“Attention assault” on Fandom
j3s.sh
j3s.sh
Fandom isn't the only bad wiki site though. Fextralife had (has? I haven't kept up) an issue where they were embedding Twitch streams on each page load, which was boosting viewer counts for whatever streamer they decided to embed on every page.
I'd love to see a world where more companies self-host the wiki for their game/TV show/etc, especially given the relatively low cost of deploying and hosting, but I also understand that most companies don't have the motivation to do that as it doesn't always directly impact their bottom line and it can take effort to moderate.
https://www.stardewvalleywiki.com/Stardew_Valley_Wiki
https://bulbapedia.bulbagarden.net/wiki/Main_Page
https://memory-alpha.fandom.com/wiki/Portal:Main
But you know what the weirdest, greatest one of all is? You'll never guess: it's the Transformers wiki: https://tfwiki.net/wiki/Main_Page Some absolute madlad has taken it upon themselves to seemingly caption every single image on the 30,000+ pages with a unique sarcastic quip.
My dream is somebody takes the data from Wowhead and ports it into MediaWiki and the community rallies behind keeping that in date, but I know it's a bit of a pipe dream.
I played a game where the developer did that. The wiki and all of its painstaking user contributions vanished the day the game reached its end of service and is lost to history forever.
Oh well, I don't understand the sunk cost phenomenon and do it for things besides money, but most people like being financially rewarded for these kinda herculean efforts. So we get what we deserve. And what google gives us.
I think the next best thing is when a company hosts their own forums where the community and in some cases the devs are active in answering questions. As long as forums are well-moderated then they naturally evolve into an knowledge base. X3:Terran Conflict comes to mind, along with IL-2 Sturmovik, which has good official forums and some excellent unofficial forums.
The most active users split from it and created https://minecraft.wiki/
"On September 24, 2023, after growing frustrations with the Fandom platform, the wiki completed its process of moving from Fandom and is now hosted independently at minecraft.wiki by Weird Gloop, with the old Fandom wiki now deprecated. The move also re-introduced a skin similar to the one used on Gamepedia, re-enabled anonymous editing, and introduced a new logo." https://minecraft.wiki/w/Minecraft_Wiki_(website)
> This is the official Path of Exile wiki, maintained by the community and hosted by Grinding Gear Games.
But for duckduckgo if you use the !poe bang... It takes you to the fandom page, even though the fandom page doesn't get updates anymore because there is an official wiki elsewhere.
As an example of out of dateness of fandom: https://pathofexile.fandom.com/wiki/Version_history claims the latest update to the game is version 3.17, from feb 2022... https://www.poewiki.net/wiki/Version_history says its version 3.24.0 from April 23, 2024. They don't even bother scraping the official wiki to update their own. It's actually dead.
Ive submitted to DDG a few times that the !poe bang points to an unofficial bad source when an official better source exists, but search engines don't care. The out of date fandom page is still the second result, because people will keep clicking it and it has years of accumulated SEO.
I saw few cases where a longstanding wiki project of a particular topic created by fans faced a competition in form of a fresh project created within wikia/fandom that was filled with low quality articles. There even were situations where content was blatantly copied over. Pretty sure that was done only to hijack position of the fanbase project so ads could be displayed on fandom wiki and user tracked.
Luckily there are projects which managed to avoid being sucked into this fandom blackhole, like evawiki for Evangelion franchise or both Guild Wars games wikipedias
---
And on as site note, I really don't like this recently introduced font change on general Wikipedia - previously default font is now the "small" size and current default is inconveniently the bigger "standard".
I'd rather that communities around games remain more independent of the developers to be honest. The incentives of the two are not always aligned and having at least two independent "authorities" helps keep each honest.
You can really see the difference with UESP vs Fandom for the Elder Scrolls
It also draws would-be contributors to the Fandom site, make changes there, and since they are no longer supervised by the community 'elite', the quality of the Fandom version rapidly declines.
EDIT: Hell, even 'Minecraft Wiki' still goes there.
The IP owners can request a takedown though & they'll usually comply, some game studios have done this
Their legal department is willing to bet that the IP owner's legal department does not care enough about the abandoned Fandom wiki to put the time and effort into finishing the job.
