Posted by edward 3 hours ago
But the future will raise the bar. Perhaps we'll see wiki of wikis, or something else getting the web closer to truly semantic knowledge.
Anecdotally, I’ve had a lot of experiences like this with newer generations of people on topics like open source software or community projects: They’ve been trained, usually by a social media niche, to believe that the only good fight is to actively harm corporations. Their idea of the world is a very clear black and white contrast with people on one side and corporations on the other.
Such a clean separation of the world only exists in ideas they get from social media and movies. Encountering a project like OSM or an open source project where the largest users (and for OSS, maybe the largest contributors) are people involved with corporations doesn’t fit into their worldview, so they file it in the “corporate” side of the divide and dislike it.
Or maybe they just feel that corporations are actively trying to harm them and this is the logical response?
I don't think that's unfair at this point.
So I don't know that it's fair? Nearly all the material goods around me and a ton of services were provided by corporations, which are largely composed of regular people glued together at scale by bureacracy.
Corporate food products (at least from the US) seem to be designed to just get you to consume more of them.
Shock horror, people tend to care most about things that they feel personally affect them.
This is something I've noticed as well.
- Movies made after WWII that needed villians often made them Nazis.
- During the cold war, evil villians were communists or Soviet spies.
- Movies made after the fall of the Soviet Union often make their villians evil faceless corporations.
A generation or two has been raised by Hollywood and Disney on business always being evil, while simultaneoulsy choosing to consume and make themselves dependent on giant corporate content.
It’s fascinatingly devilish, isn’t it? They control the outrage and profit from it. I’m reminded of Bill Hicks’ bit on marketing.
However on the other hand, it has been drilled into your head repeatedly that wanting compensation for work is "bad" and antithetical to social movements. You work for the whole, not for yourself. Besides, you (along with everyone else) gets to enjoy the fruits of others free work.
But then after dwelling on it, you realize that 90% of the people holding you to this ideology aren't contributing much if anything at all, and just gorging themselves on the work of a few.
So now you either consolidate power into your hands (nazi mods), lose motivation (project decays), or become an "evil" capitalist(start charging money for access).
You can be compensated by "gorging" on other open-source that you haven't contributed to, working outside OSS, or asking for funding to keep working or work more. But if you work privately, you're competing against someone who will work publicly, while limiting someone else who would like to use your private work but can't afford it (like you can't afford others).
You never hear about them but there are also invitation only communities where contributing is required. The benefits are not shared outside.
I get why people might not be motivated to join us in the tragedy of the commons.
Then the project loses visibility to end users. I don't know if that is worth the occasional commercial contributions from them to satisfy their own requirements.
> younger generations raised on corporate platforms and their phone apps are less aware of non-corporate alternatives
All the new guys I work with are trapped in the app store/play store bubble. If it's not available directly through the device's software distribution system it's not getting used.
> might even be suspicious of libre and altruistic movements
All my boomer bosses think that anything that is outside the Windows/O365 ecosystem, or doesn't come with a $150 yearly license, is clearly pirate hacker software that will give your e-mail a virus and steal your bitcoin.
It is increasingly frustrating when we have a problem that could be fixed with GNU corutils and a bash script, or a raspberrypi/arduino translating between 2 serial devices, but we end up getting some $1500 black box from the 'official' vendor. It teaches the newer guys that step one is not to think about the problem, it's to grab your wallet and call a salesman.
</tangent>
It's the whole reason management consultancies exist. If Deloite tell you to do X, or Gartner say buy Y, and you do, you aren't going to have the blame. Even when you end up losing half a billion as our last CxO did, you'll just get a golden handshake and a move to another company.
I understand your frustration but I’ve given up fighting with IT and just let them sort it out
Of course, all the software in the distro repo is there because somebody went against this tendency.
Turn it all off, bury it in concrete, toss the concrete in a river and hurl the river into space, and maybe then IT will be happy with the security.
The full list is at https://wiki.openstreetmap.org/wiki/Editors
Did you ask for help in the community group for your country?
> That could be fixed easily if there was a convenient app that recorded my driving and uploaded it to a central source, processed it, and converted that into edits for me to confirm.
Mapping requires more than that. GPS traces can be useful in areas where no recent aerial imagery is available, but automated importing of that would cause a lot of errors.
Then please publish it. Others can ask Claude too but it's stupid, and one central repository will have everyone's bugfixes and feature recommendations.
Craft is Wikipedia / OSM for software. We're building an open source AI software factory to agentically remake everything.
