@cassidyjames
I'm so confused by this position. Isn't Google largely who you want privacy *from*?
As the primary internet monopoly, Google form a major part of the US govt surveillance network, as well as running the dominant ad tracking network. They can't be trusted, they will change terms as suits them.
Using Google's VPN is worse than not using a VPN at all, because you're willingly sending them your *full* internet activity, far more than they'd have naturally.
@post @wizzwizz4 @didek
I've talked about this before. It's something that you'll see me bring up over and over again.
If you are feeling aimless, a little depressed, and in need of grounding...
VOLUNTEER.
Find a soup kitchen, a homeless shelter, a senior citizen center, a hospice, a charity, somewhere you can be in the service of others.
Something happens to us when we serve others. It takes us out of our heads. It centers us.
It just helps.
Trust me.
VOLUNTEER. Help yourself, by helping others.
To conclude, ChatGPT is a very good brainstorming tool because they explore perspective spaces that human biases might ignore. It's more important now to know how to use GPT-like systems rather than spending your time thinking it's an oracle of wisdom.
In my ideal world every server would say "here's how we decide who to federate with" and it would range from "every server that isn't doing protocol-level attacks on us" to "every server on which I haven't personally observed behavior I think my users would prefer not to see, such as [examples]" to "every server not listed on [url]", and then people could decide which kind they want to be on.
There could be one or more fediblock-like tags that people could use to communicate about who to block, without it having to be an absolute condemnation that everyone agrees on.
(I'm on #qoto partly because it's in that first category, not blocking much at all at the server level, and I do like to see trash before burning it, but I'm totally okay that many people would rather not! It's okay to delegate large parts of your Mastodon filtering to a trusted admin if you want; nothing at all wrong with that.)
The problem seems to be that lots of people, including some admins, think that servers have to be divided into universally praised and universally condemned sets, and of course any universally praised server would never federate with a universally condemned one. Some people just want the universe to be simple! But it seldom is.
(And so qoto sometimes gets blocked just because it doesn't block some other server, which seems I think silly to me, although I can put together an argument to the contrary if I try.)
fwiw...
@ivory @elk @pinafore A really interesting side-effect of playing with all of these clients is seeing how they all experiment with different UX patterns for the same Mastodon features. Even when Mastodon features are similar to #birdsite, the UX of these apps is sometimes wildly divergent. A young ecosystem, I'm excited to see what patterns emerge.
@JamesGleick @peachfront No, those numbers didn't have to come from somewhere. It's doing pattern matching (though on a much grander scale than your usual Markov chain), and it's filling in a slot for "number goes here". It mash be doing better than that --- it may know the scale of the kind of numbers to be found in similar documents, but that doesn't really mean the numbers "come from somewhere"
@jmbinder @JamesGleick @peachfront People have come to treat it as an oracle, but it's not fundamentally a knowledge database that contains facts and details. It's fundamentally a language model. (That happens to accidentally remember stuff.) If you *gave* it a document that contained the knowledge you were seeking to analyze, you could ask it to extract the relevant facts. But when we ask it whether it happens to remember a fact, we're inviting it to hallucinate.
Where the state lacks means of coercion, it is important to control what people think.
-- Noam Chomsky
Carl Sagan passed away #OTD in 1996. In his final interviews he left us with two messages that are even more relevant today: one emphasizing the importance of a science literate public, the other a warning about how hard it is to extract ourselves when we’ve been conned.
Thought Leaders: "Programming will be obsolete within 5 years. All business software will be generated by AIs"
Actual Businesses: "actually we're mainly still on-premise, running on Windows Server 2003 and VBScript ASP, but our two-year plan to migrate everything into a cloud kicked off in 2015 and we're making steady progress."
Short and sweet anti-crypto documentary.
This was an hour well spent. Well written, hits all the important points, and a great launching pad for anyone to ask follow up questions.
Good video to share with family and friends, and warn them of this high tech scam.
Via Stephen Diehl on twitter:
If you're excited that things are entering the #PublicDomain — which you should be! — perhaps it's worth considering reforming copyright.
In many places currently it's 70 years after the death of the author, or longer. This is insanity.
An important piece of culture made by a young author could be locked behind copyright (with rights usually held by publishers, not authors themselves anyway) for over a *century*.
That's how you get #Disney to own almost everything.
Reform #copyright!
@stux
He is ( in this case ) taking responsibility by showing us the documents proving that Twitter ( before he bought it ) took government cash to censor people.
He should be applauded for that.
Musk is bad a lot, but not all the time.
An extraordinary amount of money is being spent on COVID-19 testing. Programs meant to ensure people have easy access to tests are being exploited by profiteers and fraudsters. Even legitimate businesses have earned millions while running troubled testing programs. Here's my latest: https://www.propublica.org/article/how-fraud-increases-medicare-spending-covid-19-testing
https://www.euromomo.eu/graphs-and-maps
EuroMOMO data for excess mortality in Europe for 15-44 years old: in 2021 it starts to increase since May and overtakes all previous years including 2020; then in 2022 it keeps increasing easily overtaking 2021.
What happened to young adults in mid 2021? We all know it and it's time to admit it.
We don't have to fight open source, we have to exploit open source. -- Larry Ellison, Oracle. Interview source https://www.ft.com/content/5f7bdc18-ce85-11da-a032-0000779e2340 🤨
#Mathematics #Macroeconomics #Engineering #ComputerScience #Programming
#Ecology #Environment #Democracy #Freedom #Equity #Liberal #Socialism #PostKeynesian
#Inkscape ·
#Gimp ·
#Blender ·
#VLC
#WordPress ·
#Mastodon ·
#PeerTube:
searchable