Cardinal Forums

somebeaver

There is a new status app at status.cardinalapps.io

somebeaver

@kadytgv You are correct, if the email verification link is opened in a browser where you are not already logged in, then you will just be redirected to the login page without being returned to the email verification flow after logging in.

I have a ticket for this issue in the backlog now.

somebeaver

From now on I'm going to be publishing the work-in-progress to a new beta release on Docker Hub, which will receive more frequent updates than the stable release. I've also created a new beta forum for feedback on stuff in progress.

More info on release channels here: https://help.cardinalapps.io/guides/cardinal-home-server/release-channels

Beta release: https://hub.docker.com/r/cardinalapps/home-server/tags

somebeaver

I don't use watchtower myself but that seems strange. I can't think of any settings in Docker Hub that I might have changed that would cause that.

somebeaver

Yeah there's a frontend issue at the end of some indexing runs.

Version 0.2.0 will bring a new Jobs system and it should also resolve this issue. Thanks for reporting.

somebeaver

@douglasparker said in Good start:

Jellyfin developers refuse to take money. While I appreciate them wanting to offer something truly free, it really makes me feel like it will always be on a volunteer basis.

This can be be such a touchy subject. There is nobility in building something great and giving it away for free, but it comes with some really big trade-offs. People have bills to pay and family to feed, and skilled developers will have to choose between working for free, or working for high paying tech companies. It's difficult to attract talent on open source projects. I've seen Jellyfin's "A Call for Developers" post and it highlights the difficulty of the situation.

@douglasparker said in Good start:

I've also seen pull requests for requested features get closed because they want to do things themselves rather than work with the original dev to get it up to their standards. For example skip intro.

:/

@douglasparker said in Good start:

The biggest problem is none of them besides Plex offer Watch Together.

I think I have some really great ideas for how remote watching should work. I don't want to over promise anything here but I'm really excited about this.

@douglasparker said in Good start:

I think Cardinal Apps have a very good chance of doing extremely well in the market. Paid plans is important so Cardinal can be worked on full time.

Thanks. I'm looking forward to growing things organically.

somebeaver

@argyle said in Good start:

Looks like the app has a good foundation but functionality is fairly simplistic.

Thank you

@argyle said in Good start:

An all in one media manger is interesting & ambitious. I do wonder if a more focused scope would be more realistic.

Oh it's definitely ambitious. It was originally just going to be music, but I decided to include photos and cinema because all of the apps I use for these things seem to be getting worse, or are trying to lock me into financial subscriptions that I don't think are fair.

The way I see it, I've got another 50+ years of personal media management left in my lifetime. I can't even predict where we'll be at in 10 years but I know that I'll still be watching my old TV shows that will no longer available anywhere but on my own hard drive, so I'll need some good software to rely on. When I look at how other platforms are maturing, I feel compelled to build something that I can protect. The scope of work is necessary to prepare me for any future.

@argyle said in Good start:

We have well established self hosted alternatives for music and video. I don't know that cardinal will have much success competing in those areas. Photos on the other hand is intriguing. I don't feel there are any self hosted photo applications that are truly great.

I actually see it the other way around, and I think Cardinal will have a harder time competing in the photos space than music and movies/TV.

When it comes to photos, users expect mobile apps, photo editing, album sharing, object detection, 3d photos, drawing/annotations, filters, videos, photo bursts, and the list goes on. All things that are consuming because of the technical complexity. It's extremely difficult to compete against these features as a solo dev when Apple and Google have first dibs on everyone's photo library. Photos go way beyond self-hosting.

Music on the other hand has a smaller set of features and users are spread across mainly three platforms with many similar features: Spotify, Apple Music, and YouTube Music. I won't be able to offer Obama's Christmas playlist like Spotify can, but I think I can build a better desktop UI.

And movies/TV is even more consolidated than music. You get the same set of features across Netflix, Disney+, HBO Max, Apple TV, etc. I can build those features in a reasonable amount of time. The thing that sets movie/TV streaming services apart from each other is their content, something Cardinal doesn't have to worry about because users bring their own content.

@argyle said in Good start:

Speaking of the photos app. Indexing is slow, really slow. One of my smaller collections, (36000 photos) is going to take 48 hours to index. 20 minutes in and less than 350 images have been indexed. This seems to be much slower than other products.

This comes down to my development philosophy. When building a complex feature, I build the simple but unoptimized version first so that I can deliver value quicker. Once the basic version of a feature is stable, I begin iterating over it. Getting to use the apps with my own media keeps me motivated, and it allows for me to take user feedback into account when optimizing.

The initial photo indexing implementation is a synchronous operation. It's good enough to get photos into the Photos app so that I can build photos features and see things come together end-to-end, and maybe it'll be just good enough to interest new users.

But the optimizations are underway now. I'm currently working on a job system for the Home Server that will handle all of the long running tasks. The long part of photo indexing is creating the thumbnails, and if you enable the DEBUG_INDEXING environment variable, you'll see the exact amount of time it takes to create each thumbnail. It's like 98% of time spent indexing the photos.

Soon, thumbnail creation and HEIC->JPEG conversion will be offloaded to background jobs, so that the file queue doesn't get held up. I know that nobody is going to be happy if they have to wait a few days for their photos to index before the server even gets to their music, so optimizations are going to come quickly.

@argyle said in Good start:

A folder view is desperately needed. Anyone serious about photography will have their photos organized. Not having a way to browse images as they've been organized really makes the photos application unusable. The timeline is a nice feature but additional control is needed.

Absolutely. It's a core feature that I need for myself as well, and it will be built soon.

@argyle said in Good start:

I could go on and on with my wish list of features but I think attention to indexing & folder management would benefit the app most.

Happy to hear more suggestions any time.

somebeaver

Hi, I've sent you a private message.

somebeaver

Hey there, Cardinal won't modify your EXIF data or your files in any way. The errors you see happen when the indexing service runs into something unexpected when reading, not writing, EXIF data.

Are you seeing that error for every photo, or just a few?

I'm still working through the edge cases that can come up when indexing photos. For these photos, if you are willing to post one here or send me one in private, then I'll fix whatever bit of EXIF parsing is causing that error.

If not, no worries I have a few files of my own giving me issues and an upcoming update might fix it for you.

Cardinal Forums

administrators

Posts

Member List