[Rand Al Thor]

Hey fekker,


Great stuff you posted there. Just thought I would comment on some of it and maybe add more.

Usability
Tie in to the xbmc database (if xbmc is closed, the database will be unlocked)

I think that filters, update media info, and auto pilot are essential. It looks like you already have that under control. I did want to comment on the DB tie in. I would love to see some sort of user management as well. This would be even more beneficial if we were to have a central db location as I mentioned above (yes, I am going to plug that again ). It would be great to be able to add/remove movies and shows on a per user basis. So you could set the default action to be "add new movies to all users" or "add to users X,Y,Z." Pretty self explanatory. Then, when editing a movie you could have check boxes with all the users names where you could manually add or remove a movie from a users library. It should work the same way with watched/unwatched status as well. That way you could have an easier way of separating and managing a "master" account and a "kids" or "guest" account.

Database
I think a database is good for some areas and not so much for others. I really like XML files myself. A mix of both would prob result in the best overall usability. Cached images/data and information about those images/data, XML.. specifics about just the media itself, database.

I think here you were referring to the database for the program itself, which I totally agree with you on. I would like to address the actual XBMC db again. As we all know, databases can be lost or corrupted. I think we should still write all media specific info to an .nfo file but ALSO manipulate the XBMC db if/when possible and even back it up. Better to be overly cautious in my opinion.

TV Show Specific
Since most tvdb data comes back from the api, it's easy to show what episodes are missing, when the next one comes out, and if the show is already canceled or is active.

Links to tvdb, the network website, and others is very handy

These are also really great suggestions. I totally agree. I would also like to add the ability to take custom screenshots for episodes. You can do this in MC which is pretty handy. It would be great if you had two levels of control for this.

1) Set a default timecode where it would snap an image. User could specify ex. 100 secs.

2) If the image taken at the default time is bad/blurry you could manually pick the image. Click "Manually Grab Image" and you get a video window where you can drag a slider frame by frame until you find a good one. Click "capture"

To add to that, in the spirit of the community it would be nice to be able to contribute images to thetvdb.com. In the settings menu you could enter your userid and pass. Then if you find an episode that is missing an image you can manually create when then click "contribute to TVDB" and it would upload the image. We would not want to have an option to automatically contribute missing episode thumbs because it would result in a lot of poor automatically generated thumbs. If a user is ok with have an auto thumb locally, fine. We should make it clear that only manually captured images, which should be of better quality and relevance, should be uploaded to thetvdb.

These are just a few things that popped up when I read through your suggestions.


@mafis90

There only one thing, that i have to add and that is seperating the core from the interface. So it runs great into the background and you can programming a create a webinterface, windows form interface, but all seperating from the core.
Somebody write something about this in this thread.

I couldn't agree more. That is what I was referring to in my last post. I think there is so much potential for this application to be more than just "YAAMM" (Yet Another Average Media Manager). It could be the center of all media instead of just a media center helper. Do you see what I mean?


Cheers,
Rand

[Gamester17]

I really like gamester's idea of basing it on mediatomb.

I only suggested MediaTomb for its UPnP MediaServer function, not for its media management feature, nor GUI.

So I am not suggesting to base any of the media management features on MediaTomb.

You have to understand that the media management bit have to come first, then the UPnP MediaServer part could maybe come later (or not).

Sorry for the misunderstanding.

[MDPauley]

You have to understand that the media management bit have to come first, then the UPnP MediaServer part could maybe come later (or not).

Gamester17,

Wouldn't adding upnp just be double work of what is already working in xbmc?

[Gamester17]

Wouldn't adding upnp just be double work of what is already working in xbmc?

True, but I am not suggesting it for XBMC but for other players such as the PS3, Xbox360, WMC, MediaPortal, MeediOS, Elisa, MythTV, Tvix, and Popcorn Hour which users then could then also enjoy (and contribute) to this media management software. I also said that UPnP should not be looked at right away but much layer, so maybe not even this year.

[Rand Al Thor]

@gamester

I understand that you were only talking about the upnp features of mediatomb. Your suggested just gave me an idea that there were other features of mediatomb that might also be beneficial. Such as the built in webui and streaming. I do agree though that these features should be second to the main features of media management

Cheers,
Rand

[seedzero]

If I can contribute in any way to this project let me know. My strong point is photoshop (logos, GUI mockups etc,). Has any thought been put into a name for the project? What about something like XMI (Cross Platform Media Info)? Or XMM (Cross Platform Media Manager)?

[mafis90]

@redtapemedia we can need help all the time :-D

@zirytowany I think we do the project in the following order
1. Collect ideas in this thread
2. Planing ( Milestones,...)
3. Planing :-D
4. Programming
5. Tests

I think this order is a good order and with a good planing i think there is no probleme.

Thanks
mafis90

[althekiller]

this sounds like an extreme undertaking. I have a feeling there needs to be some financial compensation for this. Are you all willing to donate to the cause for this? .

HAAAAH!

EDIT: Apparently I need more than just "HAAAAH!" or the forums tolower() my post. Caps are definitely needed in this case.

