This is most definitely wrong. People volunteer and crowdsource all the time, driven by ideals. And the async nature of the web will only make such projects stronger. There's OpenStreetMap, TV Tropes, fandom wikis, movie dbs, and a thousand others with no sign of stopping.
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.
Many longtime OSM mappers have sensed that there has been some decline in interest and a risk to the long-term future of the project as a truly amateur-enthusiast phenomenon. Causes include: younger generations raised on corporate platforms and their phone apps are less aware of non-corporate alternatives, and might even be suspicious of libre and altruistic movements (this is a concern in Free Software circles, too); as the map got filled in, there were fewer obvious blanks that struck a total newbie as a place he/she could contribute; corporations got onboard both as OSM users and stakeholders, and then some idealistic anti-capitalistic people have asked why they should contribute if the map can benefit corps, too, etc.
> then some idealistic anti-capitalistic people have asked why they should contribute if the map can benefit corps, too
”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.
GPL should absolutely be the default choice for software. It is quite weird when people choose the MIT license and then complain when others use their open source software and turn it into a proprietary for profit software when the author explicitly allowed them to do this in the license. The MIT is really not long. People should read it before using it.
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.
I look forward to human curation to be reinvented in some way. Social media algos are a tyranny. I think something new will be driven by people's desire to be different and niche.
That’ll just have to be 100% analog, and that’s until Physical AI has a massive breakthrough where they’ll make the analog also possible to be automated away.
It’s not over… “OVER” but it’s sure feeling that we are heading there, though.
But who's going to filter out the good AI contributions from the bad AI trolls and bargain-bin-model slop?
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.
It’s the infinite value extraction hustle that’s killing the fun of projects.
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.
The food shelf I volunteered at yesterday is a lot more than 20 years old. There'll probably be another Wikipedia, because we already have Wikipedia. There's no point in having a second. Assembly for other things listed. We have one. Why would we need a second one to do that?
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.
And there is no need to, because we already have Wikipedia.
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.
_Because_ Wikipedia is community driven and independent of any corporation, shareholder, or stock exchange, I tend to view it as something no commercial encyclopedia could ever have competed with.
Now, let’s imagine the same thing happening with LLMs and AI. Local, open-source, community-driven models will build better things than any commercial LLM or AI could ever create.
Yes, this is mostly a dream as of today. But will The Economist still be alive to validate this idea that humans, stripped of the money schemes, will have built greater (aka create a project like Wikipedia) things than ever ?
So why have we stopped building pyramids for thousands of years? We have the technology and knowledge to do it, so where are all the new pyramids? And no, that Las Vegas monstrosity doesn't count.
You might be surprised to learn that not all challenges are purely technological. The biggest challenges with such megaprojects are political, economical, and social. A Wikipedia clone is completely worthless if nobody is willing to contribute - it would just be an empty database!
Wikipedia needed a very small tech infra and legal support to run back in the day. I have a feeling that now any project needs much more infra AND world wide legal support given the larger amount of laws and complexity involved. Just a feeling.
Can you expect less from a magazine that prays at the altar of capitalism?
People make brilliant things out of altruism as long as they don't have the feeling they're being abused by corporate or monetary interests. That's why the BSD Unixes languished and Linux thrived. Programmers didn't feel the need to contribute for free when companies where the ones reaping the benefits of their work.
Here's a shout-out to OpenStreetMap and its contributors.
If someone wants to learn to accomplish something tangible in the real world, there's probably no better place than YouTube, TikTok, etc.
Yes there are ads, yes there's slop, and yes it still remains a trove of practical human knowledge which would astound a 1926 American far more than Wikipedia ever did.
I love to read and have spent more of my life reading than probably anything else. It's my favorite way to learn. Most people do not love to read and would rather learn from another person.
I think humans will never create another Wikipedia because we're already creating something better. YouTube is just the lucky torchbearer, I think. Now that people know they can film themselves teaching or doing something they're good at, and other people will positively engage and maybe even pay them real money, there's no stopping the firehose.
