3 ms·

Interesting that the author doesn't mention neither Wikipedia nor pinterest. They are the two biggest curators led website. The problem OP is mentioning is ex

by polote 5y ago

Interesting that the author doesn't mention neither Wikipedia nor pinterest. They are the two biggest curators led website. The problem OP is mentioning is exactly why I read HN and that I have a newsletter about the blogs of HN [1]. If you want to have opinions on a topic, the only thing you can do is read personal blog posts about the topic. I'm not sure you can make a search engine out of that, because at some point being the #1 will provide money to someone and he will be able to game the ranking. Look at Wikipedia on a business topic and you will see only the most popular tools are mentionned and on pinterest the results are now full of pinterest SEO experts. The person who will be able to prevent manipulation on the internet internet is going to be a Billionaire. I'm very bullish on curation, especially at work, because employees should be less interested about their self interest and could curate the knowledge of the company, but for the web I don't really know how to do it [1] https://hnblogs.substack.com https://hnblogs.substack.com

lrem 5y ago
I find the fact you mention Wikipedia and Pinterest in the same breath surprising. As someone who doesn't really use the latter, I always knew it as the site to exclude from search results. Whatever the query, a result in Pinterest will look like utter spam that only used my query to get into SERP and then try to make me go to something barely related. On the other hand, Wikipedia seems to be always on topic and usually high-quality content. I understand that one medium is more vulnerable than the other. But still, with somewhat similar content creation structure, I'd expect more similar outcomes. So, what is the main reason for the perceived quality difference?
- polote 5y ago
  They have different goals, Pinterest's goal is to let user create collections of things, Wikipedia's goal is to be right. Pinterest is useful if you are looking for ideas of things that look similar. If you want to choose a pillow, just type pillow and you will find a list of collections of pillows. It doesn't make sense to show Pinterest results in Google because that's not what Google users are searching. The fact that you want to block Pinterest is just because Pinterest is sucessful and has overflowing google images has a SEO strategy.
  - bsdubernerd 5y ago
    Ideas as in "search some term, get a not-even-vaguely-related picture collection you cannot see unless you log in"? Or the "ideas" that also happen to hide where that very idea came from? Pinterest is scourge for a text search engine. It's also scourge for an IMAGE search engine, because it's just hiding the real source. Pinterest is in the top 10 websites I would blacklist from search results.
1vuio0pswjnm7 5y ago
Compared to the garbage results that the popular and wannabe popular web search engines serve up on a regular basis, Wikipedia search is vastly underrated, IMO. The results are qualitatively different when the search engine project is not being sponsored by advertising. And Wikiepdia is happy to give you as many results as you want. No easily triggered rate limits for consuming too many results too fast. By no means an expert on Wikipedia searches, but here are a few examples. There are many more searches and options. This is a standard search for Wikipedia entries exactly matching a query. It will redirect to an entry if one exists. Sometimes it may be necessary to navigate to the disambiguation. alias links="links -no-connect" q=podunk; links "https://en.wikipedia.org/w/index.php?limit=500&profile=all&search=$q&ns0=1" This is another similar search. It will not redirect to an entry if one exists. It will return all entries that have text matches the query. q=podunk; links "https://en.wikipedia.org/w/index.php?title=Special:Search&profile=all&search=$q&fulltext=1" This is a search I use to explore websites. It searches for domains found in external links. For example, one could search with the query "podunk.edu" and find all the links in Wikipedia articles to pages at Podunk University websites. q=podunk.edu; links "https://en.wikipedia.org/w/index.php?title=Special:linkSearch&limit=500&profile=all&target=$q&ns0=1"

by polote 5y ago

lrem 5y ago

I find the fact you mention Wikipedia and Pinterest in the same breath surprising. As someone who doesn't really use the latter, I always knew it as the site to exclude from search results. Whatever the query, a result in Pinterest will look like utter spam that only used my query to get into SERP and then try to make me go to something barely related. On the other hand, Wikipedia seems to be always on topic and usually high-quality content. I understand that one medium is more vulnerable than the other. But still, with somewhat similar content creation structure, I'd expect more similar outcomes. So, what is the main reason for the perceived quality difference?

polote 5y ago
They have different goals, Pinterest's goal is to let user create collections of things, Wikipedia's goal is to be right. Pinterest is useful if you are looking for ideas of things that look similar. If you want to choose a pillow, just type pillow and you will find a list of collections of pillows. It doesn't make sense to show Pinterest results in Google because that's not what Google users are searching. The fact that you want to block Pinterest is just because Pinterest is sucessful and has overflowing google images has a SEO strategy.
- bsdubernerd 5y ago
  Ideas as in "search some term, get a not-even-vaguely-related picture collection you cannot see unless you log in"? Or the "ideas" that also happen to hide where that very idea came from? Pinterest is scourge for a text search engine. It's also scourge for an IMAGE search engine, because it's just hiding the real source. Pinterest is in the top 10 websites I would blacklist from search results.

1vuio0pswjnm7 5y ago

Compared to the garbage results that the popular and wannabe popular web search engines serve up on a regular basis, Wikipedia search is vastly underrated, IMO. The results are qualitatively different when the search engine project is not being sponsored by advertising. And Wikiepdia is happy to give you as many results as you want. No easily triggered rate limits for consuming too many results too fast. By no means an expert on Wikipedia searches, but here are a few examples. There are many more searches and options. This is a standard search for Wikipedia entries exactly matching a query. It will redirect to an entry if one exists. Sometimes it may be necessary to navigate to the disambiguation. alias links="links -no-connect" q=podunk; links "https://en.wikipedia.org/w/index.php?limit=500&profile=all&search=$q&ns0=1" This is another similar search. It will not redirect to an entry if one exists. It will return all entries that have text matches the query. q=podunk; links "https://en.wikipedia.org/w/index.php?title=Special:Search&profile=all&search=$q&fulltext=1" This is a search I use to explore websites. It searches for domains found in external links. For example, one could search with the query "podunk.edu" and find all the links in Wikipedia articles to pages at Podunk University websites. q=podunk.edu; links "https://en.wikipedia.org/w/index.php?title=Special:linkSearch&limit=500&profile=all&target=$q&ns0=1"