So do I; they're one in the same in my case, for a wiki that exited Wikia in 2010. The only thing Fandom even responds to are takedown requests for copyrighted visual artwork.
Sincerely, congrats on getting one through, but I think that's the exception and not the rule, and something you might get that standalone indie wiki admins and game publishers don't necessarily have.
1) Started in 2004 as independent "WoWWiki", then "moved" in 2007 to Wikia aka Fandom
2) Forked in 2010 due to Fandoms blatant disregard to Wowpedia, under Curse
3) Fandom acquires the Wiki part of Curse and with it Wowpedia in 2018
4) Forked in 2023 as Fandom made the site practically unusable
The biggest fuck you of it all? They continued hosting WoWWiki all this time for the sweet SEO, even if it hasn't been relevant for decades. Even when they started owning the new version in 2018, they would not disable the old one - you know, it was still ranking in that Google algorithm!
For all these years, there has been a single correct choice for what link to show a Google user searching for relevant content. And yet going on 20 years Google doesn't manage to do it, instead rewarding these utter dickheads.
I've personally been approached to build tech to monitor and revert changes matching keywords using residential proxy systems as a service for low-stakes clients (and declined).
Can you suggest some alternate approaches that you think would actually work? How would you approach removing the notability requirement without immediately opening the floodgates to hundreds of thousands of additional sketchy articles about niche topics that don't have enough interest to be vetted by more than one person per year?
These differing voting systems (stars, points, etc) could be normalized through some standardized process into a final reputation score. Other services could then either consider the aggregate score, or only focus on ratings from particular other services, which could be grouped into service types for convenience. Ratings are cryptographically backed and could be fuzzed to increase anonymity. Scores would also be weighted by the relative reputation of other entities. Participating services would also themselves need to maintain a high reputation for their scoring to have much weight in the aggregate.
Services could use these scores to wall of certain site features to entities without sufficient reputation, or even wholesale deny read/write access. Obviously there are particulars to work out to prevent brigading, etc. but there is a balance to strike such that farmed accounts could still quickly be punished the moment they begin spamming/astroturfing. Perhaps a flag/vouch system similar to HN could be employed. Voting rings could also be detected, though with limits given that each vote comes from an anonymized sub-key.
This system would even protect against a future where the majority of content on the web will likely be aided or entirely produced by LLMs or AI. I welcome such synthetic content, but only if it matches or surpasses the general quality of human posts on the internet (which is much worse than one might think). On top of this, users could themselves use these scores as a way to filter out content which is signed with a reputation score below some threshold.
I imagine some kind of hierarchical web, where some services are harder to gain write access to than others, allowing users to gain reputation in low-stakes environments with some level of decentralized moderation, but without compromising the integrity of higher-stakes information.
Each piece of information on the internet should be signed with some sort of introspectable reputation score, and this kind of system would make it possible without compromising privacy and security of users in any political environment.
That said, I think there is a distinction between fact and made up.
The short answer is, because it costs time and effort, and a little bit of money, to maintain all those articles, and they cause more problems than they are worth.
But I'd like if Wikipedia had 'subwikis' with that kind of stuff.
0. https://en.wikipedia.org/wiki/Wikipedia:Notability#Notabilit...
It would be weird for wikipedia to host, for example, every minecraft crafting recipe.
See also https://meta.wikimedia.org/wiki/Inclusionism https://meta.wikimedia.org/wiki/Deletionism
If they could be assured of never needing to reduce pages to not run out of money, I would have no logical objection to having a page for everything, but as someone who's donated quite a bit to them, they are most assuredly not in that position.
If the day comes that their hosting costs match or exceed their available funds, I'd much rather know that what is there is all the most important information, and not a glut of random pages that no one ever looks at.
Not to mention courses, articles, blog posts, discussion and youtube videos
The main difference would be history, the split between books and online resources, and how the older generation sees it.
It's probably easier for all parties to let the Minecraft community draw their own line on their own wiki(s) (which is exactly what happens).
