an expert on the subject but I’d be surprised if their school book censorship regimes are also based on an allow-list.)

There are plenty of other differences too, like the amount of military spending, or the number of people in prison. Don’t worry, USA is still #1 in these categories. It also remains one of the last places in the world where many freedoms from government interference are still cherished.

That sounds quite reasonable.

So basically these teachers are making a big drama out of nothing.

I believe my summary in the title of this post is completely accurate, even if the manatee school district is not taking the law as seriously as their teachers are.

Florida has literally moved their schools from a system which banned some books to one which bans all books by default until each book is individually approved.

Here is another story about it.

Here is the actual law.

Here are some excerpts of the law:

148       1. Each book made available to students through a school
149  district library media center or included in a recommended or
150  assigned school or grade-level reading list must be selected by
151  a school district employee who holds a valid educational media
152  specialist certificate, regardless of whether the book is
153  purchased, donated, or otherwise made available to students.

…

171       3. Each elementary school must publish on its website, in
172  a searchable format prescribed by the department, a list of all
173  materials maintained in the school library media center or
174  required as part of a school or grade-level reading list.
175      (e) Public participation.—Publish on its website, in a
176  searchable format prescribed by the department, a list of all
177  instructional materials, including those used to provide
178  instruction required by s. 1003.42. Each district school board
179  must:
180       1.  Provide access to all materials, excluding teacher
181  editions, in accordance with s. 1006.283(2)(b)8.a. before the
182  district school board takes any official action on such
183  materials. This process must include reasonable safeguards
184  against the unauthorized use, reproduction, and distribution of
185  instructional materials considered for adoption.
186       2.  Select, approve, adopt, or purchase all materials as a
187  separate line item on the agenda and must provide a reasonable
188  opportunity for public comment. The use of materials described
189  in this paragraph may not be selected, approved, or adopted as
190  part of a consent agenda.

So basically these teachers are making a big drama out of nothing.

Reading the actual law, do you still think so?

I wish I could see a list of what I have upvoted and downvoted myself. The fact that admins of lemmy (and other AP-compatible) instances can but I can’t is pretty irritating.

We’re not intruding on this space. We’ve been in the fediverse for just as long or longer; the fediverse has been scrapable since 2008.

Totally. And while it was scrapable, and scraped a lot, I wish there had been a lot more systematic public scraping of the “federated social web” (as it was called before the terrible name “fediverse” was adopted) back then - I had a lot of public conversations on identi.ca and StatusNet which I wish I could still see, but they now exist only in a bunch of private databases I don’t have access to. 😢

What about public parks? Is it okay to walk around you while you’re having a conversation and record you, and then post that conversation on-line?

No, that would certainly not be okay. When I’m walking in a public park I have some expectation of privacy. If you’re walking close to me when I’m having what is intended to be a private conversation, I might notice and pause.

You are conflating private and public conversations. When we’re having a conversation in a public forum like this online, we are both posting it on-line already.

I hope archive.org posts another copy on-line so that if I want to refer to this later, after lemmy and the whole cargo-cult-deadend activitypub architecture has gone the way of the dodo, I will still be able to. And I hope they make it searchable!

Is it okay to use directional microphones to record you in such a setting?

Of course not. It’s also not possible to be sure it isn’t happening, but, if/when that is happening it is an unambiguous violation of social norms (and the law, in most places).

Doesn’t the whole recording-in-the-park thing from the Conversation give you the creeps?

Absolutely. (And now I’m wondering if you’ve noticed the reference to this film in my profile here or are bringing it up independently… 😀)

Are you saying that the fact that something is difficult to enforce against makes it okay to do, even if the person you do this to does not want it done?

Not at all. I think publicly archiving public web content is okay because I think it has a net public benefit. Better than okay, I think it is a good thing to do.

It is not because it is difficult to enforce against that I think it is okay. The fact that it is difficult to enforce against is why I think that it is not okay to give people who don’t know any better the false impression that it is not difficult to enforce against.

I could be wrong, but I interpret this post as being about Mastodon’s culture of being against search technology, which I find depressing and irritating for reasons I explained in that other thread as well as this one.

However, I just noticed a place where there is some lack of informed consent here on Lemmy: in the Lemmy UI, it appears that upvotes and downvotes are anonymous. I checked a long time ago, and realized that they weren’t really; the identity of the up or down voter is federated, but it is simply not shown by the UI.

I would assume that many (probably most) lemmy users do not realize this: admins of your own instance and all federated instances have the ability to see who upvoted and downvoted what.

It just now came to my attention that Friendica actually is showing this information publicly, in the form of “$username does not like this” for a downvote! https://rytter.me/display/4c906314-4763-d3aa-4584-11a516756414 🤣

(hey @OptimusPrime@lemmy.ml … why did you downvote that? I myself am also listed there as not liking it; I downvoted it as a test to confirm my assumption that it would show up as “does not like”, and then when I undownvoted it that event apparently didn’t get federated.)

imo these are the kind of “informed consent” issues that fediverse developers should be thinking about, rather than “how can we increase the power imbalance by making it so that only the elite are allowed to have fulltext search… in the name of justice” as so many seem to be hell-bent on doing.

i clicked a button that most lemmy users would assume is an anonymous up/down vote and now my name is listed on a 3rd party website saying i “don’t like” something (even though I tried to undo it). #thisisfine ?

thanks :) i made it after reading your comment.