In 10,000 years I imagine the 2020s will be seen less like the end and more like the beginning of human knowledge sharing.
The wikipedia images are incredibly easy to follow for me and I like that it is easy to copy and paste the info or download a copy, search it, annotate it/modify it, etc. Just recently, there was a thread here about different ways to ties your shoes laid out similarly with images. Many people talked about easily trying them out.
Even if you are the type of learner who requires a video to learn, online video platforms, like youtube and tiktok, with ads & overlays when you pause them, with no out of the box way to download the videos, autoplay to who knows what next video, etc., are terrible vehicles for that. You can't search or annotate them. The quality varies as you stream it.
Youtube and tiktok will not be better than wikipedia because the primary goal, the raison d'etre, of these products is to make money, track your behavior, sell your data to the highest bidder and show you ads based on it. I'm not naive to the fact that Wikipedia does this, but it's obvious that wikipedia is primarily a collective effort and not a corporate product, and I have more faith in a collective group of people than a tech CEO.
Lastly, a video is a video. It is not another person. A person you can ask questions to, they can offer real time advice. I learned to tie a neck tie from family. Teaching someone how to tie a tie, your child, grandchild, nephew, etc., learning how to be an adult from a role model, is famously a rite of passage. If you think a video is a substitute for another person, especially in this age of AI generated content, I suggest you take a step back and reflect on what human connection really means.
Wikipedia was made from the free labor of IT admin staff and disabled people, both had free time in a typical day to build and curate it.
If I were to make wikipedia today it would contain only bullet pointed facts with sources for each. Articles would have unlimited length. Pagination if an article gets over a million lines.
"#1 result for most searches" would literally mean that I could take all the searches for a given day, then find a 50%+1 subset that all have the same top result. There is no such top result.
The comcept of the end of history, including Fukuyama's thesis, is an interesting intellectual topic, and I would argue that it's become even more interesting and relevant with the threat that AI could become.
I predict the opposite: Humans will collectively create something like an LLM, but curated. That will be next level for Wikipedia-like project.
Think about it - LLM is just a computer program that predicts a next word. So it has a very detailed, formal description of each concept it can represent in its residual vector.
I think humans will understand (perhaps with AI help) these formal descriptions and put them into a giant database.
I wonder... not rage bait btw... but PHP was super easy to program. I wonder if things like this are harder to build today. Like, before you could setup a LAMP stack and program a wiki in the time it takes to begin to wrap your head around the current programming paradigms.
I don't think the languages changed; our minimum standards did. Standing up a bare-minimum user login in PHP for a homework assignment is easy. Now fix all the SQL injection, encrypt the passwords securely, lock down the database, make sure your cookies are always first-party, etc. There's a lot to hold in your head for secure web development in general.
It’s not just the code, but the collaboration to write the actual pages.
Regarding PHP, a small wiki or cms was always a starter project back in the day. For sure it’s a good way to wrap your head around a language and project
PHP can hosted in so many places for so cheap. It still has a place, like everything
Wikipedia became popular despite its programming. Anyone who has ever tried to edit a Wiki page knows its editor is absolute garbage, for example. Want to create your own basic DIY wiki in 2026? Just set up a GitHub Pages which auto-accepts pull requests, you'll be up and running in less than 30 minutes.
Besides, nobody is forcing to to use the current programming paradigms. A classic LAMP stack will still work.
Oops, guess it was rage bait. (not OOP btw). I just figured the economist article was about programming. Don't have JS on this computer, so I couldn't read it.
1) I doubt The Economist is suggesting that technical complexity is the reason they think that something like Wikipedia won't happen again.
2) You still have LAMP, even if the A is now often Nginx and the M Postgres. That stack is at least as easy as back in the day (signed someone who in my teens loved PHP so much that I made CLI utilities and IRC bots with it, auch). But that's besides the point, because several programming languages and web frameworks now come with their own built-in web-server for development, so you don't even need a stack to get started! This part is easier today.
