Replying to
@andrmr@mastodon.social
@andrmr@mastodon.social ja, in so einem Dorf habe ich auch schon zwei mehrstündige Spaziergänge damit verbracht die ganzen Hausnummern und so einzutragen. 😅
Remote
Duco
@duco@norden.social
Software Entwickler aus Hamburg
0
Followers
0
Following
45
Posts
Joined September 23, 2018
Posts
Replying to
@andrmr@mastodon.social
@andrmr@mastodon.social wenn ich sehe, dass da Hausnummern fehlen, trage ich die nach, sofern ich das kann. Andere Karten kosten halt Geld und sind nicht unbedingt genauer. Aber der Staat könnte auch mal hin gehen und vom Grundbuchamt oder so jemanden die Daten in OSM eintragen lassen.
Replying to
@amxmln@mastodon.design
@amxmln@mastodon.design I think Atom was killed, because Microsoft bought GitHub which resulted in Microsoft owning 2 Electron based code editors.
Replying to
@hochbahn@norden.social
@hochbahn tragt ihr das auch in der OpenStreetMap ein? Eingänge können mit ref=* die Buchstaben hinterlegt kriegen. https://wiki.openstreetmap.org/wiki/Metro_Mapping
#hvv #osm #openstreetmap
Replying to
@epicenter_works@chaos.social
@epicenter_works ich glaube für die meisten ist es irritierend von OpenStreetMap zu sprechen, weil sie denken das wäre das gleiche wie Google Maps, es ist aber etwas vollkommen anderes. Google Maps ist eine App/Anwendung, mit der man navigieren, Orte finden und bewerten kann, OSM ist eine Datenbank wo man höchstens Orte finden kann. Deswegen würde ich davon sprechen Google Maps durch Organic Maps, CoMaps oder OSMand zu ersetzen.
Replying to
@weizenspreu@chaos.social
@weizenspreu@chaos.social @FemmeNoire@chaos.social @kventil@chaos.social ah, danke, das hatte ich übersehen. Aber wtf. Warum starten sie §2 Abs. 2 nicht mit §4 oder negieren die Bereiche von Abs. 3? Dann hätte Abs. 3 nicht gebraucht.
Replying to
@weizenspreu@chaos.social
@weizenspreu@chaos.social @FemmeNoire@chaos.social @kventil@chaos.social stimmt das aus dem Artikel denn überhaupt? Laut dem neuen §2 (2) aus dem Bundesgesetzblatt gelten §§ 3 bis 52 ja nur im Spannungs- oder Verteidigungsfall. Solange wir so einen Fall nicht haben, muss man also nichts beantragen, oder?
Replying to
@kventil@chaos.social
@kventil@chaos.social hmm. Ich glaube ich bereite schon mal Anträge für all unsere Nachbarländer vor, dass ich im Zweifelsfall eine Genehmigung habe, falls ich spontan für ein paar Monate in eines der Länder reisen möchte.
Replying to
@alexmu@social.vivaldi.net
Replying to
@emacsomancer@types.pl
@larsbrinkhoff@mastodon.sdf.org @petealexharris@mastodon.scot @tml@mementomori.social @Yendolosch@23.social @emacsomancer@types.pl in the sense of life hacks or food hacks this is an AI hack. So the AI has been hacked.
Replying to
@mawr@plush.city
@mawr@plush.city while Spotify pushes AI generated music, Deezer is doing the opposite and tags AI generated music, and filters them out for their recommendation algorithms. So you only listen to AI generated music there, if you explicitly decide to listen to them. https://newsroom-deezer.com/2025/01/deezer-deploys-cutting-edge-ai-detection-tool-for-music-streaming/
Replying to
@theVedek@chaos.social
@theVedek@chaos.social @geeeero@mastodon.social @marcuwekling@cultur.social speziell für Linux gibt es auch die Kampagne #EndOf10, wo man viele Leute findet, die einen bei der Installation von Linux helfen. Mehr Hilfe ist da auch gerne gesehen. Am Besten mal beim lokalen Repair Cafe, LUG oder chaos Club fragen.
http://endof10.org/
Replying to
@trochee@dair-community.social
@trochee@dair-community.social @beeoproblem@mastodon.gamedev.place sorry. Was reading and answering chronologically. Will untag you after this answer. I'm not saying all information must be free. I understand that artists, writers, etc. are concerned about their job. I'm happy to listen to them. I'm not happy if people lie about something got stolen, if that wasn't the case. So far no one explained how the stealing happened.
Replying to
@trochee@dair-community.social
@jleader@mastodon.social @Daniel_Blake@mastodon.top @helge@mastodon.social @trochee@dair-community.social it's not stealing. I can imagine, that it is maybe possible to get some text blocks out. Like when you ask a friend what the article was about and they repeat it. Whether it's to an extend, that it's copyright infringement, judges have to decide. There is a scale of plagiarism between copying word by word and speaking the same language. Not all plagiarism is illegal. But I can imagine, that AI companies might have violated some copyright laws at some point.
