Messages

This is a good idea.  I am hoping to contribute.  Where should stuff that is found be posted?  Here in this thread?

Some of the items on the list are good ones, but fairly challenging to do without investing some time to determine usage of templates/extensions etc. 

rsync would certainly work.  That would be more of a mirror than an archive/backup I suppose.  When I have done it for our wiki it has been to just create a tar and then upload somewhere else.  This is what is described on the mediawiki site (http://www.mediawiki.org/wiki/Manual:Backing_up_a_wiki#File_system).  However, that can be very time consuming to do all at once, which is why I thought there might be progressive backup methods that use crons, etc. that some wikis might already have implemented.

An excellent write up Justin.  Can you also share your thoughts on backing up the file system?  Since this can be many, many GBs of data, I'm curious how often you do a tar of the files and then store them remotely somewhere.  Weekly? 

Bump.  I've been messing with jQuery some more using the updated version that comes with Mediawiki.  Here are a couple useful tips:

http://www.mediawiki.org/wiki/Manual:Collapsible_elements

As of Mediawiki 1.18, using jQuery for collapsing elements is now supported natively.  Just add the class 'mw-collapsible' and 'mw-collapsed' to a table or div and it just works.  This allowed me to get rid of 4 separate javascript functions in my Common.js.  The jQuery features a little animation when expanding/collapsing things.  You can customize it a little bit, but it mainly is for hiding navigation bars that are so widely used.  Here it is in action: http://dragon-quest.org/wiki/Category:Navigation_templates/Sandbox.

I also made use of this feature to redo our spoiler template, which is used to hide spoiler information.  Check it out in use here: http://dragon-quest.org/wiki/Dragon_Quest_V#Characters.

Finally, our wiki was dealing with a race condition with jQuery libraries being loaded for Common.js.  I am using jquery.ui.tabs and jquery.ui.accordion to format things.  Sometimes the libraries would get loaded and look fine, other times they would not get loaded in time and the pages looked bad.  I think I have fixed it now by forcing mediawiki to load them as dependencies.

jQuery( document ).ready( function( $ ) {
        mw.loader.using( 'jquery.ui.tabs', function() {createJQueryTabs();} );
        mw.loader.using( 'jquery.ui.accordion', function() {accordionVideos();} );
} );

So at the end of page loading, the resource loader ensures that each library is loaded before calling the functions which then go and create the tab/accordion formattings on the page.  I need to do some testing in different browsers still.

The only bad news I found is that the resource loader by default does not load any extra modules when using the MobileFrontend extension (aka mobile browser formatting).  That means that any special formattings won't happen and the pages will likely look bad.  Not sure if I will do anything to deal with that at this point or not.

Anyone have any tips for configuring the WikiEditor extension?  It was working pre-upgrade to 1.21.2, but is not now.  I seem to remember in the past that it has been very flaky.  The extension talk page seems to confirm that, as it is full of hacks and work arounds to get it working.

Found a nice blog post on creating dynamic CAPTCHAs using the ConfirmEdit extension.  I've implemented on our wiki and so far it is working great.

http://thingelstad.com/stopping-mediawiki-spam-with-dynamic-questy-captchas/

We have not had problems with users spamming edits on pages, but have had major issues with rogue account creation.  I like the random sequence of character solution because it doesn't require the user to navigate to a link or speak a specific language.

I also agree with Tucayo suggestions.  I don't have a problem with having redirects to member wikis in the case of many games (in case someone is looking for things on the wrong site), but otherwise game articles should only exist for games not covered somewhere else (Ice Climbers, Game&Watch). 

I'm not a fan of having lots of different color schemes on the same wiki, but that is not an issue if there are not articles for all the games.  That might be just a personal preference; I prefer a wiki to look consistent in terms of color scheme on all pages.

I didn't use git to retrieve the new version of Mediawiki.  I downloaded the zip and then uploaded it to the server.  I know using git could make it one step simpler, but either way works.

Also, I assume you are then manuallying updating your extensions in the newly installed version?  Your script could obviously just copy the extensions directory over, but that's probably not a best practice since some of the extensions may not be compatible with the new Mediawiki version. 

