Ridiculous article. Organizing a wiki effort, just to spite the original won't work, because that's not enough, and not the right kind of fuel to last for such an undertaking. I feel like these people have no idea what kind of effort is to run an organization, instead seeing the issues as part of the technical, or ideological underpinnings. It doesn't work that way, and I invite every hopeful to look up the list of alternative Wikis that have sprung up over the years.
EDIT: Basically that's how these projects turn out. Having a theme park was never the point; blackjack and hookers were the point. Having a federated Wikipedia alternative isn't the point; having an encyclopedia where you can push an insane description of a topic and present it as reality is the point. And really, that makes for a shitty encyclopedia. Shitty enough for it to disappear before doing damage to the shared sense of reality society needs? No, but still shitty.
Beyond the Wikipedia definition, I've heard the term used to describe the phenomenon of people looking at an ugly codebase with tons of special cases and expectations, thinking they can create a leaner version only to end up going nowhere because all those special cases and expectations that made for ugly UML diagrams had reasons to exist and the overconfidence led to feature creep to justify the time investment to stakeholders. I.e. a good explanation for why it's often better to refactor than rewrite.
Regardless of whether reddit made the correct decision banning those suspect subreddits, it turns out having a community composed almost entirely of those cast-offs is not a pleasant place to be.
Can confirm firsthand. The wiki project I founded was pretty niche (no overlap with Wikipedia) but took hundreds of hours of page creation just to get off the ground. It has to be a labour of love.
This article doesn't even try to explain or convincingly make this argument, it just takes it as given.
I am afraid that the new clone would be mostly full of pages about "Curing Cancer using Magic Healing Crystals", because people writing amount more conventional knowledge would prefer traditional wikipedia.
That article is the classic "Wikipedia criticism": butthurt they can't spread their favourite flavour of shit on Wikipedia without criticism.
I'm not saying Wikipedia is perfect or doesn't have problems, but that is definitely not a good description of anything.
The proposed alternative here is to basically distribute the taxonomy? Anybody who's touched microservices for even a minute will know how hilariously difficult it is to co-ordinate not just the services itself but the teams around them, including both the tech and people level ICP.
Find me one example of institution that managed to go for any significant amount of time (say, longer than a decade) without facing corruption. I double-dare you.
One look at the world of decentralized crypto-currency more than proves that corruption is not limited to institutions.
Corruption seems inherent to humans, the difference is that institutions are well aware of this, and frequently set up controls and systems to balance power and fight the most obvious forms of corruption.
Humans are selfish, fallible, lazy, unreliable, etc. what is every Shakespeare play about? what is the bible about? What is Greek mythology about? What are Aesop's fables about?
- jimbo making CoI edits. He got caught, and edits were reviewed for appropriateness. System worked as it was supposed to
- college drop out on arbcom (for those not in the know, arbcom is kind of like an internal court to solve user disputes. They are not allowed to solve content disputes or say what an article should say, only user behaviour problems). How is that a problem? You don't need a degree to mediate user disputes.
(In fairness, the Gibraltar thing they mention was pretty bad)
Wikipedia certainly isn't perfect, but for some reason they chose some of the silliest controversies.
The problem with Wikipedia is not consumption of content, but the contribution of content.
You probably shouldn't be lying about having one though
Here's a better example:
https://en.wikipedia.org/wiki/Talk%3AAlan_Dershowitz%2FArchi...
Jimbo essentially moved the whole history of an article elsewhere, and rewrote it. So if you clicked the History on the article, you wouldn't see the prior edits.
I mean he made the edits under his own name, not sure there's much to "catch" there. This made it sound like he was editing via a sockpuppet.
The question is, should people be allowed to make their own mistakes, or should Jimmy Wales' mistakes take priority?
Moderation is when stuff is filtered out for you, because you don't want to see it. If you got to see it by mistake, you would agree that you didn't want to see that - maybe because it was factually wrong, maybe because it was disgusting, maybe because it was just irrelevant. Doesn't matter. All moderation is justified in terms of what the user wants to see.
Censorship is when someone hides something for you, because they don't want you to see it. Maybe because they're afraid it'll turn you into a racist, or into an idiot, or into a slave of consumerism, whatever. Doesn't matter. All censorship is justified in terms of what it will do to the minds of the recipients.
Now, maybe there are some people who are afraid for their own minds. Who are afraid that they will turn into a racist, if exposed to racist propaganda, for instance. But it can't be many. So there's little actual overlap between moderation and censorship.
This does not render at all without javascript.
