General

Yeah, I think that long-timers really don't understand just how confusing things are for new people. On Mastodon for example everybody's internalized the "if you click a link and it takes you to another, just cut-and-paste the URL to the search box" trick, and of course you don't see all the conversation because of how federation works so you have to remember to check the original post, etc etc. In A faux "Eternal September" turns into flatness I wrote "Accuracy aside, the analogy to Eternal September highlights the gatekeeping and disdainful attitude many (although certainly not all) long-time Mastodonians had for newcomers. In The Year September Never Ended in net.wars (1997), Wendy Grossman had noted that “AOLers weren’t (necessarily) stupid; they were software-disadvantaged." The same is true for newcomers to Mastodon. Mastodon's official mobile apps are medicore, and Mastodon's web UI has some very rough edges and counter-intuitive functionality – even for somebody like me who's been using it for years. Ernie Smith's No More Eternal Septembers discusses the gatekeeping aspects of the term in more detail, and Don't tell people "it's easy" goes into It's complex, because it's certainly true that newcomers bring some unhelpful habits from old social media -- for example not putting alt-text on images. It would be great to find ways to help them realize that's not how we want to do things here. But long-timers have some problematic habits too, and not just the gatekeeping. In Erin Kissane's [Mastodon is easy and fun except when it isn't](https://erinkissane.com/mastodon-is-easy-and-fun-except-when-it-isnt, an informal survey of why people bounced from Mastodon, the #1 item on the list was "got yelled at, felt bad". And Don't tell people "it's easy" goes into detail on one specific bad habit. More positively, though, federated music platforms haven't yet experienced this mass influx, so have a chance to avoid these problems! Specifically in terms of maintainers taking heat, I don't think that's where most of the criticism comes from long-timers. Looking at Dan, for example, the person who highlighted that he still hasn't done the necessary recovery from the security bug has been active here since 2017 if not longer; the people I see regularly highlighting his history of anti-Blackness are long-timers; and most of the people who signed onto the letter urging NLNet to pull their funding for Loops have been around for a while. Of course Dan gets a lot of appreciation from old-timers too; many don't know about this history, others know about some of it and see his contributions as more important (and even the open letter highlighted his many contributions).

TALER Integration Community Hub comes with free software development grants to integrate GNU Taler with your software. Coming up with a federated solution that can be used by all the projects here, or maybe combining options to serve the music community around a common proposal might be an interesting take. Certainly there is a number of music-specific use-cases that can be shared beyond the simple implementation of micro-payments for singles. Note that GNU Taler has KYC builtin thanks to the use of IBANs and strong cryptography, which drops the cost of operation down to something that enables micropayments: transaction fees are a fraction of those used in classical payment systems such as Visa/Mastercard, Stripe, Paypal... And it preserves your privacy as a buyer: nobody knows you're a fan of S̵̖͑̋̆͛ŭ̵̱̃p̷̢̥͕͕̞͒̄́͛e̴̛̘̪̣͑̈́͜r̸̬̣̳̫̀̆̆́s̶͉̟͛t̷̘̞̘͌̾͊̏̄a̷̛̭͑̋r̸̫̥̮̘̾̚.

Also on the topic of scrobbling, here's a bot account on Mastodon that posts about what song the operator is listening to. Based on signals it receives from their home assistant software. https://mastodon.social/@TomsMusic

Those are great features! wow I asked about a "syndication" capability (something like a collection of servers that could rebroadcast each other's content when nothing else is playing) and some of these features lean in that direction. Each owncast being a mini-directory is cool. A decentralized directory of sorts.

Hmm. Yeah, sometimes when a new platform is positioned as "an alternative to [corporate] platform", users of the service aggressively bring that culture with them even if the design is somewhat different. That's kind of how I feel about Bluesky (and maybe I'm being unfair, since that has algos too which may subtly manipulate behaviour without you realising, so maybe it still is part of their design), but I didn't want to be on something else where people are replicating bad patterns and desperate for engagement at the expense of everything else (or just "going where the majority goes"). I do bridge my account though and try to entice people to the federated side. I hope Loops attracts people doing interesting projects who've perhaps had enough of Instagram rather than those looking to faithfully replace it, if that makes sense. I did see as well the dev getting annoyed at a comment about security and I can understand that it must be frustrating having put loads of effort into something and seeing things being picked apart, but again, maybe that's because previous social platforms have conditioned us all to read things as if we're constantly gearing up for an argument.

Ok, this is ridiculous (but pretty funny)

Anyone have experience with LiquidSoap for broadcasting video? I'm poking around trying to figure out how to establish an always-on MTV like FediTV for music videos. I tried ffmpeg and that worked up to a limit -- it has a documented but unimplemented feature (say what?) for pulling metadata out of video files and getting them onscreen (artist, track, album). It also doesn't handle failures all that well for 24x7 broadcasting. Maybe it could with some magic but... That is what LiquidSoap was designed for. With LiquidSoap you could setup a constant stream that pulls from a randomized playlist of videos and inject periodic station identification supersede regular broadcast to allow for live events fallback to known good video when something breaks (no dead air) transcode on-the-fly But still struggling to pull metadata out of the videos and get it onscreen - which is deal breaker - this is about artist promotion as far as I am concerned. I worry this is down to the same missing feature in ffmpeg (since it uses ffmpeg internally). (p.s. one interesting side effect could be... syndication... or something like it. I've noticed that there are many owncasts that go unused most of the time... i believe you could set this up to syndicate the stream to all of them and then allow any of them to pick up a live stream on demand - talking about gravitons, tibr, labr, basspistol, being able to opt-in to syndication but that is for another day) Anyway, does anyone have experience setting up LiquidSoap and getting metadata onscreen? Complete examples are hard to find... and I declare, functional programming is not my strong suit. See what I did there?