I was able to get the EmbedVideo extension working with 1.21.2, even though it doesn't state on the extension page that it is fully supported in this version.  Seems to be working ok.  I also installed the Mobile Frontend extension, which is what Wikipedia uses for a mobile browser skin.  It sounds like Mediawiki will eventually integrate a mobile skin directly, but for now the extension works pretty well.  I'll have check out the other NIWA wikis to see how folks are handling mobile visitors.

Thanks for the full list Moydow.  I didn't see this post until now, but earlier tonight I upgraded the Dragon Quest wiki to 1.21.2 from 1.18.0.  It went easier than I thought it might be.  Looking at your install process, I was did pretty much the same steps.  Good to know I am on the right track!

A couple notes I had:
1. This was the first time I unpacked the new version to a new directory instead of on top of the old.  It went really well. 
2. I had to update the path to my image temp directory in LocalSettings.  Thumbnails weren't being generated correctly at first.
3. I tried using the web installer, but it didn't go well, which wasn't unexpected.  Using SSH, I found that the default php version was non-compatible with Mediawiki 1.21.2.  Fortunately, hostgator has other versions installed on the server, just not as the default.  So I was able to run the upgrade manually by specifying the run location for php.
4. I took the opportunity to weed out some old extensions we don't need and to make sure I had the current versions for the ones we are using.  Looks like the EmbedVideo extension doesn't work with 1.21.2, so I'll have to find something else to embed videos with.  I got that one originally from seeing it used on Metroid Wiki...which is still using it with 1.21.2.  I'll have to go see if it still works there or not.  The extension page says it is only compatible up to 1.20.something.
5. My .htaccess is fairly complex because the Dragon Quest Wiki is the combination of several previous forms which had different URLs.  I have rewrite rules for them all to make sure legacy links still work.  Eventually I'll probably take those rules out. 

Well, I guessed it was not an original idea.   It makes a lot of sense.

So, by that accounting that leaves the following wikis as independent:

Bulbapedia
Dragon Quest
Icaruspedia
Nookipedia
Starfy
Wars
Zelda

I would assume Bulbapedia and probably Zelda would stay where they are due to volume of traffic.  Some of the others could definitely be considered for sharing a Linode or similar situation.  Right now, I pay for all the fees related to the Dragon Quest wiki.  The hosting isn't much (about $5 a month), but it doesn't make sense for me to pay $20 a month for a 1GB Linode that the wiki would only need 1/3 of.  Splitting that cost 3 ways to be about $7 per month would be great however.

I really don't know how this would happen, but I'm open to the possibilities if others are interested and have some knowledge of how it might be done.

Bump.

So, as of today Dragon Quest Wiki is still hosted by hostgator.  A month or two ago they migrated to new hardware.  I really haven't noticed performance issues as much as 2 years ago when I originally created this thread.  However, doing maintenance and other stuff is still a pain because of access restrictions. 

I assume there are still other NIWA wikis running on shared servers?  I wondered if there might be interest in getting a Linode that maybe a few of the smaller/less active wikis could share together.  Seems like this would make things easier with regards to making sure all the wikis are up, the upgrade process, shared costs, etc.  As NIWA moves forward, it seems like it would make sense to have some of the smaller wikis band together.

How many wiki could be supported via a 1GB Linode?

Just did a quick check and I'm pleasantly surprised to see that at least half of the NIWA wikis are using the most recent version of MediaWiki (1.21.2).  The other half range from being a minor version behind to several version behind (WarsWiki is 1.16.0). 

Upgrading is something I have come to dread, especially when there are database schema updates.  Everything is fine if there are no errors, but there always seem to be some errors and fixing them always seems painful to me.  Another annoyance is that due to being on cheaper webhosting, the Dragon Quest Wiki host kills any process that runs on the shared server for a lengthy amount of time.  This can kill the upgrade process, or at least make it fairly difficult.

I would love to see info like this posted on this forum.  There have been hardly any posts in here.  Sharing details of how an upgrade went, what went wrong, etc. would be really useful to compare with others.  Also, sharing information on new extensions you install and why, extensions you've removed and why, etc would be really useful.  Across NIWA, we are using a whole lot of variety when it comes to how each wiki is installed and configured.  I know I could benefit a lot from seeing what others are doing.