Not an alternative.
> applying all associated edits in order. Instances can synchronize their articles with each other, and follow each other to receive updates about articles. Edits are done with diffs which are generated on the backend, and allow for conflict resolution similar to git.
I somehow got the impression that CRDT was stand of the art in multiple edits to a common document, and sincerely doubt that git's conflict resolution is going to be a good experience
Also, as I understand federated things, the real hazard to usefulness is one of discovery. So how would I, owner of a hypothetical mdaniel.wiki, discover who has the best CRDT documents in their wiki such that I could subscribe to updates on those pages? A wiki of wikis?
Your CRDT merge can be a technological marvel that will seamlessly stitch together different edits on a topic... that say wildly different things about it, resulting in a completely nonsensical article.
CRDT adds automatic conflict resolution, wikis need at most, if any, manual conflict resolution, because their page editing has little need for concurrent real-time edits, it's a slower process (ironic given the meaning of wiki in Hawaiian). About git, OP tried to explain to non-wiki editors what wikis do under the hood, conflict reconciliation in wikis is vaguely similar to git but simpler, and the tech Ibis uses, paired to a design for independence prone to content rotting (every url of the federation can differ even in context about any page), makes me smell of just importing by hand or by trust edits from peers into the backing store.
CDRT is better if you want there to be just one document that everyone is editing at the same time, with merging being an automatic process and the rare non-sensical merge being acceptable.
It doesn't matter all that much whether your servers are federated or served from a single domain, the time users spend with the page editor open is much longer than the latency of a federation request. A server somewhere might be horribly out of sync and cause a lot of conflicts, but presumably those are also the scuffy servers with low traffic that don't see a lot of edits per minute anyways.
CRDT is cool, but it's deploying a very shiny new tech for a small problem that doesn't really need anything fancy.
Manual merging is mandatory to have human meaning, so might as well use it all the time.
"Replacing Wikipedia" strikes me as one of the least essential ideas on making the web better these days, but "developing an alternative news/information thing that anyone can work on and edit" seems cool? Something between the very authoritative "Wikipedia" and the mostly "single-creator" things like githubs awesome lists, rentry's and so forth?
I read a comment here on HN recently, wish I could find it now (I think it was in one of the threads on the 'missing datatype (graphs)'). The gist was that we do large things pretty well: Operating Systems, standard libraries, etc and small things pretty well: single header file libs, specific open source projects, etc, but medium size things are missing. This seems to be the same kind of thing: you have enough complexity that coordination is hard(tm) but you don't have enough scale to build up an institutional inertia to overcome the bus-factor-of-one-ness that--- uhhh
What I'm trying to say is that the support system around that project, like if you have a medium sized project, its going to have individual experts for the parts that make it up, but they're all single points of failure.
If I had more time I would have written a shorter more coherent comment.
Especially common in the DoD where they're good at funding enormous numbers of SBIRs/STTRs, yet they never go anywhere, because all large money contracts are guaranteed to be vacuumed up by Lockheed / Raytheon / Northrup Grumman / Boeing / General Dynamics / Teledyne Brown / Honeywell / ect...
And in most cases (in my opinion, not legally binding), they purposely, slowly build cripple-ware with planned obsolescence that results in equipment that's vestigial before it launches and immediately needs 'upgrade' contracts.
[♪] DoD funding, https://federalnewsnetwork.com/contracting/2023/07/defense-i...
[♫] VC investment, https://www.investopedia.com/terms/d/death-valley-curve.asp
[ꜘ] US Biomanufacturing, https://www.nist.gov/blogs/taking-measure/mind-gap-bridging-...
The reason to make a new wiki or wiki-style technology is to serve a different informational niche, guided by different rules. There's a reason each video game has its own Wiki. You could fork Wikipedia to be more inclusionist, or have people from one particular political viewpoint - if that's your goal.
But re-creating Wikipedia to do exactly what it's supposed to do - form a body of encyclopedic knowledge (subject to copyright laws, etc) - this doesn't make sense, and isn't a convincing argument for a new Wikipedia, even if its distributed.
It sounds more like they want to implement the github fork & pull-request model of version control where currently Wikipedia uses a more SVN type of version control.
There are pros and cons to both models. However federation it is not. The mentioned controversies also seem entirely unrelated to which model you like.