This is most definitely wrong. People volunteer and crowdsource all the time, driven by ideals. And the async nature of the web will only make such projects stronger. There's OpenStreetMap, TV Tropes, fandom wikis, movie dbs, and a thousand others with no sign of stopping.
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.
Many longtime OSM mappers have sensed that there has been some decline in interest and a risk to the long-term future of the project as a truly amateur-enthusiast phenomenon. Causes include: younger generations raised on corporate platforms and their phone apps are less aware of non-corporate alternatives, and might even be suspicious of libre and altruistic movements (this is a concern in Free Software circles, too); as the map got filled in, there were fewer obvious blanks that struck a total newbie as a place he/she could contribute; corporations got onboard both as OSM users and stakeholders, and then some idealistic anti-capitalistic people have asked why they should contribute if the map can benefit corps, too, etc.
> then some idealistic anti-capitalistic people have asked why they should contribute if the map can benefit corps, too
”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.
GPL should absolutely be the default choice for software. It is quite weird when people choose the MIT license and then complain when others use their open source software and turn it into a proprietary for profit software when the author explicitly allowed them to do this in the license. The MIT is really not long. People should read it before using it.
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.
TBF corps can use and profit from GPL and AGPL stuff just fine, even if they comply with the license (though they usually just don't bother)
However, it's now operationally much harder to make sure that the contributors actually are humans.
I look forward to human curation to be reinvented in some way. Social media algos are a tyranny. I think something new will be driven by people's desire to be different and niche.
That’ll just have to be 100% analog, and that’s until Physical AI has a massive breakthrough where they’ll make the analog also possible to be automated away.
It’s not over… “OVER” but it’s sure feeling that we are heading there, though.
In the end there's no foolproof way, so we'll just have to converge on the idea that quality is what matters.
Good AI contributions are better than lazy crap or trolling from humans, right?
But who's going to filter out the good AI contributions from the bad AI trolls and bargain-bin-model slop?
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.
Right. And I consider it a blessing that Claude-speak and GPT-speak continue to be so obvious :)
You've discovered some sentient non humans have you?
It’s the infinite value extraction hustle that’s killing the fun of projects.
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.
The real test is the creation rate for those projects.
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.
The food shelf I volunteered at yesterday is a lot more than 20 years old. There'll probably be another Wikipedia, because we already have Wikipedia. There's no point in having a second. Assembly for other things listed. We have one. Why would we need a second one to do that?
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.
>There'll probably be another Wikipedia, because we already have Wikipedia. There's no point in having a second.
Unless you have a political view to push. Visit Grokipedia today!
And there is no need to, because we already have Wikipedia.
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.
I think all those famdoms, main wikipedia etc should be backed up in multiple archives and should be provided to as many libraries for offline read.
It depends. Okay, yes, probably.
BUT, let me slide on the off topic borders.
_Because_ Wikipedia is community driven and independent of any corporation, shareholder, or stock exchange, I tend to view it as something no commercial encyclopedia could ever have competed with.
Now, let’s imagine the same thing happening with LLMs and AI. Local, open-source, community-driven models will build better things than any commercial LLM or AI could ever create.
Yes, this is mostly a dream as of today. But will The Economist still be alive to validate this idea that humans, stripped of the money schemes, will have built greater (aka create a project like Wikipedia) things than ever ?
Humans will never again create a project like _______.
We can't, we don't know how to do it.
So why have we stopped building pyramids for thousands of years? We have the technology and knowledge to do it, so where are all the new pyramids? And no, that Las Vegas monstrosity doesn't count.
You might be surprised to learn that not all challenges are purely technological. The biggest challenges with such megaprojects are political, economical, and social. A Wikipedia clone is completely worthless if nobody is willing to contribute - it would just be an empty database!