I get the feeling Wikipedia will slowly become culturally irrelevant and end up in the same place as the encyclopedias, as an established and mostly frozen dataset that we'll want to keep running but won't look at 99% of the time
There would be some issues, e.g. I wouldn't want to see Minecraft recipes when I'm searching for "obsidian". But these seem solvable; niche articles could be flagged and downranked, or they could live in a satellite wiki project, or something else.
I remember seeing niche articles like this and enjoying it.
If, for instance, Wikipedia allowed articles on every human being, whether well-known or not, they'd likely have a larger problem than they already do with people creating attack articles against other people over extremely local disputes/feuds. With more notable people, they can look for other sources to confirm or deny information about those people.
Now what would the millions of articles about every TikTok influencer look like?
And then the same goes for products and companies. Every scam Kickstarter and onanistic startup would get a massive screed.
And then it all sits and rots forever once the academic retires, the influencer gets a real job, the Kickstarter vanishes and the startup folds, because no one else cares. But someone has to go around and fix the links and update templates and generally expend effort indefinitely.
On top of all that, while the article creator is still around, because the article is actually an advert, any attempt to edit it into a more encyclopaedic article is disproportionately likely to cause drama that burns up volunteer time and effort.
(1): One I saw recently https://en.wikipedia.org/wiki/Didier_Sornette
(2): Especially as many people actually good enough at collating and editing encyclopedic articles about subjects that aren't about their own media-based hobbies quickly decide Wikipedia isn't a very fun place to do that any more.
That cuts both ways. IMO Wikipedia has lost a lot of contributors by banning fun and disallowing the topics that people found interesting.
Not least, you can crosslink between Wikimedia sites, so you could just link to [[fans:Digimon:Whatevermon]] and have the content "nearby" in digital terms without drawing it under the same notability and sourcing guidelines as an article on benzene, say.
Which as you say would keep the (often very, perhaps to a fault) keen contributor to the fandom in the Wikimedia tent and might encourage them to contribute to Wikipedia and related sites as well.
Then again, the auxiliary Wikimedia sites are pretty neglected by the parent foundation which has more important things on its mind much of the time, mostly fundraising and finding novel ways to spend that money.
You mean something like Wikia?
Items can be referenced using their in-game names because there's no risk of conflict and custom templates can be created and tailored per game. Guess which one of these displays more info:
{{:Infobox:Tank}}
{{Infobox video game
| title = Factorio
| image = Factorio cover.png
| developer = Wube Software
| publisher = Wube Software
| platforms = {{cslist|[[Linux]]|[[macOS]]|[[Microsoft Windows|Windows]]|[[Nintendo Switch|Switch]]}}
| released = {{ubl|'''Linux''', '''macOS''', '''Windows'''|14 August 2020|'''Switch'''|28 October 2022}}
| genre = [[Construction and management simulation]], [[real-time strategy]]
| modes = [[Single-player]], [[multiplayer]]
| director = {{Unbulleted list|Michal Kovařík|Tomáš Kozelek|Albert Bertolín}}
| producer =
| designer = Michal Kovařík
| artist = {{Unbulleted list|Albert Bertolin|Václav Benč}}
| composer = Daniel James Taylor<br>Petr Wajsar <small>(Space Age expansion)</small>
}}
Keeping the information up-to-date is a big task for a fast-moving game. Imagine editing the card every time the dev changes the hp of the tank. The Factorio blog describes the benefits of this better than I can:
> Together with scripts, templates allow us to partially automate updating the wiki to the newest version, and alongside access to the game's source code, it is possible for me to update the wiki to a new version within a few minutes
The blog post has a lot of other interesting stuff in there as well:
https://factorio.com/blog/post/fff-233
A whole lot of work goes into making game wikis pleasant. I'm sure you could duplicate this all on Wikipedia as well, but that's a lot of additional work and communities already struggle to keep their Fandom wiki's up-to-date. The old Factorio wiki does not even have an article for the tank which has been in the game for years:
https://factorio.fandom.com/wiki/Combat
Out of date info is wrong info.
Did you purposefully pick one of the items that has an in-game naming conflict and a disambiguation line?