I'm keeping an eye on https://docs.gitlab.com/ee/architecture/blueprints/activity_... <https://news.ycombinator.com/item?id=39201453> with acute interest because I'm totally open to not even being able to see the ways federation can make ecosystems more valuable due to my lack of hands-on experience with the kinds of problems it is solving
For example, I think there's a spectrum of knowledge-base like solutions, but the middle of the spectrum is often poorly served:
- wikipedia: global, canonical reference material. Is very good, and we almost all use it in some fashion.
- confluence/notion/gh-wiki: team knowledge base. Often spotty, stale, neglected.
- logseq/obsidian/org-mode: personal knowledge base, notes. Typically very idiosyncratic, sketchy, but can work very well for the people who put effort into it.
What if a "federated wiki" was targeted at the team/personal level? I'm not saying this is Ibis in its present form, but imagine:
- You keep your personal notes and knowledge store, in a way which is always implicitly contextualized against corresponding info (or lack of info) in the team knowledge store.
- When you're noting something new, or modifying something, you always have an easy path to push your personal addition/edit to the shared store.
- Ideally notes from everyone's work around or interaction with some X drives low-effort maintenance of the community reference of X.
have a local repo of your own notes, a hosted repo serving as the communal ref. if you want to make updates, commit and push --, you can cherry pick the sections you want.
i recall seeing something like mediawiki (but it stored articles as plaintext) a while back, you could use that for a web-based portal too
Looks like a toy weekend project by someone learning web development. A worthy endeavour no doubt but not sure why it's on the front page.
What have I misunderstood?
> Instead of individual, centralized websites there will be an interconnected network of encyclopedias. This means the same topic can be treated in completely different ways. For example geology.wiki/article/Mountain may be completely different different from poetry.wiki/article/Mountain. There can be Ibis instances strictly focused on a particular topic with a high quality standard, and others covering many areas in layman’s terms. Others may document fictional universes from television series or videogames. If one instance is badly moderated or presents manipulated information, an alternative can easily be created. Yet all of them will be interconnected, and users can read and edit without leaving their home website.
This is absurd. You are describing WWW, except for the "without leaving their home website" bit, but I don't know why that feature is so important to you. You can just replace that by "without leaving their home browser of choice".
Based on https://github.com/Nutomic/ibis#federation I think the practical impact of this would be that it'd be easy to have your personal Ibis instance be a fork of some big mainstream one, where the only difference is that e.g. your version of the Pigeons article explains that birds aren't real, but everything else is automatically kept in sync.
I'm not convinced this is particularly worthwhile, or much of an improvement over the existing "run a wiki" workflow.
It isn't that you are wrong as there is some overlap but here are a few properties that federation assures, that the web doesn't.
The web is decentralized. What's served on the web isn't necessarily decentralized.
- Server Decentralization: Federated systems use a decentralized network (www), and are themselves decentralized, meaning the servers that forms the federated network are decentralized, there is no single central authority controlling the network. multiple servers run by different individuals or organizations communicate with each other to share content and data, via:
- Interoperability: different platforms can communicate and share data seamlessly, and also enables easier:
- Privacy and Control as users are given alternative servers, which may operate very different policies, or can run their own if policies from this or that server don't fit their liking, compared to centralized platforms. all ultimately run on the same web but since users can choose which server to join or even host their own, they have more autonomy over their online presence and data.
Analogy with selling buckets of paint:
- Centralization of that service is a one stop shop re-selling exclusive delux white and blue paints.
- Federation of that service is multiple shops, selling white and blue paints. By a maker not moved about the idea of shops selling to each others, not interested in making it exclusive and letting any shops offering other colors.
- the web is all the streets
As much as it's a complicated mess, a centralised system with a broad army of moderators operating to similar standards, is the feature of Wikipedia. They may be wrong at times, but they're accountable somewhat collectively for that and so hold each other to account.
This page doesn't even render on my phone properly. Probably should fix that before it can approach being an alternative
You can easily take Wikipedia's HTML output and use it, but reusing the source of those articles requires a lot of work on either compatibility or translation.
(The "why wikipedia is bad" bit at the start talks about things from 2005, 2007, and 2012.)
Anyway, this might be pretty "it's hard to make a man understand something when his salary depends on not understanding it" since I work for the WMF, but I'm not really seeing much appeal from federation in this area. It certainly seems less relevant than in social media, anyway.
As a small example (not a scandal), a local celebrity mentioned in the (local) media last week that she'd put some incorrect facts on her own wiki page in the past, but can't get them corrected any more.
I tried to suggest a useful addition onto a page recently (against knowing better), and was blocked immediately. Then I noticed others tried the same change several times already, to no avail. So i gave up, and wont bother any more. Also in the future, i wont bother. Life's too short.
