> I can only assume that’s a testament to the quality of mediawiki.

I was curious about this so I poked around both and I think I disagree. Both load very fast for me and are snappy and look pretty nice. The one difference is that the Runescape wiki has a single ad in the sidebar or at the bottom, below the content footer. While the Fandom wikis have 3+ ads, far larger, one of which covers content until interacted with (like being closed). For me, Fandom's ad approach absolutely falls within "offensively bad," while the Runescape ad approach reminds me of early 2000s, "here's an ad to pay the bills. We've tried to keep it well out of your way."

So I'd opine that it has less to do with the quality of mediawiki, and more about how much money both Wiki hosts are seeking to gain from the existence of these resources.

Try editing anything on a Fandom wiki and that's where the real differences in experience comes from.

Fandom makes it extremely difficult (nigh impossible) to do something as simple as access the page of an image asset.

Woof. Yeah, you're right. That was not great.

Fandom runs on media wiki too.

I have seen cookmeplox, one of the admins of the Runescape wiki, round these parts. Thank you for your work, as a gamer and new Runescape addict. For an MMORPG as massive as OSRS, having a good wiki is crucial and probably the reason why it's seen a resurgence over the past few years.

That's me! I also wrote the blog :)

Factorio and Rimworld have amazing wikis as well. And they're both maintained by the developers AFAIK...

The Dwarf Fortress wiki https://dwarffortresswiki.org is perhaps the most impressive I've seen, as it maintains namespaces to maintain (and update!) information about particular versions, because many players end up staying on a version for various reasons.

I wish the Minecraft wiki did that. I don't tend to play the latest version because I feel like it got overly complex and I get analysis paralysis if I play the latest version.

But being on an old version makes navigating the wiki hard. I'm never sure if some content applies to me. Sometimes they say which version a feature was introduced in, but if a mechanic changes, they often just document the latest behavior.

Tell me about it; playing GregTech:New Horizons and trying to figure out vanilla mechanics related to 1.7.10 is annoying. All the GTNH specific stuff is on their wiki, but vanilla mechanics are just assumed.

The good ol' Mediawiki look of the DF wiki reminds me of the underrated, and oft maligned Dungeon Crawl Stone Soup wiki: http://crawl.chaosforge.org/Crawl_Wiki

I haven't played DCSS regularly since probably v0.24 or v0.25, so things may have changed - but if I recall correctly, it was not kept up to date very well, character guides are flat-out wrong, etc...

One of the Crawl design philosophies is that it should be possible to play without needing to consult a wiki. E.g. inspecting a monster shows you its spells and their damage ranges, there's a searchable in-game encyclopedia of all items/spells/monsters/etc., there's an extensive in-game manual with things like species skill aptitudes, examining an item tells you exactly what skill level you need to use it optimally, and so on. There's plenty of useful stuff on the wiki, but it's not a priority to update because it's not entirely necessary.

Factorio has a similar goal, which actually works really well. It lets the wiki focus on strategies and things that aren't well described in-game, but for quick "how does this work" you can just stay in game.

I mean yeah, I agree, I'm a longtime player. I appreciate DCSS's in-game discoverability. I find wikis useful for more in-depth explanation of game mechanics though. How combat rolls are calculated, etc. Ideally a wiki would provide that kind of deeper level of information, guide materials, etc.

Interestingly, DCSS's best source of info is the IRC bots/learndb.

Could you say a bit more about that? Normally I think "have to support people stuck on old versions" is something that happens when you're selling enterprise software to insurance companies. This is the first I've heard of it in games.

DF had a massive update probably a decade or more ago that changed the game from 2D to 3D (still represented as 2D z-levels though). With such a change, obviously some people would want to stick with the old version. There have been numerous large updates since then (the game has been in development for 22 years) and with each you get some people that just don’t want to update, either because it might ruin their current games or they prefer to avoid new features etc.

Another example is the various Dungeon and Dragons wikis that allow you to toggle between versions, since it has existed for 50 years now.

As mentioned, some versions of the game introduce breaking concepts that earlier players may not want to deal with (either because it breaks save compatibility, or they don't like the mechanic, etc).

Minecraft has this somewhat also, with some people sticking on various versions because of mods, or play style, or combat, etc.

For example, one huge change was going from a 2D map to a 3D one, another was how world generation was done.

See "Eras" here for the big ones: https://dwarffortresswiki.org/index.php/Release_information

This actually used to be the norm: you'd release slowly, and support old versions for years (which still isn't that long, all things considered). It wasn't until relatively recently that six months became some sort of unconscionable amount of time to support software, because it's friendlier to the companies and developers writing it, instead of the users using it.

It used to be the norm because releasing and installing stuff was hard and expensive.

I agree we should be focused on the users, but I think the solution there is not to leave them on various outdated software, but to make it so easy for them to be on the new thing that they are happy with frequent updates.

And I feel pretty strongly about this because I've met people whose entire lives are about keeping old, broken stuff limping along so that pathological bureaucracies can never get their acts together. Sure, it's a living, but it's also a colossal waste of human potential.

Players comfortable with the ASCII graphics version (the one that existed for years) often just paid for Steam release with pretty graphics just to support the brothers. And then kept playing the "hardcore" version they are used to.

On the Steam platform for instance there is an option (perhaps developer supported) to stay on a certain version of a game. For instance, in the game Mount and Blade: Bannerlord, players notoriously stay 2, 3, or even 10 versions behind in order to maintain compatibility with specific mods or sets of mods (10s or 100s of mods). Eventually, enough of the modders move to the next or latest version and the players gradually move with them.

Games with "always on" or auto-updaters avoid this.

It's more a testament to the devs. I kept up with the RuneScape wiki Discord server for a bit and there were flamegraphs flying left and right. You can see some of there recent performance improvements here: https://meta.weirdgloop.org/w/Forum:Board_Meeting_-_2024-06-...

I think the theory is people edit more if pages load lightning fast. I can attest to that, especially if you use tools for partially-automated mass edits like https://github.com/wikimedia-gadgets/JWB