Being focused to a single game means there's only a few conflicts, but it's not zero.
An encyclopedia excluding exhaustive videogame information is reasonable.
If you don't, you end up with 20 articles for real life fish species, and 20000 articles for fish species in videogames.
Especially since we are talking about an online encyclopedia, Wikipedia is cognizant of its biases and tries to counter them.
The notability policy predates the creation of Wikicities significantly.
I can understand strict rules about reliable sources, independent point of view, or fighting SEO/spam. Still, satellite wikis could exist. Or Portals could serve as some workaround to the notability rule. Or something else.
Wikipedia is by nerds and largely for nerds, I'm sure allowing nerdy topics there would help everyone in the long run.
[1]: https://vampire-survivors.fandom.com/wiki/Vampire_Survivors_...
- more "reputable" news sources AI generate their articles
- traditional publications are facing existential threats as their revenue source is drying out ("Google Zero"), and we can't expect their number to grow in the future
- more experts and analysts are gatekeeping their work under subscription paywalls and alternative services. Paid mailing lists for instance won't be a valid and verifiable source
It feels like at some point Wikipedia could be basically frozen if they can't adapt to the new landscape, but from the outside it doesn't look like an organization that can sensibly move on and change core parts in a pragmatic way.
Notability criterion is important.
I agree on the second point but fundamentally extensively in depth information about the inner workings of a fictional piece are best acquired from the work itself. Not gleaned from an encyclopaedia.
My understanding is that the series of events was - stuff kept getting rejected from wikipedia, so jimmy and some other people made wikicities (now fandom). I dont think its coincidence that jimmy saw a business oportunity, but wikipedia rejecting stuff came first, and jimmy does not control wikipedia processes.
He and the employees of his organization obviously had an outsized influence on it.
Investors don't give a shit about making sustainable products or really anything valuable to society, as long as they can fleece unwitting consumers for massive eyeball money for a short time.
Examples:
Fandom: https://noita.fandom.com/wiki/Noita_Wiki
Official: https://noita.wiki.gg/wiki/Noita_Wiki
Fandom: https://terraria.fandom.com/wiki/Terraria_Wiki
Official: https://terraria.wiki.gg/wiki/Terraria
Fandom: https://minecraft.fandom.com/wiki/Minecraft_Wiki
Official: https://minecraft.wiki/
Two of the three Fandom wiki's above are still ranked higher than the official ones on Google.
As a concrete example, Path of Exile moved to https://www.poewiki.net/ (which is a single MediaWiki instance not associated with a larger network). The content is quite good but it took probably 18 months for it to start reliably appearing in google search results.
LibRedirect has an automatic Fandom redirect option, so if someone sends you a fandom link, or you otherwise click one without realizing it, you won't wind up on fandom. You still give the analytics that you clicked a fandom link to whoever served you the link though.
I typed “memory alpha” on the search page you linked to.
Exception raised in Racket code at response generation time:
json-pointer-value: contract violation
You can use one (or both) of these extensions to avoid Fandom:
* https://getindie.wiki/, should redirect you to any not-fandom equivalent
* https://support.wiki.gg/wiki/Wiki.gg_Redirect, is wiki.gg wikis only and will see updates for such wikis more frequently
Definitely takes a lot of effort, but can be done.
My takeaway from the recent Google search engine ranking documentation leak was that they very heavily uprank big brand names. A page on reddit with various short comments (many of them useful I'm sure) is going to always outrank a random individual's blog with a well-researched article about the same topic. Same for Quora, Facebook, and the other websites with big moderation teams and where we all know what to expect.
There is something to be said for it as well: the quality of a reddit discussion or quora answers will probably be more consistent than if you serve up random other pages (those may be content farms, ad farms, unresearched opinions...). Other pages on the web may be better, or may be worse. I'm guessing that Google figures/found they look better if they don't serve up (m)any bad results but rather pretend that big brands are the internet now, sticking to an average (mediocre-ish) quality
Perhaps that's why Google likes big brand Wikia better than this new domain
^https://pathofexile.fandom.com|https://poewiki.net
Alternatively the Redirector extension can also achieve similar results, without having Kagi, and converting all links, not just search results [1].