We're going to open source all the apps, turn platforms into protocols, build open infra, and help assist open weights.
Everything will be rewritten as high quality open source software. And once we've covered the space of everything people use, we'll turn the components into highly reusable open source libraries that can be reshaped into new workflows in real time.
We're going to make a seamless bridge between local and cloud AI, we're going to make a rich set of primitives for fine tuning and sharing.
Not just applications. Data, infra, hardware, the whole of technology.
I've been around long enough to know that sustained collective enthusiasm can sometimes blindside you.
I took a brief look at the posted projects and quickly surveyed the community behind it and I'm not ready to dismiss them yet. I'm not exactly sold either, but I am going to put them on my radar and check in every few months.
This post box is no longer there. "I'll drive round and check", "no need I'm stood there, it has gone". "I'll drive round tomorrow"...I got the impression i was on someone else's patch and gave up
But with regard to bodies of water, there are systems that monitor OSM and warn if someone has tried to "move them to the right place". These details were often added to OSM from professionally surveyed maps, so they are in the right place. However, ordinary people's consumer GPS is typically a few meters off, and they end up moving things to the inaccurate coordinates on their screens.
Meanwhile, the Dutch community also imports all official addresses, but there the tool simply allows for these to be used with other OpenStreetMap tags. It makes you wonder about the Danish approach.
”We should respect freedoms and not restrict them arbitrarily… except when I do not like the people using those freedoms.“
It is always a weird take to me. Pick a license that matches your preferences, like the GPL. Do not pick MIT/whatever and then complain when corps make use of the freedoms you afforded them.
But you can also take a noncommercial GPL software, slightly modify and put your logo and name on it, do some marketing and get cash out of it without contributing anything significant. You just can't make it proprietary. But it's still an unfair situation.
If you don't want your software used commercially, don't Open Source (or Free Software) it. Maybe a CC-NC license if you want to give it away for noncommercial use.
The rest of us will continue to use and make OSS and enjoy it without worrying about what others are doing with it.
But the GPL goes even further than that, and this appears to be the source of your confusion. The GPL stops people redistributing the software under terms that would restrict or prohibit use, including commercial use. The GPL guarantees that anyone receiving the software now or in the future has the same rights you have.
Many human problems could be solved if humans were less willing to compromise with greedy people, but unfortunately authors are too nice in general to do that.
What does stick in the craw with many of us, however, is that Meta, Mapbox, and various other bigcos have nonetheless decided to disregard the attribution requirements of what is a fairly permissive licence, simply because they're sufficiently big that they can say to the OSM Foundation (annual turnover about £3.27) "suck it, we dare you to sue us".
It’s the same thing with freedoms
What’s interesting is that most younger people I interact with now have a better idea of what’s possible through crowdsourced collaboration than older people. In the tech world, people who grew up with crowdsourced sites have seen the end state of Stack Overflow and its moderation, or they’ve seen enough controversies around Wikipedia edit wars where controversial topics become battlegrounds to see which moderators can wield the site’s rules more effectively to revert edits they don’t want on the page. On a long enough timeframe, community projects are at high risk of turning into fiefdoms for power hungry moderators.
The younger people grow up seemingly seeing the same things, but having a different experience about what it means. Talk to any younger generation gamer and they’ll have stories about Discord or Minecraft communities collapsing under power tripping moderators. But in the newer world where internet resources are cheap and now that LLMs can help migrate things, they may have established a new Discord or moved the community to a new game server by the end of the week. Their favorite game emulator or other project may have been forked 4 times already and the community will move to another fork again at the first signs of the new maintainer not doing what they like.
StackOverflow used to be one of the big ones in that area and that is as good as dead. I see the same coming to TVTropes, fandom wikis, movie dbs and Co. once AI gets a good at interpreting video and getting access to some sizable movie libraries. When it comes to /r/tipofmytongue/ style questions AI is already much better than large groups of humans together, though only if the necessary information is available on the public Web.
Lack of access to content outside the Web is really what is hampering progress so far, but that could go away very quickly whenever any of the big players decides to let AI run wild on whatever libraries they have.
As for humans, I just don't see how they could compete. Wikipedia took decades to write, millions of volunteers and Grokipedia (despite being a bit of an unmaintained mess) caught up to its article count in a few months. On top of that comes that all those projects aim for a neutral POV, which is very non-human to begin with and only going to benefit AI.
1. the culture of sharing your work 2. the sense of usefulness and purpose
I'm not sure the newest generation share the 1, but in any case LLM companies destroy the 2.
