So what kind of compilation packs are allowed to be added to a game group? One of my favorite features of MobyGames used to be the game series groups so I could find the best compilation packs for games in a certain game series but now compilations are often rejected. I think this is a bad idea. If it really is so "messy" why not just add an option to filter out compilation packs when viewing game groups?

Compilations are only allowed into series groups. For example, a compilation with The Secret of Monkey Island can go into the Monkey Island group, but not the Sea Pirates group.

I forgot to mention I was talking about game series groups.

Can you give a specific example of something that got rejected?

These entries were rejected for being compilations:

Worms series - Civilization/Worms Classic Dual Pack

Ultima universe - Wing Commander & Ultima VI: The False Prophet, Wing Commander II & Ultima Underworld, Ultima Value Pak

Day of Defeat series - Source Multiplayer Pack

Driver series - Action Pack - Driver: Parallel Lines + Far Cry

Far Cry series - Action Pack - Driver: Parallel Lines + Far Cry

Ridge Racer series - Tekken / Ridge Racer

Sonic the Hedgehog universe - SEGA Fun Pack: Sonic and the Secret Rings / Super Monkey Ball: Banana Blitz

Super Monkey Ball series - SEGA Fun Pack: Sonic and the Secret Rings / Super Monkey Ball: Banana Blitz

Mostly 2-in-1 packs.

All of those rejections were wrong. Well, maybe except the universe group. The universe groups usually have a number of associated subgroups ("Might & Magic Universe" --> "Might & Magic series" and "Heroes of Might & Magic") and if a compilation can be entered into one of those sub-groups it serves little purpose to add them to the universe group, I think.

Another point of discussion was about fangames. In my opinion they clearly need to go to the series game groups so they can be found.

That's the first time I hear that compilations of games belonging to different series can go into individual groups for those series. I clearly remember the time when people submitted corrections to remove them.

The policy exists since about two years. Before, compilations could be entered into all game groups. That's where the corrections were coming from.

Oh, and I think this calls for my annual whining about the topic: The way we handle compilations sucks.

I don't see any sense in that. The whole idea of removing compilations was to clear up groups. A compilation of Secret of Monkey Island and something that is not Monkey Island belongs no more in the Monkey Island series group than it does in Sea Pirates group.

Compilations need to go into the series game groups because we have no other way to track a compilation with a specific game in it.

We also can't track most non-series games included in some compilation, or more importantly: games with ports exclusive-to-a-compilation (main game entry doesn't mention it was ported to other platforms, since they didn't have stand-alone releases).

I'd also reject shovelware compilations (all those "Civilization+Doom+Hentai Porn") from any series(franchise) group. Series can only get series-specific compilations (e.g.: "Baldur's Gate Saga", "Jedi Knight + Mysteries of the Sith").

As mentioned above, other groups (themes, settings, vehicles, periods) apply only to basic game entries. If I can fly Bf109 in game A, Fw190 in game B, then compilation A+B doesn't get Bf109 nor Fw190 groups.

I would have rejected those too.

Well, this was your policy in the first place to include them in series groups. However, it does not matter if we finally go ahead with a good solution.

What about the fangames problem? Personally I think series groups should include both official and unofficial games, except if there are enough unofficial games to warrant their own game group (like the unofficial Flight Simulator add-ons group). Star Trek could be such a candidate, I imagine.

That's trickier, but I guess so.

A1: There is the problem of overlapping, including compilations that include hundreds of games in their repository. There was a reverse approach proposed, however it seemed to lack support and was eventually abandoned. This unfortunately however is possibly what you were looking for: Compilations: Beneath a Steel Sky. Er. I think.

A2: Any approach that requires a fix on the programming side is impossible to implement here. Don't ask why. :p

Might as well repost this summary I put forward to discussion a few months ago:

I am fed up with the inadequate way we handle compilations once again, so I need to whine a bit about it. Please bear with me.

The problem:
- The way we handle compilations is awful. We have no way to track in which compilation a game was released. Honestly, without a link between game and compilation (and not the other way around like now), a compilation entry is pretty uninteresting and of no practical use.

The current solution:
We add compilations to series game groups.

Advantages: - Not much work

Disadvantages: - Only covers a part of games - There is no way to know which game of the series was released in the compilation. This could be added in the game group description, but this has own disadvantages of its own: a lot of work, information in a place where no one looks, needs to be standardized.