[Montellese]

If you guys are still looking for developers, count me in :-) I would love to help on this media manager. I already have some experience in this area because I used my self written Media Manager scraping OFDB/IMDB, ImpAwards, Google Images and using MediaInfo before I found out about XBMC and MIP ;-)

I'm gonna read this whole thread now to be up to date.

Just contact me over PM or something if you let me join in on this one.

EDIT: Btw I just read someone suggested Qt4. The self written program I mentioned was done with Qt4 and if need be I can give introductions and help to ppl who don't know much about Qt.

[royw]

@zirytowany I think we do the project in the following order
1. Collect ideas in this thread
2. Planing ( Milestones,...)
3. Planing :-D
4. Programming
5. Tests

I'd recommend swapping 4 & 5.

[mafis90]

@Montellese your a in the team :-D, fekker add you to the list on the evening. We plan to seperating the core and the interface, so we need someone to create the interface for tests :-D

@royw

[royw]

After mulling this over I thought I'd share my thoughts on ideal media management. This is my natural workflow after backing up 1000+ DVDs.

I typically rip 5-10 DVDs (including barcode scanning the ISBN into DVD Profiler) in the background over a few days before having time to "process" them. I'll usually start the processing then check mail or surf until it is done. I then review the report for problems (title not found in one or more DBs, multiple titles found, image not found, fan art). I look up the problems on-line then correct titles or add production year, etc. and then run the processing again. Rinse and repeat until clean.

Also the production and release years don't always agree between DBs. So a little fuzzy processing is in order. I've found matching against year, year-1, year+1 usually works better.

I've noticed that I disagree with the genres from the DBs occasionally so when I rip each DVD I place it in sub-directory that is it's best genre IMO. These genre directories can be nested, for example: "action/martial arts/" will tag the media with both "Action" and "Martial Arts" genres. This also allows me to group series as genres ("/Science Fiction/Star Trek" or "Action/Bond"). So the feature I'd like is an easy way to add genres.

While on the subject of genres, I've found a lack of standardization, for example: "SF", "SciFi", "Science Fiction" or "Musical", "Musicals". So a genre mapping feature helps a lot, for example, "SF" -> "Science Fiction".

Oh, also it helps a little to remove duplicate genres (yes, some DBs do have duplicate genres on a title).

While on the topic of lack of standardizations, movie titles can vary between DBs and the DVD cover. For my library, the title is selected from the first match in order: DVD Profiler, IMDB, themovieDB, filename. So a feature to select which title and/or a default selection order would be useful.

Another issue with titles is the original title. I like Hong Kong martial arts films, but want the English title, not the original title. So an option to use the AKA title that matches the filename would be nice. In my library I get around this issue by using the DVD Profiler database which uses the DVD's ISBN as the primary key and the title is required to match the DVD cover title.

Here's an issue I have with XBMC that a management feature might help. That is the number of actors, in my library the average is 35:1 actors:titles. A straight limit sometimes eliminates a bit part played by a popular actor. So I'd like to see something like: include the first N actors and any other actor that is already in the database. I know, this adds complexity as we would need one "full" database which will be culled to what we export to XBMC (either as .nfo files or directly to it's DB).

Food for thought.

[Montellese]

I'd recommend swapping 4 & 5.

Good luck with that one

One question: Will all the development planning be done here? Coz in my opinion it will get really messy. If need be I can set up a forum on my webspace where we can start discussing different issues in an organized structure.

[Montellese]

I've noticed that I disagree with the genres from the DBs occasionally so when I rip each DVD I place it in sub-directory that is it's best genre IMO. These genre directories can be nested, for example: "action/martial arts/" will tag the media with both "Action" and "Martial Arts" genres. This also allows me to group series as genres ("/Science Fiction/Star Trek" or "Action/Bond"). So the feature I'd like is an easy way to add genres.

While on the subject of genres, I've found a lack of standardization, for example: "SF", "SciFi", "Science Fiction" or "Musical", "Musicals". So a genre mapping feature helps a lot, for example, "SF" -> "Science Fiction".

Oh, also it helps a little to remove duplicate genres (yes, some DBs do have duplicate genres on a title).

I think Genres is a very difficult area. If you'd ask 10 ppl to define genres for 1 movie you would get 10 different results.

I myself would never ever think about organizing my movie collection in genre folders. How do I decide whether to put a movie into "Action/Science Fiction" or "Science Fiction/Action"? I would end up with a total mess and I'd never manage to find my movies on a manual search again. This is even more of an issue if you have movies in your collection you haven't seen yet. I got some movies I don't really even know what they are about so how do I know where to look for them in a genre folder system if I don't even know what they are about?

[Gamester17]

One question: Will all the development planning be done here? Coz in my opinion it will get really messy. If need be I can set up a forum on my webspace where we can start discussing different issues in an organized structure.

One option is that we (at least temporary) setup a separate internal private forum here for only the developers who volunteer to join this project, you can then also in the future use that internal private forum to communicate with Team-XBMC. Later you can setup a public forum on a other website once you are ready to release the first version and have decided on a name (so that you can get both the .org and .com domains for it).