Replying to
@beeoproblem@mastodon.gamedev.place
@beeoproblem@mastodon.gamedev.place @trochee@dair-community.social so you tell me, that OpenAI destroyed the lock to the working room of some artists, gone in, took some paintings, put it into their van and drove away? 🤔
As far as I know, they used publicly available material to train the AI. Therefore the author decided to make it public. Or if not, it's not that companies, but the publisher, who did something wrong. But even that wouldn't be stealing, but copyright infringement.
Replying to
@Daniel_Blake@mastodon.top
@Daniel_Blake@mastodon.top @helge@mastodon.social @trochee@dair-community.social yes. That's not stealing. DVD sellers called it stealing, but they lied. It never was stealing. And actually download videos isn't illegal (at least in Germany), as far as I know. The only illegal thing is to share this videos with others, which is copyright infringement. But as OpenAI and others don't provide the training material as download, they don't do copyright infringement. So it's definitely not stealing. Maybe they violate other laws I'm not aware.
Replying to
@beeoproblem@mastodon.gamedev.place
@beeoproblem@mastodon.gamedev.place @trochee@dair-community.social if it's stealing, which person was stolen from and which object did they possess before, which is now in the hands of the thief?
Replying to
@trochee@dair-community.social
@helge@mastodon.social @trochee@dair-community.social if it's stealing, which person was stolen from and which object did they possess before, which is now in the hands of the thief?
Replying to
@trochee@dair-community.social
@trochee@dair-community.social
Is it's mere use a climate disaster? No. It's not good, but other things, like driving a car, using Airplaines, having gas heating, etc. Are way worse. And compared to using bitcoin, using LLMs is like nothing.
Is it founded on theft? No. Theft means that you steal something, that it's not there afterwards. It's not even copyright infringement, as they don't publish/distribute copyrighted material.
There are ethical ways of using AI, like following the green Nextcloud Ethical AI Rating.
Replying to
@StaticR@guild.pmdcollab.org
@StaticR@guild.pmdcollab.org @jasongorman@mastodon.cloud do you have an example of a question answered by the RNG list?
Replying to
@BrendaAnna@mastodon.social
@BrendaAnna@mastodon.social @brucelawson@social.vivaldi.net most trans people fit into the 2 gender status quo, though. They are men or women.
But I agree that there are people fighting trans and other queer people. There is still a lot of fighting for trans rights needed.
Replying to
@BrendaAnna@mastodon.social
@BrendaAnna@mastodon.social @brucelawson@social.vivaldi.net Genetic testing costs money. But it's not needed here because trans is defined as having a gender different from the sex assigned at birth. So when a baby is born, they look at it and decide what it is and put it on the birth certificate. If someone has a different gender from what they decided at that time, they are trans.
Replying to
@tiberiustribun@mastodon.social
@tiberiustribun@mastodon.social @kuketzblog@social.tchncs.de Wenn du ein Buch aus einer Buchhandlung stiehlst, ist das kein Urheberrechtsverstoß. 😂 Für einen Diebstahl muss dem bestohlenen hinterher der Gegenstand fehlen.
Und nein, KI-Firmen geben keine Fetzen davon wieder, sondern erfassen statistische Zusammenhänge zwischen Worten und wenden das gelernte dann an. So wie Tolkien in Herr der Ringe vermittelt hat, wie Orks aussehen und Rowling das in Harry Potter ähnlich beschrieben hat. Das ist kein Urheberrechtsverstoß.
Replying to
@gunchleoc@mastodon.scot
@gunchleoc@mastodon.scot @kuketzblog@social.tchncs.de da sie die Raubkopien ja nur konsumiert, nicht aber vervielfältigt, verbreitet oder öffentlich vorgeführt haben, verstößt das (meinem Verständnis nach) nicht gegen das Urheberrecht. Gegen das Urheberrecht verstoßen hat ja derjenige, der das zur Verfügung gestellt hat.
Replying to
@kuketzblog@social.tchncs.de
@kuketzblog@social.tchncs.de ich verstehe dieses Urheberrecht-Argument nicht. Das Urheberrecht regelt die Veröffentlichung von Werken und die Anerkennung des Urhebers. Nichts davon machen LLMs. Das einzige was diese Firmen machen ist Inhalte abzurufen, um sie fürs Training zu verwenden. Bei Menschen nennt man das lesen. Das Urheberrecht regelt nicht, wer ein Werk lesen darf. Wo verstoßen die also gegen das Urheberrecht?
Replying to
@transworldexpress@milliways.social
@transworldexpress@milliways.social could you work on making the design mobile friendly as well? It's a bit hard to read on a smartphone.
Replying to
@Lkdc@dmv.community
@Lkdc@dmv.community @MurmeltHier@climatejustice.social @transworldexpress@milliways.social it depends. There are more extreme ones that are against all immigrants, but they are very rare. I would say most anti-immigrant people are mostly against people from Africa and the Middle East and are worried about them using our social systems. They are not attacking them, just want the state to move them out of the country. Somehow they think they would get more, if there are less immigrants. I would say avoid east Germany (except Berlin).