(based on this, which is a reference to this… see also this)

I’m not actually sure if fortran is involved in any of the parts of numpy which I’ve actually used myself, but I do know that if I apt remove libgfortran5 it would also remove the python3-numpy package.

are cafés public or private spaces?

They’re fundamentally private spaces, even if open to the public. Under certain zoning ordinances they may be considered a “public place” for some purposes if they are above a certain size, but this does not negate their ability to set their own rules and deny access to members of the public who violate them.

If a cafe wants to enforce a “no phones” rule, they can do so relatively effectively. If a website wants to enforce a “no robots” rule (especially if they also want to not require any login to view the content on the site) they can ultimately only pretend to be able to do that effectively.

Can I just sit at the table next to yours and stream and record your conversation with your friends?

You technically can, and if you get caught the cafe can (and should, imo) kick you out for doing so. Pretending that a provider of an electronic publishing system can enforce the same kind of social norms as are possible in physical spaces is silly at best and actually harmful at worst.

Some of my favorite bars and cafes outright prohibit the use of phones and also don’t operate CCTV, but in many places you are in fact frequently nonconsenually recorded by other people, sometimes streamed onto something like facebook live, as well as constantly by 4K CCTV with audio (in violation of the law in many localities, yet still common).

When you’re having a conversation in a physical space and you notice someone eavesdropping, you sometimes might speak less freely as a result, especially if they appear to be filming. In a public conversation online, especially one readable without even logging in, you can’t tell when someone is “eavesdropping” because you are publishing.

I’m a big proponent of enforcing privacy in online and offline spaces with technology, policy, and social norms. I’m also opposed to magical thinking. Telling people that they can semi-publish, to have some of the benefits of publishing without some of the consequences, is misleading to the point of being dishonest.

I blame facebook for conditioning people to believe that such a thing is possible, through their years of blurring the lines between public and private.

I agree with a lot of the spirit of what they’re saying, but I’m pretty sure I wouldn’t agree with their concrete applications of it (although they are unclear in the thread).

I think blurring the lines between public and private spaces is the opposite of informing consent. Cultivating unrealistic expectations of “privacy” and control in what are ultimately public spaces is actually bad, imo.

Informed consent in the fediverse should look something like a message on the signup page that says: This is a publishing system. Be aware that everything you publish here will be distributed to a bunch of other servers which are not under the control of us, the operators of your server. When you edit or delete something you’ve published, we will honor it and relay the message, but other servers may or may not honor it. There are many other tools for private (encrypted) group communication, but that is not what this is. ActivityPub is for publishing.

ps: I, for one, am glad that the Internet Archive exists!

via imgur, the above are all non-rectangular alpha-transparency-having themes (or “faces”) used by Panic’s Audion from 1999 to 2004.

There is a sad story about how it almost became iTunes, but because the developers were already in talks with AOL they tried to invite AOL to their Apple meeting and… apparently AOL couldn’t fit it in their busy schedule, so, iTunes was instead based on an inferior player distributed by Casady & Greene called SoundJam which looked like this:

edit: TIL that in 2021 Panic released “a stripped-down version of Audion for modern macOS to view these faces”. Unfortunately all of the screenshots in the linked directory appear to be 404 now, but the faces themselves do appear to be downloadable. (Also unfortunately, there still doesn’t appear to be a player that can use them without running a proprietary OS…)

Op-Ed: Yes, Gov. Cuomo, Car Helmets Could Save Lives

Amazing, thanks for the link. I just re-posted it here.

via a comment in The Cult of Bike Helmets: The history—and danger—of a modern safety obsession.

alternative automated plagiarism engine?

I think it would definitely be “perceived as unprofessional and silly in a negative way” by some people and also “in a net positive way - perhaps a bit silly, but memorable” by others, so, if you’re very concerned about appearing serious and professional you should probably not use such an address for these purposes.

If you emailed me from this address i would consider you a dork, and as a bit of one myself i might start calling you Fratnickle.

update, 10 months later

Hexbear currently has more than twice as many users online as Lemmy.ml and Lemmygrad combined, and a mind-boggling number of posts (total, as well as total divided by total users, etc):

Lemmy.ml:

Lemmygrad:

Hexbear:

These hexbear people appear to be extremely online. Them federating with us here will have a major impact on these spaces, for better and worse.

🍿

(here are similar screenshots i took of lemmy + lemmygrad last april…)

Brave was never based on Gecko. (Originally it was based on (a fork of) Electron! 🤣)

#NotTheOnion

This is a nice overview of this absurd situation, but Tim Bray’s conclusions are a little surprising to me.

