Yes, the #ChatControl vote today was a farce. BUT. End to end encrypted messages are excluded from scanning through an amendment. So #UseSignal or other secure E2E solutions, as you should anyway.
I am undecided if the French court decision that allows Marine Le Pen to become candidate for the presidential election is a huge mistake or a brilliant act of sabotage to disassemble her party, the RN, by feeling the fires between factions inside.
The ratio of things you want to change to things you can change is typically 99:1. But focusing on that 1% is worth every minute. Do not feel inferior because you have to leave 99% unsolved. Don’t complain. Do your 1%. Trust the community. Other people will be inspired to work on their 1%. In sum we can reach 120%. Take your ego out of the equation. Do what you are good at. We will win. And celebrate your every little win. Locally. With your friends. Peace :)
New #noFIFA rule: All players in the World Cup quarter, semifinals and finals get the All New Trump Joker Card that they can use once per game to overrule a red or yellow card from the referee!
Code reviewing was never the most interesting thing to do. But it had one important element. That, if done right, it was knowledge exchange between the reviewer and the coder. That can be quite motivating. Helping a fellow coder to become better. Reviewing "AI" written code does NOT come with that potential reward. The machine doesn't learn the way a human does. This turns code reviewing into a menial, fruitless task that leads to frustration instead. That's my observation and opinion.
Interesting. Nikon is testing if there is a market for cameras that are normally only sold as special version for governments etc. „The Z6 III (No Wireless Connectivity) camera is identical to the standard Z6 III, except it lacks a Wi-Fi and Bluetooth chips and is completely incapable of wireless communication with any device, including a smartphone.“
@alan These botnets are more or less immune to rate limiting, as they use many (and I mean millions) of IP addresses fro a run and each IP address is only used for a few requests before it is being put back in the queue. The IP addresses are also from many different providers, so a (sub-)net wide block also doesn't help. I wrote about those "residential IP proxies in [1] and [2].
My timeline (which contains a lot of project leaders/sysadmins from big projects) is filling with posts about a new, ongoing wave of what most likely are scrapers collecting training data for „AI“ companies. They seem to be using botnets (or what some call „residential IP proxies“ to make it sound a bit more legitimate) with millions of IP addresses, making it really hard to defend against. Some have decided to take their sites down until this is over. This is now the world we live in :(
Age verification means identity verification. And it removes the final remnants of anonymity from the internet. Anonymity in the sense of being in control of the decision to share your identity is however a requirement for a public space. The internet can never be a public space with identity verification. That’s what is at stake, in my personal opinion.
Maybe this simplification will help some people: Modern LLMs (Large Language Models) *interpolate* based on training data that is locked in time. They can uncover sometimes surprising connections that weren't "seen" before, but did exist already. What they can not do is *extrapolate*, as in creating new knowledge, what we humans call creativity, inspiration, lightbulb moments. Thus LLMs are like The Matrix, an artificial virtual world that cannot make progress. It's stuck in time
I don't use commercial VPN (Virtual Private Net) offerings, but when a Swedish VPN company defends (in my opinion) massive donations by one of their CEOs (Chief Executive Officers) to a far-right political party with "free speech" and "it's his private choice", it tells me it's not a company I would consider to use as VPN provider. That's my choice.
@badnetmask Yes, HOWEVER, when you are in control of the subdomain and software, you can transfer those keys. This process should become standardised, in my opinion. Right now it’s quite a bit of manual work. @evan this might be something worth discussing on the standards level?
The current heat wave over Europe will force France to reduce the output of their nuclear power plants as the cooling water they take in from rivers isn’t cold enough to allow for full power production. Not for the first time. #ClimateChange is real.
@apps In my personal opinion it is a workaround. The clean solution would be to use (and have support for this everywhere)`"mediaType": "text/markdown"` as described in [1]. If `"mediaType": "text/html"` is used (either explicit or as implicit default when no mediaType is set), it really should be HTML5 according to the primer [2]. See also [3]. @pbnoxious
@pbnoxious Yes. Some apps do try to guess what the content *could* be. But as I explained elsewhere, that’s more of a workaround. ActivityPub doesn’t allow for markdown, it wants the HTML. @apps
Can we please all stop the "newcomers to Mastodon should visit this and that site, follow these rules" etc? Social networks evolve. Trying to control the behaviour of newcomers is condescending and counterproductive in my opinion. I don't want the Fediverse to become a walled garden. #UnpopularOpinion, maybe. So be it.
I am not poor. I am also not really rich. Some people say I have missed many opportunities to become rich. By changing the license of my open source code to proprietary. With Cryptocurrency. Or now with AI. My problem is: I see through the smokescreens. I prefer to do what I think is right. My conscience means more to me than the numbers on my bank or stock account. I understand and respect you when you think I'm stupid for that. It's just that I can't be any other way. My apologies!
OK. I've skimmed the decision (35 pages, I am not a lawyer). This is NOT based on the EU AI act. The court sees the "AI Summaries" that Google produces as content created and published by Google, hence they are liable when their work produces wrong and harmful statements, as was the case here. So quite a normal decision. Google's argument that they are somehow exempted failed. They have to remove the wrong statements and pay 80% of the costs.
Yes, a local court in Munich (Landgericht I) decided that Google is liable for what their "AI summaries" state. I am reading the full text of the decision to see if this was based on competition law or if it was based on the EU AI act before I go any deeper.
This morning I woke up from a dream where I discussed with @MastodonEngineering that Markdown support for posts and replies is really important and that a new feature called "Reply Delay" where you can force a delay of 1-24 hours before replies on your post are allowed would be really helpful. They looked at me. Looked at themselves. And shouted "Let's do it!" and they coded it all up and we feasted together. I woke up with a smile :)
My private account. Posts/Opinions are my own. Focused on #Open and TCP/ID, identity ownership in the real and digital world. 3D printer. Can solder and repair. Antifascist. European. Policy hacker. Rambles around at times. He/Him/His $argon2id$v=19$m=512,t=256,p=1$SWmoDffV/hOu+/Vii5Nxsw$zYZ5n+cXxZLKoLnXZJjll1JWcCFyiRVli7xOPqu63GM Blocks threads.net Works at Red Hat.