In the last years, multiple solutions to this problem were tried out, but none left the testing stage.

Solution 1: a game group for every compilation
Example 1, Example two. Solution contributed by Xoleras.

Advantages: - Such a game group is quick to add
- References the specific game included in the compilation
- Since the compilation itself is also included in the game group, in most cases it can be guessed which version of a multi-platform game is included
- We can stop putting compilations into series game groups.

Disadvantages:
- When a game is included in many compilations, it can get messy and confusing on the game page. This can be countered by using the "Compilation: " prefix in the game group name.
- The game group count will be considerably bloated.
- It is a workaround for something that should be implemented in the site design itself.

Solution 2: One compilation group per game
Example. Solution contributed by Indra.

The advantages and disadvantages are pretty much the same as in solution 1. The main difference is that the game rap sheet will be clearer and the messy part will be on the compilation rap sheet. It would be more helpful for people who want to find a compilation with the game at hand, I think. However, it is certainly more work to create the groups, at least at first.

Solution 3: Trivia
Example. Solution contributed by Demian Katz.

Advantages:
- Since trivia is free text, we can specifically list which platform or version of the game is included in the compilation
- Can be conducted by non-approver contributors.

Disadvantages:
- Since trivia is free text and coding is manual, it is the most work intensive solution, both on the contributor and approver side.
- The information is in a non-intuitive place. Compilation information has not much to do with trivia.
- It is a workaround for something that should be implemented in the site design itself.

Solution 4: Wait for the re-design
A site re-design is in the works and maybe makes all workarounds unnecessary because of a smart compilation concept.
Solution contributed by myself during the last discussion about this topic.

Advantages:
- No work for us. At least now, I don't think the already entered compilation entries can be converted into a new system without manual work.
- The information can be stored in an intuitive place and the links can be automatically created when submitting a compilation.

Disadvantages:
- Will never happen (a solution to the problem I mean; the re-design will probably come someday within the next ten years)

And as addition:

I just had another idea t counter another problem discussed in the main forums:

Solution 2.1 One game group per platform, e.g. Beneath a Steel Sky (Amiga) and Beneath a Steel Sky (DOS)

Advantages:
- Even better for people who search for a specific platform version of a game.
- It does not matter if compilations with different game selection are merged or split. The group will be correct anyway.
- Allows for factually correct linking of emulated games and cross-platform compilations with exclusive ports. I especially think of those PSX or Windows Arcade game collections. Also useful if the game in question has no MobyGames entry because the original platform is not supported yet and there are no stand-alone home ports or if the arcade version would get its own entry.

Disadvantages: - More groups
- More work
- Problematic for DOS/Windows/Windows 3.x
- Might get messy for games with a lot of ports (think Lemmings)
- Only really useful for the special cases mentioned above or games which are in a lot of compilations.

The official result: no one cares. (three votes for #2 (including myself), one vote for #4)

I still like #2 most and I agree with Unicorn that series of game groups should be kept pure and not filled with compilations that clutter it when they contain completely unrelated games. If Rob or no one else in charge has an opinion about it, we take the highest vote and go along with it, as pretty much every policy in the last five years has been approver-driven.

I agree on #2 for now, although I hope that we in the future can have a more elegant solution for showing relations between entries.

When we use Solution #2 I'm all for removing compilations from all game groups (unless the compilation itself belongs to a group, of course).

However, there are a few points we need to agree on. I wouldn't want us change our minds in the middle of the project.

• One-series compilations: I'd treat them the same as shovelware
• DLC: We established some rule of thumbs to decide if such a package is marked as compilation or not. Use them here, too.
• Emulated games (including emulations from platforms we don't support yet such as arcade): I would include them. Even make a compilation game group if it can't have its own game entry yet. It will be messy if we try to make hundreds of exceptions. In really confusing groups, we still have the game group description.
• Games only released in compilations: Now that's an interesting question. Personally I'd say the best solution would be to make no group but enter them as alt title for the compilation.

Hm, I just noticed a pack wasn't in the series group and submitted it. Guess I should pull it now?