Yes, Mastodon traffic either is already or soon will be captured and filed permanently as in forever in certain government offices with addresses near Washington DC and Beijing, and quite likely one or two sketchy Peter-Thiel-financed “data aggregation” companies. That’s extremely hard to prevent but isn’t really the problem: The problem would be a public search engine that Gamergaters and Kiwifarmers use to hunt down vulnerable targets.

Here Bray appears to be missing the fact that those people will often end up with access to those Thiel-financed private intelligence services that will have the full-text search, while the rest of us won’t. Making things public and pretending they’re private by shunning search effectively lobotomizes everyone who abides by this custom, while still allowing the worst people to have the capability (and not only the ones working in state intelligence agencies).

What success looks like: I’d like it if nobody were ever deterred from conversing with people they know for fear that people they don’t know will use their words to attack them. I’d like it to be legally difficult to put everyone’s everyday conversations to work in service to the advertising industry. I’d like to reduce the discomfort people in marginalized groups feel venturing forth into public conversation. (emphasis mine)

This is a conflation of almost entirely unrelated issues. The first half of the first sentence is talking about non-public conversations. The solution there is obviously to use e2e encryption, so that even the servers involved can’t see it, and to build protocols and applications that don’t make it easy for users to accidentally make private things public (ActivityPub was not designed for private communication, it was designed for publishing, so, it is unlikely to ever be good at this). The second sentence is about regulating the ad industry… ok, cool, an agreeable non-sequitur. But the last sentence is talking about public conversation… and in the context of the second half of the first sentence, it carries the strong implication that Bray somehow entertains the fantasy that conversation can somehow be public and yet be uninhibited by “fear that people they don’t know will use their words to attack them”.

Contrary to what the title suggests, it sounds like they aren’t actually planning on implementing ActivityPub on medium itself, but rather they are just launching a new medium-branded mastodon instance. 🥱

since you bring your own domain it is not total vendor lock-in

sure you can point your domain elsewhere, but i don’t see anything about the ability to export your data in a way that lets you actually migrate to something else (though i think that is arguably required by GDPR…)

on the other hand, and other side of the so-called “proxy” service, it’s a shame that the golang community has so far mostly acquiesced to the situation where the language’s tooling by default relies on making numerous connections to google’s servers.

serious investors

yeah, risc is good

sorry to say i don’t have a github account associated with this pseudonym and am too lazy to make one right now.

Lemmy does have an emoji picker in the web interface. To access it just type a space and a colon ( :) while editing a comment, and then begin typing the name of your desired emoji. Like this:

😂

unfortunately there is currently a bug in it (at least here in tor browser) which causes the menu to appear near the top of the page when replying to comments far enough down the page that it is necessary to scroll, so, it is easy to miss. but it does work, though it is sometimes necessary to scroll up to see it. cc @nutomic@fedibb.ml @dessalines@lemmy.ml in case you guys haven’t noticed this bug.

the fact that they’re now owned by automattic (wordpress) makes me optimistic they won’t. in fact, they might even help evolve activitypub into something that doesn’t suck as much as it does now.

the server software is non-free. iiuc it would be easy enough to reverse engineer the protocol from the client software (which is free software) but (last I checked, anyway) the server URLs are not configurable so you would actually need to patch and recompile the client to use a different server.

I’m certainly not recommending snaps, but, it is important to acknowledge the problem they’re trying to solve. “The debian model” means using years-old versions of everything, having a single set of dependency versions every program must share, and giving every package’s control scripts root access while you install it. This paradigm made sense when it was developed 25 years ago but it is far from ideal today.

i still ♥ Debian but there are tons of things I need to use which I can only get from somewhere else, so, “the Debian model” for me nowadays means a stable base system and then lots of software from other distributors (sometimes flatpak or appimage, but also a lot of podman containers of various distros).

What I am almost never willing to do is use 3rd party entries in my apt sources.list file on an actual host system (though I do in containers when necessary) - down that path lies madness.

yeah, I am aware, and I do actually think the xdg portal stuff is generally a good idea for a lot of programs… but the way it works right now sacrifices a lot of usability and doesn’t gain much security.

passing files given as commandline arguments seems like an easy problem to solve, but the linked file situation with SVG is much harder (probably requires a whole new flow for xdg portals where a program can request access to a bunch of files and prompt the user once to allow access to all of them). in the absence of any solution, imo it is silly that they’re shipping inkscape as a snap with strict confinement today.

I’m unsurprised to see lots of good reasons here why not to use them already, and none for why anyone does :)

I imagine the vast majority of snap users are using them only because Ubuntu ships a few things (like firefox) as snaps by default now.

I tried the Inkscape snap recently on an ubuntu system where i needed the latest release, and found that due to its sandboxing security theater (last I heard it is still not difficult break out…) it is impossible to open files from the commandline. And, even worse, when you use the Open command from File menu, it just passes the one file you selected in to the sandbox, so, when you open a file which has references to other files (which is not uncommon with SVG) it is not able to load them! So, I ended up using Inkscape’s AppImage instead.

better?