Nobody wants a real pyramid
I hope they don't. Wikipedia and the Wikimedia foundation is quite dodgy behind the scenes.
It shouldn't take away from the benefit the project provides.
Wikipedia needed a very small tech infra and legal support to run back in the day. I have a feeling that now any project needs much more infra AND world wide legal support given the larger amount of laws and complexity involved. Just a feeling.
There's that, but also economists are most often wrong then right on their forecasts.
Because...money, money, money!!
Can you expect less from a magazine that prays at the altar of capitalism?
People make brilliant things out of altruism as long as they don't have the feeling they're being abused by corporate or monetary interests. That's why the BSD Unixes languished and Linux thrived. Programmers didn't feel the need to contribute for free when companies where the ones reaping the benefits of their work.
Here's a shout-out to OpenStreetMap and its contributors.
What do you mean? Linux is thriving and companies benefit.
Once AI starts being a superhuman doctor everyone will want to contribute their medical story
If someone wants to learn to accomplish something tangible in the real world, there's probably no better place than YouTube, TikTok, etc.
Yes there are ads, yes there's slop, and yes it still remains a trove of practical human knowledge which would astound a 1926 American far more than Wikipedia ever did.
Compare the Wiki article with step by step PNGs. https://en.wikipedia.org/wiki/Half-Windsor_knot
To the thousands of examples one can find on YouTube. https://www.youtube.com/results?search_query=tying+a+half+wi...
I love to read and have spent more of my life reading than probably anything else. It's my favorite way to learn. Most people do not love to read and would rather learn from another person.
I think humans will never create another Wikipedia because we're already creating something better. YouTube is just the lucky torchbearer, I think. Now that people know they can film themselves teaching or doing something they're good at, and other people will positively engage and maybe even pay them real money, there's no stopping the firehose.
In 10,000 years I imagine the 2020s will be seen less like the end and more like the beginning of human knowledge sharing.
The wikipedia images are incredibly easy to follow for me and I like that it is easy to copy and paste the info or download a copy, search it, annotate it/modify it, etc. Just recently, there was a thread here about different ways to ties your shoes laid out similarly with images. Many people talked about easily trying them out.
Even if you are the type of learner who requires a video to learn, online video platforms, like youtube and tiktok, with ads & overlays when you pause them, with no out of the box way to download the videos, autoplay to who knows what next video, etc., are terrible vehicles for that. You can't search or annotate them. The quality varies as you stream it.
Youtube and tiktok will not be better than wikipedia because the primary goal, the raison d'etre, of these products is to make money, track your behavior, sell your data to the highest bidder and show you ads based on it. I'm not naive to the fact that Wikipedia does this, but it's obvious that wikipedia is primarily a collective effort and not a corporate product, and I have more faith in a collective group of people than a tech CEO.
Lastly, a video is a video. It is not another person. A person you can ask questions to, they can offer real time advice. I learned to tie a neck tie from family. Teaching someone how to tie a tie, your child, grandchild, nephew, etc., learning how to be an adult from a role model, is famously a rite of passage. If you think a video is a substitute for another person, especially in this age of AI generated content, I suggest you take a step back and reflect on what human connection really means.
Wikipedia was made from the free labor of IT admin staff and disabled people, both had free time in a typical day to build and curate it.
If I were to make wikipedia today it would contain only bullet pointed facts with sources for each. Articles would have unlimited length. Pagination if an article gets over a million lines.
No exceptions.
An absolute source and collation of human truth.
nor GNU, nor Linux. This is the truth.
Not with that attitude!
Not with this attitude.
I can't find a way to read the article, anyone have a gift link?
(Edit: archive.today and wallhop.io don't work)
AI would need projects like this to harvest human knowledge so...
Not enough humans even care about the snake eating its own tail. We are rapidly approaching post-truth.
AI, however ...
Wikipedia has been an excellent resource for LLMs, and seems to be one of their prime resources. Make of that what you will.