Does seem pretty weird though, I think there should be a way to see all options for getting games from a series in one place, not going to each of them, then checking the compilation group they're in. It's definitely a... less bad choice than #1 and #3, but...
Maybe a compilation group for the series too? Compared to making one for each game, it shouldn't add all that much to the workload...

#2, definitely. Simple, elegant, non-cluttering.

See? You only show up once and suddenly we make progress! Now just imagine how far we can go if you pop up on a regular basis again!

We'd all be discussing tentacle porn!

My vote is for tentacle porn. Oleg being molested is optional. Eww.

Regardless, what Sciere said. It's been the approvers (and concerned contributors) that have been pushing the wheel. Make it formal, contribute a poll (or preferably a thread for votes) for a final public consensus. Whatever result, stick with it.

*goes back whining to Facebook and ignores the grammar errors in 2,432 game group descriptions.*

We had a poll last week and the majority of voters (65%) were in favor of the idea. So I think we should just go ahead and do it. My suggestion: from now on we reject all compilations in game group submissions (except when the compilation itself belongs into the group and compilation exclusive games). In turn, the rejecting approver opens up a compilation group for the rejected game. Basic research for the included compilations can be conducted over the site's full text search.

This way we don't have to rely on OCD alone to get it rolling.

I'm sure this could be in good part automated, scanning compilation entries, checking the game links included in description, creating groups for those that exist in database if said groups don't already exist, and adding the compilations there. Of course, some may still slip by (bad links and such) and an approver would still have to have a look to weed out false positives (game links included as an example of similarities, related projects/compilations, etc.) but would take care of most of the work at least.

But... that'd require a backend programmer...

Let's keep it realistic here. This won't happen and we all know it.

I know that the proposed solution is only the lesser evil, yet I still voted against. We have over 3000 compilation entries, does that mean 3000 new groups? If there are differences between compilation ports (say, Amiga gets game A, B, C and D while Atari ST gets A, B, C and E) that means even more groups...?

@Patrick Bregger: isn't there some (video)gaming institute in Germany? If only they could acquire MobyGames and hire us all, I'd be gladly & meticulously going through those 3000 entries :) Too bad ICHEG.org wasn't interested...

Otherwise the new proposal borders on insanity. It will slow down database growth as approvers & veteran-contributors will be up their necks with sorting out those new groups. The benefits are little (do people hunt down compilations? I guess standalone releases are what constitutes as collectible item). If the site will get an update, this group solution will probably be made obsolete by a new database structure.

I'm all for removing "10 bestest games evar" compilations from game-series.

One game, one group; just like the example. I am pretty confident that people are able to read up the game entries to find the differences - if they are able to find them. Which is pretty much impossible now.


Sorry, but this is the kind of thinking which causes MobyGames to stagnate: Never change anything. Vedder's new genre proposition? Nope, too much work. Remove genres from compilations? Nope, too much work. Replacing bad and/or plagiated descriptions? Nope, too much work. Just because a change in policy may result in work does not mean we should stop improving.


May I laugh?

I really only need a "yes" from Corn Popper to overhaul the genres, admins have access to it. It's just a matter of inserting the new ones and removing the obsolete genres from all games that still use those, before removing the genres the new system does away with entirely. Just three letters. I'd settle for a "ya" or a "yeah" too. The initial hurdle was that genres should be edited through a revision system, but we all know that's never coming.

Do not put words I haven't said into my mouth. It's not up to me to make changes to MG. Many times I proposed enhancements to database structure and site's interface, just dig through my posts.



It's just that the compilations->groups solution is a makeshift one, and a bit sloppy too.

We've trying to slow down Sciere's contributions for years now. That isn't happening anytime soon. Not when there's crazy Belgians and Germans in 24/7 bot mode. :p

I blame Alkaline batteries.

Hey, wait a second. The question in the poll was "Should we make game groups for a game and it's compilations". There was never mentioned a "and kick out all compilations from other games groups". Using this approach you can add a "Company of Heroes: Antology" to the "Company of Heroes"-game group but not to the "Historical conflict: World War II"-game group.

That's been put into practice for years already. The only reason it hasn't been applied everywhere is the tiresome process to get them removed.

Wait, we cannot add a compilation with all games set in WW II into the WW II group?

Not if the included games are already part of the group. The exception is game groups for series of games, like a Command & Conquer compilation in the series game group for instance.