Recently, we’ve merged PR#4412 (repo, app page), which clearly labels itself as a small university student project.
I have mixed feelings about this app. While I’m glad we can add small apps / projects to Flathub, if we merge all students small projects, we’ll end up with a ton of unused and unmaintained apps, and that will dilute the value of Flathub.
The nice thing with Flatpak is that anyone can setup their own repo and it does not need to be on Flathub to exists, so maybe we should redirect people to that for such explicitly small and student project apps.
I also understand that the difference can be very fuzzy between a small student project and legitimate small application.
A nice thing about Flathub is that it allows tons of apps to get out there, and get shared. If anything, we should be a bit stricter on what’s required to get on Flathub, and provide suitable resources to allow others to host their own repository if they get rejected.
A definite checkbox should be “is the project actively maintained, and does it have a good likelyhood to be continue to be maintained in the future?”. If that can’t be met, it should definitely be scrutinized a lot more. This way, student projects can still get on Flathub, but they can’t just be one-offs; they must be updated appropriately as time goes on.
I also had a look at just serving a flatpakrepo using github pages, and it seems to work. Not sure how happy github would be about it though, at least from their docs it doesnt seem like they care what you serve, so long as its not commercial.
You could even generate a small website as part of the deployment process, and make a little static flatpak store.
Game preservation gets to be a bit of a problem if nobody maintains the wrapper for it - the criteria here would just be “Is someone available to make sure the game still works, and update the runtime when one drops?”
That is up for discussion. I’m sure everyone would be fine with discussing it at some point in the future; @hub definitely has some thoughts on it.
@TheEvilSkeleton’s Flatpak remote repo was exactly what I was thinking of, though it would be nice if GitHub worked with Flathub to properly provide native ostree repositories on GitHub infra, or OCI remotes that work better for use with Flatpak.
While that would be ideal, it shouldn’t break if that’s not the case? The worst thing would probably be the download size, as it’s unlikely that you have a runtime that old.
My opinion would be to be accepting of such projects, but simultaneously ensure we’re being clear throughout the stack when something is outdated. That means:
Flatpak CLI
Flathub website
App store front-ends like GNOME Software
The Flatpak CLI already warns when an app uses an EOL runtime, for example. elementary AppCenter also flags apps using an outdated runtime as well. So in effect, Flathub could accept these apps and then if they become outdated over time, they will naturally fall out of favor in all of the actual user-facing places.
There’s nothing ensuring any app on Flathub won’t suddenly become unmaintained, so this all needs to be considered anyway to ensure we’re not encouraging people to use outdated apps. But I don’t think gatekeeping people from getting their apps onto Flathub because we don’t know if they will maintain it long-term is a great policy. Let’s onboard them into the community and encourage them to stick around, instead.
Hello. Tabela is my app. And I didn’t think it would start a discussion here
I don’t know if I should comment if it’s my application that is the starting point of the whole discussion, but I thought it was worth leaving my post.
The `table’ application is part of a master’s thesis in which the topic of discussion is a comparison of distribution methods for open source applications on Linux systems. Why did I indicate right at the beginning of the application description that this is a university-related project? Firstly, I want to explain why the application is so very simple, and secondly to encourage other users to become curious about the history of the application’s creation. I must point out, however, that if I had not written at the beginning of the description that this is an app created for university purposes, such a discussion would not have arisen, and the app would probably have been added to Flathub anyway.
My opinion may not be very objective in this discussion, but I think that not adding apps just because they were created by studnets is difficult to verify, and also difficult to be fair. Well, a student can create an interesting app that they will develop for a long time, and a non-student can just add an app and never develop it further. I’m not saying my app will be developed for the next 10 years or more, but just because somebody create something during studies doesn’t mean it will die in a moment. Many interesting projects are created by young people at university or still in high school.
Furthermore, this is an element that unnecessarily divides potential developers. Let’s be real, Linux is not as popular as Windows or Android, and developing apps for it is not easy - mainly due to lack of tutorials and other tips on the web. Closing down a popular repository like Flathub for young people (students) is simply not very fair.
My application may not be developed further for a long time, but the lack of further development will not necessarily be strictly because it was developed during my studies. In addition, there are many other apps on Flathub that do not mention studies in their description and have been underdeveloped for a long time.
Exactly! Lack of development can affect any application, not just student projects. And getting students and other enthusiasts interested in Linux and programming for it shouldn’t be fraught with limitations.
I am sorry, but I feel slightly attacked by this post. My app has been selected from a big collection of apps as a source of mixed feelings, pointed as an example of something bad. The purpose of my app was not to abuse Flathub’s resources, but to promote Linux and Flatpak through my master’s thesis. Publishing the app was hard for me, but I understand that because Flathub has high standards. I’m glad I gained this experience, which should help me a lot in the future. Unfortunately, this post demotivated me a lot and I feel disappointed.
Is this an actual problem? Are there hundreds of useless applications cluttering up Flathub? Or is it just a possibility, something that might happen in the future?
When I first created an application I submitted it to another {nameless} packaging system. The so-called {helper} system did not exist in practice. Errors that testing revealed got curt and patronising replies. I saw a comment (not aimed at my app) about peoples’ “Vanity programs”. The “Vanity” was on the part of the commenter. I got the impression that the aim of those supporting the packaging system was to boost their own egos, rather than enable the publishing of applications.
I switched to Flatpak and found helpful experts whose aim was to get applications published. Please don’t destroy this by telling people that their app is inadequate/second-rate. If there really is a problem, it would be possible to set up a “Study” category that people could choose to put their own application into if it has been created as part of a learning process, doesn’t have any real function or is unlikely to be maintained.
I think unmaintained apps are different and separate issue than this.
This one is more about demo apps aka hello world apps which aren’t really meant for general public usage but are made for some private purpose.
This isn’t much problem for flathub yet but anyone can see how it looks in snap store where there are hundreds of such apps, effectively cluttering it.
It’s possible that it won’t be problem as long as flathub does manual approvals since it takes too much time for most people to bother and introducing some hard rules may present flathub as unfriendly place for beginners.