It would be less curious if the post was saying something around like "I object to fandom monetizing at all", but also that also probably feels like a less sensible thing to say.
Does it? It's not like Fandom produces the content they monetize.
It can't be all bad, right? More and more, it seems like VC is the only way you can go in tech unless you have a really specific business model.
Opening the wiki now on the domain they host themselves, it is also very noticeably faster to click around. Sounds like a big win for the Minecraft community and kudos on doing such a large migration! Having tried to migrate Telegram groups to Signal, among a group of people studying IT security no less, it became very clear to me how hard it is to migrate communities
For paid offerings, with a little more freedom than a wiki farm host can allow, WikiTeq, MyWikis, and WikiWorks are all good offerings in the MediaWiki space. I currently work for WikiTeq, but they're all pretty good, and the owners are on friendly terms with each other, with a lot of them coming in through the Wikimedia community or staff.
Cloudways for a while had a managed MediaWiki VM offering, but they discontinued that I think. Are there standard best practices for hosting MediaWiki these days, or is that part of your secret sauce...?
Generally I would go with Canasta, a Docker image that's collectively maintained by a few MediaWiki companies. Canasta upgrades with every LTS release, which is a much easier pace for maintainers. https://github.com/CanastaWiki/Canasta
Parsoid got ported to PHP and made a core feature in 2019 and no longer requires being run as a separate Node.js service. The Visual Editor (which most people ran Parsoid to get) just works out of the box now.
Update difficulty scales about evenly with the number of third-party extensions you rely on, especially if CirrusSearch or Semantic MediaWiki are among them. But if you stick to stock and to the LTS version line, it's relatively painless since 1.19 compared to before it.
Between general PHP performance improvements and php-fpm compatibility, it's frankly overkill to run anything beyond the official mediawiki:lts-fpm Docker image[1] while applying whatever PHP config changes you need for your use cases to the Dockerfile. Throw in SQLite improvements and even bare-metal admin isn't much of a hassle these days compared to the bad old days of juggling PHP versions and extensions, configuring Varnish, running Parsoid, patching Mediawiki extensions and skins, and doing MySQL/MariaDB/Postgres upgrades.
A handful of well-supported responsive Mediawiki skins also means the stupid mobile-domain tricks aren't necessary anymore for most users to have a decent experience.
I'll have to check back in with that wiki I used to help with and see what they're doing these days...
Cargo's still easier to work with and its query syntax is a lot more SQL-like, even with the reduced feature set.
Having a couple Fandom pages open in tabs can drive my laptop fan crazy and turn my phone into a griddle.
It's all the FanCoin it's mining in situ
Dynamic user generated content is going to do two things:
1) it's going to result in you needing a database and a web server, capable of a few concurrent users (in addition to bots and guests) but likely sized to handle your spikes in traffic... i.e. over-sized, let's ballpark at $50 per month (people will argue some specifics, but in general this is indicative)
2) it's going to result in you needing to stay up to date, moderate new users, and manage spam... i.e. it requires a disproportionate time investment
and the content is likely crappy, you were just having shits and giggles and nerding out on some small thing that you got excited about... not quite worth those costs.
the solution then is a multi-tenant platform that distributes the ops, cost and moderation problem evenly across many sites, so perhaps the $ cost to you is $1 per month, and time is hardly anything.
but now you've got a different problem, now there's a central entity that "owns" the platform and likely helped itself to a license allowing it to use all of the content, monetise away and profit of the fandom of others.
there are anarchic platforms to counter this, but they're small and cannot be the destination for people to migrate to, and so we're stuck with either the high overhead, costs, and burnout of individual sites, or these crappy platforms taking their users for granted.
though now I wonder... what license is fandom content under? can an independent entity scrape it and start over?
Generally, CC BY-SA https://www.fandom.com/licensing
> can an independent entity scrape it and start over?
Sure. The difficult part is migrating the community and competing in search results. Fandom can and does fight efforts to direct people from fandom-hosted wikis to elsewhere. See this other sub-thread: https://news.ycombinator.com/item?id=40711726
It's not really that intensive when you pretty much just serve cached html to 99% of the users.