Yeah, I think they use cargo as a distributor in germany, and cargo uses audiosalad. There wasn't an explicit ask, but so yeah it sounds like they use audiosalad to then send music on to DSPs (spotify etc). It's interesting that AudioSalad is part of Merlin etc. I think that this is probably the standardization--if you can work with AudioSalad, you can work with DistroKid, etc.

Isn't the goal here to find ways to break our dependence on extractor money, rather to find ways to make it easier for more artists to fall into their financial pitcher plants?

FYI Team Bonfire are working on ways of doing privacy-respecting geolocation in the fediverse, and Evan of SWF is seeking interop testing for various location-aware fediverse tools he's been working on. The work they're doing could be relevant to making this idea work in practice? EDIT: Here's an example of a kiwi musician doing a cultural experiment with tying musical works to geolocations; "Alongside being a stand-alone album, Driving Creek is also available as an on-site listening experience for visitors to Driving Creek Railway. A QR code link to the album is accessible to visitors on the station platform."

Mel: I feel strongly that artists should NOT have to use social media, for their work to be seen and heard. Even apps like Mastodon Thank you for calling out that Fediverse social media is still social media, and this comes always with trade offs. So... in the old days of music creation, "all you needed" was a meeting place where interesting artists would get together and share. Could be a music club, a magazine, a label... First they met, others would join, and they would figure the next steps after meeting, creating and sharing music. Is it possible that we are putting too much emphasis in software architecture and tech and too little on actually choosing a place (online) and call it our home, our shared studio? Look at the Bonkwave thing, their made... a hashtag their home and you have what looks like a fun and thriving community now. Mel: continue using CC music only CC music makes prototyping and testing easier, because money and restrictions to play and share don't play an important role. Once some good exists for the CC music use case, and a first user base exists, building the features for money and restrictive licenses should be easier. But they should be in the roadmap of the open source project from the start, so nobody feels cheated, enshittified down the line.

@lorenzosmusic If you've posted about it on the fediverse, link us so we can boost it. Or if there's an official account for the event, link us to that. Or both.

There's been some questions so I'll answer them here. Firstly I'll say that this is just the submissions stage. Tracks submitted obviously remain the sole property of the rights holder and the artists maintain rights. The tracks can be exclusively made for this or an older teack as long as the submitter owns all rights with no conditions. Once curated the artists retain all rights and are in effect giving their permissson for the compilation to prsent their work. There are no fees or charges from them to me, or from me to them. Exclusivity: Tracks made specifically for this compilation will be used by mutual agreement that they arent released by the artist for 3 months after publication. This is simply so that if artists have their own fanbase it helps boost the cause of the compilation. The more original tracks the better. It builds a momentum in the community. Format: bandcamp download, vinyl and CD. Profit and payments: as mentioned above, theres no fee or charges in either direction. When people buy the compilation, all profits will go to Amnesty International. This means that every penny/cent after sellers fees and bandcamp cut etc will go to the charity. Elasticstage (vinyl and CD) takes about 80% of the money paid (because they are producing the physical media, on demand, for no charge). I hope this clears some things up?

I said I was going to simmer down for a while in the search and discovery topic. But I do want to address this, because it's the crux of the issue. Our analysis of the problem is going to determine our theory of change. So, what I said was; Discoverability on the Fediverse and Thought Dump We started with a world of record shops and many small websites. If that was adequate to the purpose, the DataFarming platforms would never have taken hold in the first place. And @timglorioso replied; Discoverability on the Fediverse and Thought Dump I contend the success of DataFarming platforms is due to the stickiness and frictionlessness of their commercial design, not because they are better for music and musicians I agree that they have never been better for music or musicians, in the long term. The comparison to junk food is apt. But they came to dominate because at first, they provided a much better experience for audiences. Once audiences were locked in, they started making it worse for them, but a better experience for their commercial partners. Mainly the corporate music publishers. As they too got locked in, the platforms could make it worse for them too, and start extracting the value for their shareholders. This is Cory Doctorow's basic theory of how enshttification works. He also talks about 4 forces that can prevent or roll back enshittification; credible exit, self-help, labour power and uncaptured regulation. I don't always agree with Cory, especially the casual misrepresentation of other schools of tech criticism. But his thesis on how and why platforms enshittify matches what I've been observing online since the DotCom bubble.

They organise on Matrix & Fedi, a few of us here are in the group!

Mel: We were talking to them for RFF community a while ago. But it transpired that they needed us to sort funding for it first! This isn't uncommon for tech projects in the solidarity economy. They don't have access to the VC pipeline, and the potential for future returns through IPO or acquisition by a larger competitor. So money needed for R&D, beyond the time and skills that enthusiasts volunteer, generally comes through grants, funded by governments or well-heeled not-for-profits. One thing I've noticed by watching this space for a while; there isn't a huge overlap between people who are good at developing new kinds of software, and people who are good at writing grant applications. Even in that overlap, writing multiple applications (to increase the chances of getting one) takes time away from software work, which is kind of counterproductive. So it's often better for people who need software adapted to their use case to write the grant applications. But they often don't get the grants either, because they struggle to prove that a working solution can be delivered for the money granted (why is another story). Meanwhile, another thing that makes it really hard for developers to get grants is the inability to prove demand for what they want funding to build. What I'm ambling towards here is that if a group could be coordinated here to put in a grant application for building a Bonfire Music flavour, with thorough technical input from the Bonfire team, there's a good chance you'd get funded. I'd suggest making a number of small grant applications to both tech-based funders (eg NGI0 and NlNet), and music-based ones. Random idea that may or may not work; include the full list of funders you intend to apply to in the grant applications. Encouraging them to backchannel and co-fund a grant big enough to cover the work needed.