Meta/Instagram launched a new product called Threads today (working title project92). It adds a new interface for creating text posts and replying to them, using your Instagram account. Of note, Meta has stated that Threads plans to support ActivityPub in the future, and allow federation with ActivityPub services. If you actually look at your Threads profile page in the app your username has a threads.net tag next to it - presumably to support future federation.
Per the link, a number of fediverse communities are pledging to block any Meta-directed instances that should exist in the future. Thus instance content would not be federated to Meta instances, and Meta users would not be able to interact with instance content.
I'm curious what the opinions on this here are. I personally feel like Meta has shown time and time again that they are not very good citizens of the Internet; beyond concerns of an Eternal September triggered by federated Instagram, I worry that bringing their massive userbase to the fediverse would allow them to influence it to negative effect.
I also understand how that could be seen to go against the point of federated social media in the first place, and I'm eager to hear more opinions. What do you think?
Personally I'll support instances that choose to tell Zuckybags to fuck right off, and I think the fediverse is pretty well set up to be able to do that.
But I guess the bigger question how is how we protect our information, since it seems like everything that happens here is pretty wide open.
The big companies will all come for places like this and trawl for "genuine human input" to feed their AI cashbabies, and what we create has value. Maybe even the shitposts. So how do we protect that?
There's no way to protect against that. They can still have a scraper set up to crawl the fediverse without threads ever federating with any of the instances.
Yeah, I get that, and I don't think the solution is a walled garden. But I do think it will become more of an issue as these companies get more aggressive about going after any data they can find.
The difference is that Reddit was looking to IPO off that data, here that won't happen. We were the product there.
If you're ok with your data being used, maybe there's a way to opt into that. But I would rather it be something data trawlers had to forcibly ask first rather than just stealing it and then saying "oopsies"
At this point, I really don't know. But I guess the difference is that here, or possibly other instances at least the users can have a say in it. Posting topics like this helps.
That for me feels way better than just having some company give out the terms and having us accept it whether we like it or not.
I feel like this is all still new enough to where we can shape it in a different direction than "growth at all costs"
It's not really fair to call it stealing.
This is an open source platform so you can use the code freely, the developers decided to make it open source because creating tools and sharing them for everyone to use enables us all to work together and build better technology which improves the quality of life we all enjoy.
I take the same attitude when posting here, I'm making a public statement because I believe we as a society make better choices when all voices are heard and that I have something to add to the discussion - if someone wants to take an idea that I have and build in it or incorporate it into a speech they give then I'm happy about it because that's the point of expressing an opinion, for it to spread. If someone learns from what I'm talking about then I'm happy, I don't feel robbed or like I lost anything.
Coming on a free and open source platform then demanding ownership of your text would make even less sense than Bob Ross getting angry that people are following along with his painting tutorials. I get the sentiment and yes ai is changing the world and people will lose their jobs but it's also bringing endless positive advancements which will improve and extend billions of lives - if datasets are locked down by oddly selfish rules like having to track down the original poster and ask for permission then only big corporation's will be able to train LLMs and other ai thus giving them a monopoly.
Also another interesting hypocrisy that a lot of people are missing, meta might be awfull in many regards but they're actually pretty good with open source and wrote a lot of the tools that made things like stable diffusion possible. Taking what they give then complaining when they use content posted publicly just feels so odd to me.
Just to rag on the Bob Ross metaphor, because I absolutely love him, he might not have minded if I followed along with his paintings, but he was absolutely selling some pthalo blue (which is not a bad thing its own right) and what happened to him after he died was absolutely terrible in terms of monetisation without his permission.
I guess this is all sorta tangential to the main thread here, but I think we should start seeing our information in terms of actual value, and even if we choose to contribute to an open platform like this (which I absolutely support) how much is that actually worth.
Kinda like if you work pro bono, it's still good to keep a record of what your billable amount would have been.
Zuck and co have a vested interest in making us think that what we provide is worthless
I really am split on this, I think that modern corporate media has a big incentive to convince people that we shouldn't participate in community projects and that working together for the betterment of all is somehow a bad thing because this is what directly affects their monopoly and control.
That said you're right that other portions of corporate evil benefit by us thinking it has no value.
I do think we need to move away from the pushed ideal that the only benefit to work is money because that really does just benefit the rich - take Wikipedia as an example, writing pages for that isn't going to earn me money but it allows me to live in a world where Wikipedia exists, and it's not just me that gets to use it but everyone trying to do anything now has access to that resource which means whatever I do it's possible that the work I put into editing articles has somehow made it a bit easier or a bit more possible.
This is even more true with open source software, it's possible that a project I've done some small thing to help has grown to benefit the people making point of sale systems that allow a cafe I like to reduce costs and stay in business or maybe the veg was grown by someone that learned how from a YouTube video made using open source software...
AI is already making so many more things possible and when natural language control is better evolved it's going to have benefits like giving every human on the planet access to world class healthcare, education and tech support, it's going to allow anyone with an idea to make it reality and to allow open source developers to create really amazing things that we can all use.
If harvesting my throw away comments and old wiki edits can create a tool that will allow me to sit down and describe the electronics projects and coding ideas I want to make then they're welcome to it.
Hey I just wanted to say first of all, this has been great discussion, I really appreciate it. I guess it all goes back to the original purpose of the internet, which was never meant to be turned into some heavily monetized cash cow for a handful of companies. Open source has created some amazing tools, and I'm a big fan of things like GNU and cc licenses. For me, individuals using whatever I write as a tool is no problem at all. But when it comes to a big company that just wants to scrape data for a LLM, that's not the same.
This is wild thinking, but like say if they wanted to scrape the lemmyverse, ok, but they had to pay so we could maintain servers etc. We'd all benefit.
Yeah maybe a share-alike clause would be best, use the data if what you're making is properly free otherwise pay