One thing that I hate about MacOS is that CMD+Tab does not cycle through windows of an application. Navigating windows is so fucking clunky on this machine.
(Please do not tell me about your favourite Window manager, my personal machines have run different flavours of Linux for more than 20 years now, I am speaking about the laptop I have to use for my job)
Thinking about the OSI "Open Source AI" Definition and how to proceed. Like say they changed their definition to demand that all training data was "available" (right now you only need to describe it) meaning there are URLs that you can access. Think YouTube Videos or social media posts or whatnot. But not all content is under a free license, some explicitly copyrighted with "all rights reserved".
Would you consider a machine learning system trained on that data still "open source" in the intention of the Open Source definition (https://opensource.org/osd)?
"Still, it’s unclear whether “Just Go Independent” is a sustainable career path for the number of journalists who we need to have a functioning society. But I do know that relying on the passing interest of billionaires to keep journalism alive is not sustainable. And I know that 250,000 subscribers could fund a lot of independent journalists."
(Original title: The Billionaire Is the Threat, Not the Solution)
Another Masto Instance going down. But not because moderation or something but because running bigger instances on Mastodon gets really expensive.
I think this is an issue that we need to put more work into: Not just finding better ways to sustainably fund the operation of instances but also the technical means to make running it on smaller hardware easier. This includes mechanisms to maybe push certain data into "archives" as to not have it in the live database/asset store.
The federated social web is my main social platform (using this account). It's not my identity to defend or shit on it. But I do have many criticisms with regards to culture, product and project management as well as many questions about sustainability (financially and socially). It's important to address those questions honestly, curiously and openly. To accept that currently this is a social sphere that excludes many groups for a whole bunch of reasons (many of them coming from the dominant culture here). And we're not really doing a great job.
On the other hand I find it a bit weird how many people publish their hagiographies of Bluesky here without any deeper analysis. Federated Web vs. Bluesky is way too often an identity thing. About defining one's own crowd as the actually cool ones or whatever. It's not helping anyone really.
Personally I do think that going the VC route again is a structural mistake that is very hard - if not impossible - to recover from. The Bluesky team disagrees. On the other hand I don't feel like the "but we are open source software, you can run your own thing - that lacks core features you expect" is a great sales pitch either.
TBH: While I think that the underlaying platform here needs a lot of work (especially Mastodon the dominant one) personally I'd really love for more people to go a bit apeshit with weird experiments using ActivityPub. This can start with experiments like SFO's "turn every museum exhibit into an AP account/bot" https://millsfield.sfomuseum.org/blog/2024/03/12/activitypub/ but could go a lot further. I think it would do this place some good to be a bit weirder.
If you have a process that reasonably can be replaced by an LLM you either don't fully understand the process (like working on a text is not superfluous but the act of thinking) or that process should not exist.
(Very specific exceptions to the rule exist: Translations or transcripts for example serve a purpose and can reasonably done in a sometimes passable way with LLMs. But in general the rule still stands)
"But we should be extremely clear about this, because it is an inane and even maybe dangerous notion: AI will never “solve” climate change. Even if OpenAI successfully builds an AGI tomorrow, it will never, under any circumstances, produce any kind of magic bullet that will “fix” the climate crisis. "
Beim Spiegel wird geleitartikelt, für ein Verbot der AfD sei es "zu früh", "Zunächst müssen alle anderen Mittel der wehrhaften Demokratie ausgereizt sein."
Dann kommt ein konfuser Mix aus "ja, die haben aber so viele Mandate, und deren arme Wähler fühlen sich dann sicher betrogen" und "die Demokraten müssen sich einfach anstrengen".
Man sieht hier die bürgerliche Denke: Angst vor Unordnung und echtem Konflikt (die armen AfD Wähler die werden dann grantig) und Finger-in-die-Ohren stecken (Wenn C*USPDGRÜNE nur brav ... was eigentlich ... tun, dann geht die AfD weg).
Wie kann man mit einer so geschichtsvergessenen, feigen Position in einen Leitartikel stolpern?
"Wehrhafte Demokratie" heißt nicht, "nichts tun, was jemanden nervt". Wir sehen wie die AfD vorgeht, wenn man sie läßt und wir sehen auch dass einfach zu glauben, dass die anderen Parteien das schon hinbekommen, peinlich naiv ist.
Es ist mir auch egal, dass die Wähler von Faschisten sich von einem Verbot der Faschisten betrogen fühlen. Mir sind die Gefühle von Faschisten ganz grundsätzlich egal. Ums mit Wiglaf Droste zu sagen: "Das Schicksal von Nazis ist mir komplett gleichgültig; ob sie hungern, frieren, bettnässen, schlecht träumen usw., geht mich nichts an. Was mich an ihnen interessiert, ist nur eins: daß man sie hindert, das zu tun, was sie eben tun, wenn man sie nicht hindert: die bedrohen und nach Möglichkeit umbringen, die nicht in ihre Zigarrenschachtelwelt passen."
When logging into Wordpress.org (login.wordpress.org) you have to confirm that you are not affiliated with WPEngine.
While this looks petty and childish it's another sign the Wordpress is in the hands of people who can't find less destructive ways to handle their fucking midlife crisis.
"Taking a scaffolding built out of LLMs to regulate other LLMs’ and calling it “reason” is a surefire way to create a weird, incomprehensible system. Building infrastructure on top of this Frankenstein’s monster of automated statistics should be a terrifying proposition for anyone, not because it will decide to kill us — but because it’s an invisible bureaucracy incapable of deciding anything but potentially given authority to exercise power as if it does. "
Sociotechnologist, writer and speaker working on tech and its social impact. Communist. Feminist. Antifascist. Luddite. Email: tante@tante.cc | License CC BY-SA-4.0 tfr"Ein-Mann-Gegenkultur" (SPIEGEL)