Replying to
@sarajw@front-end.social
@sarajw@front-end.social yes. That's a great feature. There is just one problem. There are some people I would like to follow, because they write sometimes interesting things, but they post 20 things every day. I don't have time to check so much posts of multiple people. So I would like to have a filter, that just adds the most important posts of them.
Replying to
@derickr@phpc.social
@derickr @ed_hawkins @skaphle Yes, thanks. Actually the CC-BY licence requires the user to also link the license. So if I'm correct 5 things have to be provided with each use:
- creator of the image
- link to the website
- the licence (CC-BY)
- the link to the licence
- if it was modified, a hint, that is has been altered.
That's a lot if I for example want to use it in the profile picture, like others have a 🏳️🌈 there.
Replying to
@ed_hawkins@fediscience.org
@ed_hawkins as this graphic is CC-BY, I'm wondering which author has to be named. Also it seems you shared it without naming the author, just providing the link. Also you are required to add a link to the licence and to name it. It would have been cool if it would be CC-0. Maybe it is actually, as it's quite simple. At least in German law content is public domain if it's very simple.
But I would avoid using it, because of the hustle with the licence.
Replying to
@ink@turtle.garden
@ink@gts.turtle.garden @yabellini@fosstodon.org @skylarkingmullet@piaille.fr not according to German law, I think. As far as I understand it, in Germany you can not give away all rights. The author always is the owner of their creation and can only give away usage rights. Therefore public domain exists here only for content that is either too small/simple to fall under copyright or the author died more than 70 years ago.
When I change a WP article it says (in both de and en), that I agree to publish that under CC BY-SA 4.0 and GFDL.
Replying to
@yabellini@fosstodon.org
@yabellini@fosstodon.org @skylarkingmullet@piaille.fr at least under German law, the author of any text has the rights on it. Every Wikipedia article has authors with copyright on it. But they licenced it under a free licence, so everyone can use it. So as every text written by a human, every photo taken by a human and every image painted by a human is copyrighted, OpenAI is correct, that they can not train the AI without that. That does not mean, that the texts are behind a paywall.
Replying to
@skylarkingmullet@piaille.fr
@skylarkingmullet@piaille.fr @yabellini@fosstodon.org so they sued OpenAI. Well people sued government for legislation of masks against Corona. Just because someone sues someone doesn't mean they are right. Let's wait for what the judges say. The second part seems to be about data protection, not copyright.
Replying to
@yabellini@fosstodon.org
@yabellini@fosstodon.org I can not read the article as it's behind a paywall and the other document is 69 pages long. I will not read that. If you want to say something with it, say waht you want to say. Depending on what you will say, I will think about if I want to check that with the provided sources or not.
Replying to
@yabellini@fosstodon.org
@yabellini@fosstodon.org SciHub makes papers public that are behind paywalls. I agree, that they shouldn't be behind paywalls, but it's completely different to OpenAI.
I think they used mostly sources that are public anyway, like Wikipedia, etc. They also didn't publish them but trained an AI with it, that creates new texts. So they did a remix in a way. Remixes are handled differently in copyright law.
"The corpus [GPT-2] was trained on, […] 40 [GB] of text from URLs shared in Reddit" https://en.wikipedia.org/wiki/OpenAI
Replying to
@christian@social.jcs-net.de
@christian @t3sserakt @johl Algorithmen können wir ja in Mastodon beliebig einbauen. Oder in einer anderen Software, die ActivityPub spricht. Selbst im Client kann man da viel machen. Ich sehe da jetzt keinen Vorteil bei Bluesky.
Replying to
@reina@cutie.city
@reina @AmyZenunim your posts are public. Everyone can already access them, if you block them or not.
If you DM someone on Threads that's an interaction. So my claim wasn't false.
XMPP was never mainstream. Before and after Facebook an Google used them it's used by a small group. History can repeat with Threads, true, but it isn't necessarily the case. And blocking them doesn't make the Fediverse more popular either.
My claim is, we should use them to improve the Fediverse.
Replying to
@reina@cutie.city
@reina @AmyZenunim of course my Szenario is a positive one in contrast to the negative Szenario it reacted to.
Why should anyone block Threads for privacy concerns? If you don't interact with people from Threads nothing is shared with them, that isn't public. An instance doesn't have to block them, so every single person can decide for themselves. Even if you interact, there isn't much shared.
Replying to
@AmyZenunim@unstable.systems
-
Meta embraces ActivityPub with their Threads App: "Look all that people"
-
People on Threads speak with all the nerds on Mastodon that speak bad about Threads and explain that Mastodon is much better because of privacy and no ads.
-
Over time people join Threads but also leave it after they understand the Fediverse.
-
A fork of Mastodon looks like Threads and makes switching easier.
-
Meta quits Threads and the rest of their users switch to other Fediverse applications.