Why would you thanklessly work for free and for the community if your work is pillaged and sold for a lot of money by AI companies before the project even reaches a critical mass ?!
It’s not over… “OVER” but it’s sure feeling that we are heading there, though.
I won't post a link here, but check out KamTape.
Good AI contributions are better than lazy crap or trolling from humans, right?
but worse than no contributions at all
> we'll just have to converge on the idea that quality is what matters.
for me it is not changed at all, just that volume of spam increased and it is now harder to detect
With AI there is an infinite stream of contributions to curate. No human is ever going to volunteer for that, so unless you keep AI contributions very tightly restricted you'll rapidly lose all human contributors. At which point nobody is left to ensure only good contributions get in, so it is only a matter of time before it devolves into a giant pile of crap.
Besides, who's going to pay to run those AIs? Good models aren't free, so you are effectively turning it into a pay-to-win project. You can be absolutely certain that some tech billionaires are going to burn a few million to have "good AI" constantly rewrite their pages the way they want it.
I used to contribute to [X] project because it felt good to make something, but realizing that someone is getting a dollar on the back of my work that I am not is just not worth it.
Wikis by article count:
Wikipedia: 65 million (English: 7.2 million)
Baidu Baike: 30 million
Hudong Baike: 15 million
Namuwiki: 8.7 million
Fandom: 40 million
In fact, the Everquest 2 wiki dwarfs the Yongle encyclopedia at ~400k articles.
"AI’s absorption of Wikipedia means that more people than ever are reading its content. Yet it also means that fewer people are visiting the site itself. When a Wikipedia link is suggested by a search engine, nearly a tenth of users click through. When Wikipedia is cited by an LLM, the clickthrough rate is only 0.33% (and not all credit their sources)."
Individuals may keep blogging out of the force of habit or to cater to narrow / walled-garden audiences, but there aren't that many people you can attract to a planetary-scale project with the pitch of "let's edit a training dataset for OpenAI together".
Some folks are saying that AI will only accelerate the creation of future Wikipedias, and that's obviously missing the point. Yes, there might be Grokipedia-style entities, probably a number of them.
They do show signs of stopping.
From your list, I think most were started 15-20+ years ago[1] so your argument isn't contradicting the original point.
* * *
[1] OSM, Fandom and TV Tropes were started 22 years ago. Wikipedia itself was started 25 years ago.
I don't know what the future will be, but the future people who live in it will know and they're going to do something. I can say that with confidence, even without knowing exactly what they're going to do.
edit: grammar - see reply for old/wrong version.
Unless you have a political view to push. Visit Grokipedia today!
"Would you like to know more?"
Let’s take 2 of your examples: OSM still has major gaps that seem to be getting worse actually (so much outdated or missing stuff and I’ve seen the community worry about interest) and fan wikis have almost all consolidated onto the ad riden, unreadable monstrosity that is fandom.
I think this will be the case for other things too. OpenStreetMaps won't need as many contributors as more sensors come online, eg. sensors and footage from hobby drones and people's phones augment how much information a single person can enter about a city.
But then it'll just be your AI slop and no one will care. If they want one they'll slop one up themselves instead of using yours.
The people building these technologies do not have a good understanding of humanity (they have some oversimplified view focused on reducing friction), and even if they did their goal is not to help communities from but rather extract the most profit for themselves out of society. They won't maintain the friction that helps communities form and keeps them cohesive, just like they won't maintain the friction that keeps us healthy as individuals.
> There are all sorts of possible ways we can do some large scale crowdsourced effort and its probably going to be realtime news e.g video streaming live events on an open platform not owned by corporations. Where there is no misinformation. No one can lie to us, we can see it happening.
That's a fantasy. The technological trend is towards more corporate ownership and more misinformation (or if there's less, it will only be because the corporate overlord AIs will suppress misinformation that's against their interests, and promote other misinformation that suits them better).
I think you're generalizing a brief summer, when the web was new and the power-that-be had not yet muster their resources to control it, as the natural state, when it was really a happy but unsustainable state.
Yes, some projects like stack overflow may be losing contributors by a large degree, but it doesn't have to be this way
I run SponsorBlock and DeArrow which are two human powered crowdsourced browser extensions that are only growing in scale in opposition to the want to slopify everything
We have 17% of submitters every week being first time contributors and 19% being old, a nice balance.
