I like FOSS (free and open source software). Have contributed bug reports and feature suggestions to open source projects that got accepted (e.g. the ability to block instances on lemmy). Check out my github if you’re into that kind of stuff.
Pretty sure you want to use the AGPL and not GPL.
The ideal IMO will be some kind of experiment (probably something you should opt in). give people feeds and have them be ranked on a 1-10 scale and see which is better and do a Statistical hypothesis test to make sure the differences are not due to luck.
That is also a product of many variables (including software quality). maybe time to go back to the drawing board.
For this project I’m mainly testing distribution models
What do you expect to learn from those so called tests? (no offence)
The big questions in closed vs open is that there are different scenarios :
for closed source -> less competitor -> more users -> more money -> more investment in the project -> better product.
for open source -> more users want to use it and contribute to it -> better software -> more users -> more potential for making money.
The problem is that for the outcomes you want to track (more money or better software). there are so many variable involved that influence those outcomes so it’s hard to deduce that the license is improving the outcomes or making them worst.
It’s another case of OSPNS - Open source projects names sucking .
You can register on ibis.wiki, open.ibis.wiki or other instances.
Correction, they are not listed as open to registration on fediverse.observer and they don’t have a register button on the front page. unlike fedipedia which is opened.
Also some instances appear to be missing from fediverse observer (i see at least opentextbooksolutions.org not appearing). you could feed it manually which should work (i did it for opentextbooksolutions.org) but a automatic addition will be better. also a lot of the other instances listed on the page you linked appear dead.
List of servers (pretty small)
Not really, i did play around with it and added a few reviews and saw nothing like it.
neodb has other stuff besides books (games , movies ,podcasts, music ) , also bookwyrm license isn’t OSI approved which i don’t like (not looking to start a holy war on this, form your own opinion).
it’s just that it doesn’t work efficiently enough.
Yeah, but we live in a world of limited resources. in particular labor and specifically knowledgeable linux nerds willing to answers questions for free. If everyone will have that mindset there won’t be a lot of time left to answer the difficult questions .
With that said i agree that occasionally if its done its probably no big deal, there is also linux 4 noobs for those who want to ask some questions to help getting started with linux.
Active users and number of servers also recently reach all time high.
If you want to learn how to create a good open source project for the fediverse. i think peertube is the best role model for that.
TBH i think you’re overthinking it, funding software development and running businesses like open source software development is often driven by self interest (even if it’s not easy to accept) . Like in software development part of it is throwing stuff at the wall and seeing what sticks. So trying to transition to more closed model is expected (some of the projects you mentioned went back to being open source).
Sure i have my opinions about software licensing but for me open source is good enough. if something like that will happen and the software is good a fork will be made. That is a acceptable risk-reward calculation to me.
I am considering applying for a grant to fund the development, but there may be complications, especially given that I am a Ukrainian citizen currently living in Ukraine. I am not yet sure what specific difficulties might arise.
Maybe contact framasoft, the creators of peertube. maybe they could help or even provide employment .
Therefore, as an alternative, I would consider fundraising directly from interested users and the community to support development. Since you mentioned fundraising, I would like to ask how you feel about having an option to financially support the project’s development
I recommend giving in a shot, while trying to use lemmy approach of having a good fundraising process with a pop up specifically . I actually tried to write a learning resource for funding open source so there is a lot to say about that. but you would still need a relatively large number of active users. Lemmy for example has around 37K monthly active users and earns around 3.6K a month. so around 10 active users per dollar. I looked at other fediverse platforms and i think it is one of the most effective fundraising processes when measured using active users per revenue ratio.
You could also just put ads in it and maybe a option to pay to remove them . openfront does this and still seems relatively popular. FOSS is nice but eating is better.
Would you personally consider becoming a financial contributor? If so, what monthly amount would you potentially be willing to contribute?
I am afraid i am not a very good case study. i am not really a peertube users and already spend a relatively large chunk of my time on FOSS so i don’t want to invest in it too much.
Regarding your points about benchmarking the recommendation system and curator-based feeds, I am not entirely sure I understood what you meant. Could you please clarify?
Regarding elo ranking something like a page on the website where you are shown two pages with recommendations from two different recommendation systems and you would vote which page and system provided better recommendations . You could also just have one feed and see which recommendation system has a better viewed video to liked video ratio.
Regarding curation. some account invites a bunch of people to act as curators. then something like the most liked videos this group of people had in the last 24h or week get shown. something like that.
If you are looking for free hosting, maybe checkout tuxfamily.
You could also put some links for donations (liberapay , open collective etc) and fundraise the money for better hosting outside of ukraine (and maybe some could go to ukraine government though taxation ). I saw a patreon but no mention that the donations could be used for better hosting which could motivate people.
If you don’t want to deal with taxes, i think using some non profit could help with that (opencollective , Software Freedom Conservancy , spi etc ).
I would like to see some benchmarking of the recommendation system, maybe with a elo ranking like done for AI .
allowing feeds by teams of curators is an idea i would like to have explored (it could be benchmarked using surveys or elo).
This seems really good btw.
Why not link the source code from the website (or did i miss it?)
Which is not useful if those users are people who try out the platform and then abandon it, or worst bots for state actors.
They have a explanation here. they claim part of the problem was banning people who are harmless and not homophobes (they show at least one comment ).
Maybe it is better to avoid duplication of effort and contribute to OSGL.
You just need to add a “mature” tag to it and a option to filter by it. contributing to it is easy as far as i can tell.
I guess maybe set up a community on lemmy. have the client automatically make posts to it. or comments to a specific post?
Or maybe have a bot clients will send messages to?
Lemmy has a bunch of libraries for automation so that could be used to extract the information you need for the test.
Of course that probably makes it easier to feed fake data (e.g. if china wants to make lemmy.ml more popular).