Things like this happen all the time without having any visibility.
Its good to be protective, and necessary to avoid spam. But in a few encounters it felt like there are many tiny kingdoms on wikipedia being guarded by their monarchs.
For me Wikipedia is consistently amongst the least biased sources on the internet.
[0] https://en.m.wikipedia.org/wiki/New_Zealand_Labour_Party
At worst it's going to have hundreds of conflicting versions of a document, where each federated node disagree on which is the current master of the document, because each node is fighting for their bias, or it's going to be Wikipedia, where there is one promoted version of the document with whatever bias.
I ran a wiki for local music artists. I started it on an ancient version of mediawiki for use with an ancient extension that wasn't even used, so overall bad idea. It no longer works on a modern version of php so I'm having a lot of trouble getting it back up.
But on that same note, a federation of local wikis, like for small towns/cities, artists that sort of thing, I think would be very cool.
Trying to extend, mirror, or bolster it is potentially beneficial.
Trying to replace it is foolish.
As if the constant begging for donations they don't need and the shithole that is Fandom wans't enough to lower my opinion of ol Jimmy.
i.e stopping scandals & improving trust?
if anything, this seems to be more prone to scandals and less trustworthy.
It's just basic accessibility what we're talking about here.
The most successful alternative to Wikipedia is probably Conservapedia which is largely edited, I believe, by American right wing political activists and evangelical Christian homeschooled teens. Their articles reflect the changes in American right wing politics over the past 20 years or so meaning some are out of date with the current party line because they haven't been touched in years. That's one model of competing with Wikipedia but an encyclopedia consisting of outright partisan political propaganda and outright religious propaganda isn't useful to most people even most people who agree with its viewpoint[0]. Counterintuitively, I think Conservapedia has probably been successful because it rewrote everything from scratch. That also ensures you won't get dinged by Google under the duplicate content policy and basically delisted from search.
I don't think federation is the right way to make a Wikipedia alternative viable because it is already an open source project run on open source software that anybody can fork. It's solving the wrong problem. The right problem to solve is getting a critical mass of editors who aren't just editing to push an agenda. That probably requires paid professional editors whose job is to maintain the encyclopedia. You can probably hire sufficiently smart people for $15 or so an hour because there are probably many smart people already working for that rate at McJobs[1]. That also makes your alternative appealing to current Wikipedia editors who are generally paid $0 to edit Wikipedia. If you're looking to hire expert editors, you'd probably have to pay more but you probably don't need experts. The downside of this is you have to have money to pay editors and probably to advertise that you're paying editors to create a competitor to Wikipedia. That also ensures you have enough editors to maintain the articles. I think Wikipedia alternative with paid editors would probably work as a relatively low risk but also low reward startup idea assuming there was sufficient funding behind it.
[0]: Wikipedia does have its own biases and "neutral point of view" is now basically equated with "objectivity" but it is far less in your face about its biases than Conservapedia which reads more like the timeline of a Twitter account that follows all of the big right wing influencers to Wikipedia's Google News.
[1]: Specifically, high school and college students. Also, many smart people who didn't go to college for whatever reason. They'd probably prefer writing over flipping burgers or running a checkout line.
It didn't?
Compare https://www.worldbook.com/world-book-encyclopedia-2024.aspx :
> As the only general reference encyclopedia still published today, The World Book Encyclopedia 2024 provides authoritative content on almost every topic to learners of all ages
What do you think it would supplement?
I also hear wiki.gg is pretty good, but they're focused on gaming wikis. A lot of Fandom wikis jumped ship to them (of course Fandom is like Hotel California, you can check out any time you like but your wiki content will never come down).
If you think “fediverse” isn’t “toxic” when it comes to controversial issues, you haven’t really checked it out.
- The band: https://www.ebay.com/itm/225562259251
- The hotel chain: https://all.accor.com/brands/ibis.en.shtml
- The anime drawing program: https://ibispaint.com/?lang=en-US
- The beer: https://ibisrice.com/ibis-pale-ale/
- The whatever this is: https://ibis.ibo.org/
- The ibis python library: https://pypi.org/project/ibis-framework/ (disclaimer: I work on this library)
- The ibis python library: https://pypi.org/project/ibis/
- The software for reverse mortgage management: https://www.ibisreverse.com/
- The song: https://www.youtube.com/watch?v=mO-OpFjHRbE
> The only solution is a distributed architecture, with many smaller websites connecting with each other and sharing information.
That is incorrect.
What you want is a block chain.