Unfortunately the talk isn't available yet but you can see the slides here https://docs.google.com/presentation/d/1nGArM7c3i9cVGGvkh6ER...
That doesn’t sound like necessarily a positive thing; it sounds like it could well be unsustainable churn (impossible to tell from such limited numbers, I don’t know what the general growth figures are or proportional trends).
I have contributed in the past, but don’t any more because of how you went out of your way to make it tedious: https://github.com/ajayyy/SponsorBlock/issues/2106. I suspect there are a lot more that have been driven away by the unnecessary and not even accurate mandatory preview step.
And attempts like everything2 and h2g2 were neat art but nothing like Wikipedia. The spinoff attempts to do something like Wikipedia also failed.
This is characteristic of something that is probability zero in its particulars but rather likely as an example of a class.
Humans do like getting serious and accomplishing something before the twinkling light of our lives fades.
Wherever there is a community, people will contribute; people love being social, and many platforms do this broadly and well enough.
- OpenstreetMap
- GitHub ( people contributing Code, issues and bugs)
- All Reviews Platform ( this creates such a sign that today, whenever people buy or use any product, they would like to see reviews)
Thousands of others with no sign of stopping, and the future will keep on bringing more.
Peer-to-peer is a social culture that will always be there, and whenever new things come up, a similar peer-to-peer culture will back them.
Isn't that availability of information the entire point of Wikipedia? Heck, wasn't that the point of Google before they turned evil?
Probably half of their edits are flame wars between editors with different political slants.
With a LLM, the system prompt can tell it exactly which way to bias the search space, and you only have to do this once. And no, open weights models won't save you because someone will figure out a way to embed the system prompt into the weights themselves (if they haven't already have).
I think they've failed at that, then. I regularly come across articles that have every slant you can imagine - political, technical, and corporate (written as marketing material for a product).
The value that I get from Wikipedia is in its aggregation of sources - not it's neutrality, because they're regularly cherry-picked to favor whatever the spin is - but having some sources is better than having none, which was the pre-WP state.
You can't defeat confirmation bias, but LLMs at least have the ability to provide more objective evidence than Wikipedia (while also being more tenacious when trying to find new sources).
I’m not sure it helps with the “AI absorbs it, nobody visits” problem, but every card links to the page, and I think players will go read them to understand what they've just collected.
I also suspect a lot of collectors like correcting things. I know “I was there” isn’t a citable source (see the OSM post box story above), but photos are a real contribution: many Wikipedia photos are poor or outdated, and I’d like players to be able to submit theirs to Commons in the future.
Built with MapLibre + Protomaps. I just added a waiting list if you're interested trying it out!
Nothing public to show, but I have learned a lot about my local area by testing it out.
Yours looks much more polished, I'll look forward to trying it out!
For an alternative use: the OSM app CoMaps fetches those articles and also includes them in the offline maps.
Last time I was walking around a tourist city, I didn't even bring a tourist guide book. I just followed OSM to the sights, and read the Wikipedia entry about the things I saw. All without using my phones roaming data plan, or even leaving the maps app.
If we get to the more general idea of "can humans collaborate in the same way they collaborated to make Wikipedia". I see no reason why not. Collaborative projects similar in spirit to Wikipedia pop up all the time.
Spirit is not sufficient. I can start a new project with the same spirit and a user base of one.
Scale is part of what makes Wikipedia. What projects are similar in spirit and scale?
I became a Wikipedia editor not long after the 2nd version of the site, before they even had logins, IIRC.
I was banned from the site probably 7 years ago or so by a troll who collaborated with the admins. The review "process" was more like The Trial and the ban was quite terrible, because not only did they install some deep cookie in the browser, but they permabanned the IP address.
Wikipedia is gone. Whatever it even was, it is not now.
The title evokes a romantic sense of wikipedia as a well governed, neutral egalitarian multicultural collaboration. But anyone who has been marginalized by the cabal knows it was not most of those things.
We dont use "deep" cookies, just normal cookies. If memory serves, 7 years ago Wikipedia didn't use cookies at all in the block process, it was purely ip based back at that point.
Name one started after 2022, with at least a modicum of success.
After a quick search I stumbled upon Wikifunctions (2023), but do Wikimedia projects that are not Wikipedia count? A bit more controversial would be Anna's Archive (2022).
We have to think about the needs too. There was a need for an online encyclopedia, Wikipedia took it. What need do we have now that is currently unanswered and can be answered by a collaborative effort similar to Wikipedia?
Also, open source projects, so many come out every year that it is hard to count.