Second wiki roadmap

[Garrett]

OK, now we have a secret forum, so let's get brainstorming!

Here's a rough list of what's needed to make this second wiki a reality. Feel free to add or correct anything listed.

Scope
What will be included, besides characters? Real-world collectibles? In-game items/weapons/vehicles/etc.? Films? This will all have an effect on what name would be suitable. While the scope can change over time once the second wiki is running, having a good idea of what will be included will make the initial growing stage much easier.

Software

• OpenID support. The existing extension has minor problems installing on MW 1.13+, and it will need some modification to work with GamerDNA's OpenID server. Ideally, this would be implemented right from the start to avoid any potential problems.

• Semantic MediaWiki. Not a necessity, but it allows for dynamically generated lists to be displayed. I don't know the full extent of what it can do but it seems to have many features comparable to DPL, and Skizzerz says it's not as hard on the server.

• Account sharing. The easiest way to implement this without security issues would be to set up a wiki family and handle accounts centrally, using subdomains for the sites (e.g. en.strategywiki.org and gamepedia.strategywiki.org). The Wikimedia Foundation has worked around this limitation by remotely triggering images that write to the other sites' cookies (or something to that effect), but I don't know how much work that is to implement. Regardless of the implementation, the second wiki will still have its own domain(s) even if they only act as redirects.

• Commons-style media repository. Not only would this allow the second wiki to access the huge archive of existing files, this only has to be done once to allow access from any and all potential future projects (e.g. a Japanese StrategyWiki). This feature works just like on Commons, including the message about the upload being on another site (see this Bulbapedia image for an example). Local media uploading can optionally be disabled, which would probably be best to ensure all images are available to all sites. I don't know about the work needed to move existing images to a repository, but the long-term benefits should far outweigh the effort required.

Anyway, I can't think of anything else at the moment (the proposed change of forum software is a matter that needs its own thread), so there you have it.

[Skizzerz]

For the software, would it be a good idea to have global rights (aka global sysops and the like that can manage all wikis)?

[prod]

Stewards would be good, but that other stuff sounds a lot more complicated now.

[Dukeruckley]

For the scope, do we want to focus on in-universe information or not? Personally, I'd say yes because Wikipedia already provides the out-of-universe information very effectively (not saying we can't include it, but we need something to set us apart). The problem with in-universe is that now you've opened a can of worms to a ton of information. One game could be a full wiki on its own, nevermind the thousands of games out there.

Once again, to help mollify this, we could use subpaging. For instance, Chrono Trigger could have its own page with subpages like CT/Items. Items could branch off further (should we determine each individual item deserves its own page). This would be great for games with a ton of in-universe information (say Star Ocean with its myriad of characters, places, etc.).

However, on the flip side we have the idea of more out-of-universe approach. A Mario page would be great but not really possible with the subpaging idea.

I'm sorta torn on this, the more I think about it.

[DrBob]

I actually think the new wiki would be better off without subpaging. It's got such a broad scope, and all sorts of random topics, that I think it would be quite difficult to consistently subpage things sensibly.
Instead, I think sensible use of top-level pages, with bracket-based differentiation where necessary, would work fine.

I'm fine with the scope as it's currently defined (i.e. not too strictly). I don't think we can define it much more solidly without actually having the wiki and seeing what happens.

On the software side of things, I'd quite like to give Semantic MediaWiki a go, since otherwise we'll drown under the weight of categories.
I think we should set up the commons-style media repository first, since otherwise we'll have a real job moving across all the images which will be initially uploaded to the new wiki.

[najzere]

Will we move all non-strategy guide content to the new wiki, such as company and engine pages? Will the wikis be intertwined enough to share category tags? If not, maybe there should just be a link to the applicable VGopedia page. If they can share categories, it would be neat to have the date categories include stuff from both, although then I don't know which wiki it should reside on.

Also: is there a mockup or anything of what the new wiki would look like? Has a color scheme been selected or a logo made? I'm interested to see how it's coming along.