And before that it was the #1 result for most Google searches.
Not "most searches." Wikipedia was (probably still is!) the the most common #1 result.
"#1 result for most searches" would mean 50% of all searches would have Wikipedia up top, and no site clears that bar.
That’s incorrect. If there are 10 websites competing for #1, Wikipedia could show up twice in 11 searches while the others showed up once.
"#1 result for most searches" would literally mean that I could take all the searches for a given day, then find a 50%+1 subset that all have the same top result. There is no such top result.
The word you're looking for is plurality.
Don't tell me, tell the guy who said "most!"
I suppose humans will never again create a project quite like the old printed encyclopedias, either ¯\_(ツ)_/¯
These are also the geniuses who thought Fukuyama was right when he declared the end of history.
The comcept of the end of history, including Fukuyama's thesis, is an interesting intellectual topic, and I would argue that it's become even more interesting and relevant with the threat that AI could become.
"Humans will never again create a project like Wikipedia" aaaaaand it's paywalled. The irony.
I predict the opposite: Humans will collectively create something like an LLM, but curated. That will be next level for Wikipedia-like project.
Think about it - LLM is just a computer program that predicts a next word. So it has a very detailed, formal description of each concept it can represent in its residual vector.
I think humans will understand (perhaps with AI help) these formal descriptions and put them into a giant database.
You say, building ontologies with a specific intent will actually lead to intelligence at a similar level of current AI?
This would be a severe blow to the bitter lesson, and generally vindicate the AI researchers of the 80s that put a bet on symbolic processing
I wonder... not rage bait btw... but PHP was super easy to program. I wonder if things like this are harder to build today. Like, before you could setup a LAMP stack and program a wiki in the time it takes to begin to wrap your head around the current programming paradigms.
I don't think the languages changed; our minimum standards did. Standing up a bare-minimum user login in PHP for a homework assignment is easy. Now fix all the SQL injection, encrypt the passwords securely, lock down the database, make sure your cookies are always first-party, etc. There's a lot to hold in your head for secure web development in general.
> encrypt the passwords securely
Uh oh.
Whoops :) Don't forget the salt and hashing steps!
It’s not just the code, but the collaboration to write the actual pages.
Regarding PHP, a small wiki or cms was always a starter project back in the day. For sure it’s a good way to wrap your head around a language and project
PHP can hosted in so many places for so cheap. It still has a place, like everything
No.
Wikipedia became popular despite its programming. Anyone who has ever tried to edit a Wiki page knows its editor is absolute garbage, for example. Want to create your own basic DIY wiki in 2026? Just set up a GitHub Pages which auto-accepts pull requests, you'll be up and running in less than 30 minutes.
Besides, nobody is forcing to to use the current programming paradigms. A classic LAMP stack will still work.
Today you can use LLM to get most of that. Whether anyone is willing to commit to the success of a LLM-driven project and a different problem.
Oops, guess it was rage bait. (not OOP btw). I just figured the economist article was about programming. Don't have JS on this computer, so I couldn't read it.
1) I doubt The Economist is suggesting that technical complexity is the reason they think that something like Wikipedia won't happen again.
2) You still have LAMP, even if the A is now often Nginx and the M Postgres. That stack is at least as easy as back in the day (signed someone who in my teens loved PHP so much that I made CLI utilities and IRC bots with it, auch). But that's besides the point, because several programming languages and web frameworks now come with their own built-in web-server for development, so you don't even need a stack to get started! This part is easier today.
A LAMP stack is still possible. Some lean toward alternative "frugal" stacks, like https://learnbchs.org/
It’s not like choice is not there.
In 2003-2004 the benefit of php was really php.net letting you comment on any function
Also here, that’s still an option, and comments are still there and (I guess) open to new entries. So what is the concern pointed at here?
Economists (not the website) and accurate predictions are two very different things, first.
The economists, the website site, second.
Humans will never fly, etc. Third.
(Extra point for the paywall).