That's going to host a gigantic wiki even with something as bloated as Mediawiki running on PHP, at least if you have a remotely sensible setup with (micro)caching for anonymous users.
I'd like to hear if by change someone had archived lyricwiki before its shutdown. I've looked in to it a couple of times but AFAIK the only pages that were preserved are the ones found in wayback machine. There is a dump of various wikia pages in archive.org but IIRC lyrics wikia is missing from that.
I actually find it appalling that browser don't have a proper 'do not autoplay video at all' setting.
There's this thing called "click laundering" though. Where the website will take any qualifying event you trigger anywhere on the page and make it play the video. (I did it myself to make a gimmicky webpage work before)
It's aggravating, but I think there's browser addons to fix it.
Audio isn't the only problem though. I'm starting to think that any non-trivial javascript execution should be behind a permission prompt.
I don't know how you'd define non-trivial javascript in a way that couldn't be circumvented.
One approach would be to start out with the interpreter only with extreme throttling. Then to remove the throttling and enable the JIT you'd need explicit permissions.
Some do, Google Chrome - non Chromium version - does not, because Google/Alphabet owns Youtube.
If the host doesn't let you bring your own domain, find a better one or self-host.
* Okay, most of the editors moved there, and apparently the ones that stayed on Fandom have developed a siege mentality and started banning anyone who mentions the new wiki on their Discord. It's very strange. Fandom is worse in every possible way, where's the loyalty coming from?
>download Indie Wiki Buddy
Seems like a nice plugin.
There are dozens of tools to block ads on all platforms and nearly all of the internet is a cesspool of ads. I don't know how anyone can browse it without an ad blocker. Not really an interesting topic to spend most of the blog talking about.
Pages with ads will rank higher because of Google. It decides what is seen and what is not seen. It will definitely like pages with ads more.
Google does not have to care about quality, as it does not have competition. Yet.
Google search drowned other smaller communities.
The Internet is run by ads. The search is run by ads. Why do we wonder that our content has ads? Now we observe how much a user can tolerate.
There is no alternative business model. The volume and frequency of ads will increase as well, due to Google slashing premiums over the years. The days when the PlentyOfFish guy could bring in millions in AdSense revenue in a single month are long gone.
High-traffic websites will still make more with ads than they will with paid subscriptions. If your website has 1 million users/month and no moat, you'd be doing very well to convert even 0.1% (1,000) of them into paying subscribers. Even that would put you at what, $10k / month? Would that stop you, the site owner, from double-dipping and running ads on paid subscriber accounts? "All the other companies are doing it!"
If we're talking about a wiki, then there's no point in paying for the subscription if you still get access to all the content AND the ads. Plus for that kind of website that's pretty much entirely static text and images $10k a month should be more than enough to cover the cost of hosting.
The problem is i got stuck at that game and searched for a quick solution. Then google straight up made me end up in that pile of ads. I hope people will start realizing what that website is and hopefully migrate their wikis to a better place, although it probably won’t happen.
On interested, here's a 2005 - archived copy: http://web.archive.org/web/20051110015311/http://tvtropes.or...
^https://minecraft.fandom.com/wiki/(.*)|https://minecraft.wiki/w/$1
If they have .wiki domains then they were not the first by a long shot.
you lost me here... Why would you do that?
Surely it's pretty easy to understand.
I also really liked the name wikia, it was like a wiki for anything too detailed for wikipedia.
But today, yeah...
P.S. I never knew it was founded by the same founder as wikipedia!
Because this road leads back to the AOL days of paid forum access.
And such could be said that without an ad-block that this road leads to a gaping chasm full of forced malcontent and sponsored information.
Watching something that you loved and was genuinely useful and important in your life being utterly destroyed by this garbage is pretty motivating to "complaining" and "writing articles".
What compelled you to comment about the compulsion?
Geez.
You pay for services. Some services are free and funded by ads.
Be thankful for the free shit instead of being salty about having to "pay" with ads.
There is no reason to be thankful to Fandom in particular.