4 ms·

But the thing is... there's also a web that isn't like that, but no one has yet built a tool that helps you find it. Like, here's a site that (among other thin

But the thing is... there's also a web that isn't like that, but no one has yet built a tool that helps you find it. Like, here's a site that (among other things of interest to fans) chronologically orders Korean historical dramas: https://thetalkingcupboard.com/resources/sageuk-timeline/ https://thetalkingcupboard.com/resources/sageuk-timeline/ No dark patterns, no clickbait, no videos, no ads, just a labor of love by someone with a lot of time on their hands. The problem for search engines is that it's trivially easy to copy this list and put it up some garbage blogspam domain, but even if the content is exactly the same, I want this list, not the copy, because if I find this list, I can explore the rest of the site and find more great stuff. Even a spammer copies the entire site, this is the one I want because this is the one with a real person behind it who will be updating it, engaging with readers, appreciating the traffic they get, etc. It starts to become clear that what I want my web discovery tool to do is less find individual pieces of good content but more find content from good creators. I don't see any way a search engine can ever do this kind of discovery as well as a human curator. The web has changed so much from when Google launched. Google is optimized for finding content about some keyword because there was actually a time where the was a legitimate problem. You searched "flowers" and Google helped you find the places to buy flowers online. You searched "guitar tabs" and Google helped you find where to get them. Google is still really good at this. Other search engines too. So good in fact that it's become almost invisible to us how easy this is compared to how it was in the past. But now instead of buying flowers, I might want to know... where are the communities where professional florists hang out online and talk about their craft? The Google results for that query are pretty much equivalent to a random listing of websites. But I can almost guarantee there is such a community. Is it a subreddit? A facebook group? Some old vbulletin? A mailing list? Or I might want to know... where are the blogs/channels of the top authorities on web accessibility? What are the best craft coffee roasters in Sweden? What are the essential reads on the history of Botswana? Who are the most consistently good travel vloggers in Spanish? Where can I get help with sourdough starter problems? I can 100% guarantee there are people out there that have great answers to all these questions, but they seem unlikely to be solved by search engines any time in the near future. If someone can crack the problem of leveraging genuine, interested-motivated human curation to deliver results without falling victim to abusive, quick-buck-motivated spamming... they could replace search engines.

the_snooze 5y ago
>If someone can crack the problem of leveraging genuine, interested-motivated human curation to deliver results without falling victim to abusive, quick-buck-motivated spamming... they could replace search engines. That's the tension right there. The moment someone solves the first part is the moment someone will try to hijack it for their own gain. The Internet is dark and full of husslers. They want to exploit people's goodwill and trust---that stuff is costly and doesn't scale, which makes it valuable. If you want genuine interactions, you go to a quiet corner with a handful of folks and be very very careful about who you let in.
- NoGravitas 5y ago
  People don't want to hear that the problem is capitalism, but the problem is capitalism.
- BizarroLand 5y ago
  For instance, I'm imagining a wiki-search engine. The builder would need to load the first few hundred thousand results and then add some sort of user moderation system, where search results get a reputation and site submitters get a reputation as well. Donation funded, with maybe one sidebar section for curated ads that are tied to search results and not with a tracking cookie or fingerprinting. With a few tweaks, that would be a really good start, right? Once it reached a certain point, it would be overrun by the same systems, using human and bot armies to bias results towards the results that would make them the most money. As long as there is a profit to be made somewhere, people will undermine any obstacle placed in front of them to make it, and I can't imagine any fix for it that isn't "run your own servers and pay for it out of your own pocket".
anderspitman 5y ago
> It starts to become clear that what I want my web discovery tool to do is less find individual pieces of good content but more find content from good creators. Bingo. What's interesting about this is that while maintaining a list of all useful content on the web is probably intractable, maintaining a list of all useful content producers might be doable. You could have moderators for each niche, and as the list of producers for a given niche grows, it could split into sub-niches which would get their own moderators responsible for keeping track of quality content producers. Sort of like an awesome list for content producers. I'm certain this already exists in some form. The original web was build around curated content like this. But people aren't used to looking for content this way. EDIT: Saw dmoz.org mentioned below which sounds like exactly what I'm talking about. Apparently they're called web directories[0]. Sadly DMOZ is dead but appears to have a community fork called curlie[1]. [0]: https://en.wikipedia.org/wiki/Web_directory https://en.wikipedia.org/wiki/Web_directory [1]: https://curlie.org/ https://curlie.org/

the_snooze 5y ago

>If someone can crack the problem of leveraging genuine, interested-motivated human curation to deliver results without falling victim to abusive, quick-buck-motivated spamming... they could replace search engines. That's the tension right there. The moment someone solves the first part is the moment someone will try to hijack it for their own gain. The Internet is dark and full of husslers. They want to exploit people's goodwill and trust---that stuff is costly and doesn't scale, which makes it valuable. If you want genuine interactions, you go to a quiet corner with a handful of folks and be very very careful about who you let in.

NoGravitas 5y ago
People don't want to hear that the problem is capitalism, but the problem is capitalism.
BizarroLand 5y ago
For instance, I'm imagining a wiki-search engine. The builder would need to load the first few hundred thousand results and then add some sort of user moderation system, where search results get a reputation and site submitters get a reputation as well. Donation funded, with maybe one sidebar section for curated ads that are tied to search results and not with a tracking cookie or fingerprinting. With a few tweaks, that would be a really good start, right? Once it reached a certain point, it would be overrun by the same systems, using human and bot armies to bias results towards the results that would make them the most money. As long as there is a profit to be made somewhere, people will undermine any obstacle placed in front of them to make it, and I can't imagine any fix for it that isn't "run your own servers and pay for it out of your own pocket".

anderspitman 5y ago

> It starts to become clear that what I want my web discovery tool to do is less find individual pieces of good content but more find content from good creators. Bingo. What's interesting about this is that while maintaining a list of all useful content on the web is probably intractable, maintaining a list of all useful content producers might be doable. You could have moderators for each niche, and as the list of producers for a given niche grows, it could split into sub-niches which would get their own moderators responsible for keeping track of quality content producers. Sort of like an awesome list for content producers. I'm certain this already exists in some form. The original web was build around curated content like this. But people aren't used to looking for content this way. EDIT: Saw dmoz.org mentioned below which sounds like exactly what I'm talking about. Apparently they're called web directories[0]. Sadly DMOZ is dead but appears to have a community fork called curlie[1]. [0]: https://en.wikipedia.org/wiki/Web_directory https://en.wikipedia.org/wiki/Web_directory [1]: https://curlie.org/ https://curlie.org/