I should also mention that Square Enix dumped the DQVC online service for Dragon Quest IX last month without any sort of advanced warning.  It just quit working, with many players thinking it must be down for a day or two (which has happened in the past).  Nope.  Square Enix just pulled the plug. 

Too early to say for the Dragon Quest 7 remake.  If it did happen, it likely will be a year or longer until it gets outside of Japan.  Due to the Dragon Quest X debacle, they will likely wait until they are absolutely sure it will be a success.

As for the Pokemon/Mario comparison, they have both had many more titles released than Dragon Quest and both are first party.  I probably shouldn't have said "best selling" but instead "one of the most popular" series.  Dragon Quest in Japan is pretty crazy.  http://en.wikipedia.org/wiki/Dragon_Quest#Cultural_impact

No Dragon Quest news came out of E3 from either Square Enix or Nintendo.  Underscored by the poor sales of Dragon Quest X on the Wii U, Dragon Quest has officially been abandoned again in North America.  Square Enix simply cannot get their act together and made a horrible decision in making Dragon Quest X an online-only, subscription based game. 

Dragon Quest was dead in North America from 1992-1999.  Looks like we are entering the second "dark age" for the game series which is arguably the best selling series of all time in Japan.

Wii Wiki does not meet the requirements to join NIWA.

http://www.niwanetwork.org/join.php

However, so long as you give credit where credit is due, there is nothing wrong with using content.

Yeah, the Wars wiki has/had more of an uphill battle than the Nintendo wiki really.  Since it was not a well established long-term series, the overall interest declines greatly for both editors and visitors when there are no new titles released.  The 2 are very connected.  A long-running series will most likely have more titles, and the fact that it already has many existing titles only boosts its hits from search engines, etc.

Nintendo wiki is underdeveloped, but has potential.  Wars wiki is facing more dire circumstances. 

There is a lot of loose terminology being tossed around here.  Wikis are not "dead" simply because no one is editing them.  Some articles might be getting stale and new articles might be missing but those factors do not mean a wiki is dead.  A wiki is only truly dead when it is not visited by anyone on the web or it is taken offline.

Here is the Wikipedia page for dormant articles.  Many of these articles have not been edited for over 7 years.  Are these articles dead?  No, they just might be in need of updates.  I'm sure lots of them have been used millions of times in those 7 years by those searching the web.

With that said, I still think the Nintendo Wiki has a strong reason for existence.  There is plenty of information about Nintendo which is not series related.

Nintendo and its corporate history
Important Nintendo executives, Shigeru Miyamoto, lots of others
Nintendo hardware and peripherals, where else do articles about the Power Glove, Super Scope, and Virtual Boy go?
Development studios which Nintendo controls and their history
Games which do not merit their own standalone wiki (Wii Sports/Play/Party/Music, Game and Watch, other Nintendo ventures which were Japan only)
The Nintendo Store in New York
Channels on the Wii and Wii U
Nintendo Power magazine history info, etc
Nintendo @ E3 (I think it would be awesome to have articles detailing what Nintendo unveiled each year, reaction, etc)

Nintendo is identified by the products it makes, which is why the Nintendo wiki is really a "support" wiki to the game series wikis.  I don't believe it is dead or useless though.

   

What exactly would be the benefit?  Being a member means you have the support and are included in the hyperlinks on the other wikis.  I see nothing worthwhile about putting DQ and StrategyWiki in a separate list of links.  You are either a member or you aren't.

I can check in for Dragon Quest Wiki. 

We are mostly running in hibernation mode right now.  Since there are no titles planned for release in the USA anytime soon, and the prospects of Dragon Quest X online making it out of Japan are slim, there is little interest in the wiki at the moment. 

We also were having major issues with spam account creation.  I am glad to see some advice posted above.  I will have to try out that extension when I get the chance.

So, overall not a lot to report.  SEIWA also is in somewhat of a hibernation mode.  We contacted just about every eligible wiki that might someday join, but virtually every one of them is understaffed, stagnant, or otherwise.