3 ms·
Think there's a bit chicken and egg with that one. Back in the day Google relied on DMOZ (and perhaps relies on Wiki a little nowadays, WRT knowledge entities).
by ricardo81 6y ago
Think there's a bit chicken and egg with that one. Back in the day Google relied on DMOZ (and perhaps relies on Wiki a little nowadays, WRT knowledge entities). Search (and social) rely on signals to indicate that something is of relevance... you're sort of implying that the monopolies and their single perspectives on how information should be ordered is satisfactory for us right now.
Sometimes there's 10's, 100's or 1000's of pages/sites that are relevant to what we are searching for. All I'd say is that if we diversified the way in which we search/discover new information a little bit more, there's less reliance on a platform or a single algorithm- to me that sounds healthier.
It is totally worth remembering that as much as the likes of Google and FB control a huge share of browsing time and clicks, their algos don't make them the arbiters of truth and information discovery. Sounds quite daft to say it but that is pretty much what the reality is of the modern web- if the portals don't want to rank you, you aren't seen.
- wpietri 6y agoI am definitely not implying that. Twitter lets anybody publish small things. And it lets anybody select a set of accounts and get a time-ordered list of what they've published. That's how I use it. I follow a couple thousand interesting people, ones who tend to point to other interesting people and articles. If at any point I'm bored, I can drink from a firehose of good content. I also maintain a feed reader, which is how I follow people who have their own sites. But I don't need to use those sites for discovery, the pattern you describe in the old days. Discovery just isn't a problem I have much. These days, my problem is filtering down to what I actually have time to read.
- tacocataco 6y agoSo something like stumbleupon with a search bar to search for certain topics?