Finally chiming in. I think there are two things at play here:
SaaS
Plugins for interaction between LT and other platforms
As for SaaS, I think it should be decoupled from the current LT team. A company offering LT SaaS would be welcome to contribute back upstream like normal, but the requirements of developing and maintaining a SaaS system often are in conflict with host it yourself requirements (e.g. tie-in into billing and feature limiting systems). I intend (in the coming year) to investigate LT SaaS as an extension of my Linux hosting company. I don’t know the feasibility at the moment, both in terms of code and business financially. I know we are stripping much of the old SaaS code out of LT, but I doubt it would have been useful anyway. There is lots that require designing and integrating different platforms and LT, but having those integrations are useful for more than just SaaS.
Which brings me to thing 2, plugins. A stable, full API for LT would go a long way to allowing people to develop integrations with LT, such as extending the Radio Station WordPress plugin. I think the next step in this regard is to start fielding a discussion around what would be useful in the API and plan out how to go about re-working the existing (semi-functional) API to achieve this. I have created a thread to start that discussion
Regarding SaaS, as far as I understand it there are already some managed service providers based on legacy upstream that have started trialing updating to LibreTime for select customers. What I’ve seen so far really looks like managed services which IMHO lack some of the automation usually associated with actual SaaS. Preferably such solution providers would also be involved with the community, so please speak up if you are one of them and lurking
I’m fully on board about the API, it’s one of the main issues that has been holding off me implementing a proper Wordpress plugin. While workable, live-info-v2.json simply doesn’t cut what would be needed for a full integration. I’m sure lots of other (even SaaS builders) will profit from a proper API that could also be useable as a control plane and to extract billing data.
BTW, the FreeIPA integration I did some time ago could also be a good fit for anyone wanting to build out a SaaS while having their users in a System like Azure Active Directory. At RaBe we’ve only used it with our FreeIPA/SSO environment but there is no reason it shouldn’t work with any generic IPA system. All in all it is rather enterprisey and I would probably focus on something OpenID Connect based nowadays if revisiting the auth thing (if anyone wants to work on OpenID Connect: ping me I have some ZF1 specific snippets that I can either point you to or publish and then point you to).
It would be good to come out of this discussion with a clear idea of what would be in scope for the LibreTime team - from what I can gather, any sort of hosted service offering would probably need the following -
Plugins for interaction between LT and other platforms (maybe WordPress and Drupal to start)
Not a product manager or developer myself, I’m probably missing a lot. What else could be considered in-scope for the LibreTime team? Could/should we create a campaign on the LibreTime collective to help contributors deliver whatever is required?
I think the ideal would be to provide a solid base for partner service providers to extend. We also want to make it easy and obvious for them to support and help sustain a stable, actively maintained and updated LibreTime.
It is still very much just a thought in my head - when I get a chance to start ironing things out (probably closer to beginning to middle of next year) I will certainly start talking about it here
A proper API would be very helpful for many aspects of LT and LT integration
Our friends at AzuraCast just made their app available. The installation and updates is fairly straightforward. Think this would be good exposure for the project.
Based on these Twitter and subsequent discussions, we could work with global and local organizers to leverage or extend the global do_action hackathon program as a way to direct resources towards WordPress integration and related plugins such as Radio Station.
We can also work with WordPress and Drupal organizers to extend the format to building Drupal based sites and directing resources towards modules and integrations.
Considering we’d be able to tap into an established, well documented format and collaborate with the WordPress global community team, maybe we can test the waters by finding non-profit community radio stations in need of a new LibreTime powered site.
I’ve been pitching LibreTime + OpenProducer as a Twit.tv type of solution to Open Collective and partners like SustainOSS. Anyone interested in providing something like this might want to join #openproducer and #sustainoss in https://slack.opencollective.com/
The beta that inspired this post was more of a platform support offering with the option of a DIY SaaS as an outcome.
The initial description needs some clarity, but the idea is similar to what Robb is mentioning.
What if stations or podcasters paid a TBD monthly amount for setup, platform updates, and maintenance until LibreTime was at the stage where it ‘could’ be offered as SaaS? What if the beta also included early-stage access to an open-source website platform as an option? What if hosting and bespoke development was decoupled from a beta?
So as far as I know no stations running LibreTime have expressed interest in paying for support or development. This could be due to a number of reasons. One possibility is the type of stations using LibreTime are happy with it as-is and don’t have any money in their budget to support a project they can download for free. Or perhaps we haven’t been proactive in soliciting donations from users of LibreTime and they would be willing to use money to support the project if they were asked more frequently.
The real question I think is even if we were able to get money from a number of stations is whether this would be a sufficient amount of money to hire developers full-time. I suspect that it would be a rather small amount in the grand scheme of things and nowhere near the amount required to sustain payroll.
I suspect that the users of LibreTime would be able to help more if they were to submit testimonials about how LibreTime helps them serve their mission and helped us connect with potential non-profit funders that support development of independent media projects, like those who originally funded the development of Airtime.
One department we are lacking in with comparison to the originators of Airtime is in terms of marketing, promotion and branding. Plus as an organization we are primarily focused on the code which I think is fine and appropriate. We just need to remember what happened to Airtime and how the free & open-source version was more or less abandoned because development was funded almost exclusively by a single entity.
I think probably the most useful thing that we could do rather than trying to make money from hosting would be to reach out to stations that we know are using LibreTime and compile a contact list so that we can get in touch with them regarding new versions, donating to the project in terms of code, word of mouth or promotions. Just searching the web for LibreTime shows that there are a number of unique stations not listed anywhere on this forum that have picked up the software and are using it. Since we are free software we don’t want to require people to sign-up for a mailing list or even register that they are using the software. We have no telemetry which isn’t something I want to change. I am happy with how LibreTime respects the freedom of the users but it would also be useful to be able to reach out to the users who wanted to receive updates from us regarding development, funding and other things. I’m still not sure the best way to organize this data but that is my thoughts on the matter.
A paid beta, where 3.0 + API + ?, are goals could be one part of the equation. Based on feedback and other examples in the wild, I think there’s enough interest for a partner such as OpenProducer to test the waters with a WordPress MVP that integrates LibreTime.
…a couple folks involved in this discussion are planning some kind of SaaS offering for later this year. While not LibreTime owned, the service should generate funds for the project and contributors via Open Collective.
Does anyone know of other LibreTime service offerings - existing or are in the works?
