5 ms·

It seems like nobody at Facebook is allowed to play the long game. In the long game, quality wins. In the long game, an organization that puts quality of infor

It seems like nobody at Facebook is allowed to play the long game. In the long game, quality wins. In the long game, an organization that puts quality of information at top priority persists and becomes a long-lived resource for civilization. Like Google has become a persistent Web search resource, setting aside its short-lived ancillary forays into more clickable, trivial services. So long as Facebook prioritizes viral clickbait, so long will it be in danger of collapse, irrelevance, and extinction. That's ok with me.

rdruxn 9y ago
Except their moves and announcements in recent weeks suggest that they're in fact prioritizing higher quality content e.g. local news (https://www.axios.com/facebook-to-prioritize-local-news-in-news-feed-1517258545-f4b20fdf-6f6d-4511-abec-d5bf334ebca1.html https://www.axios.com/facebook-to-prioritize-local-news-in-n...)
bonniemuffin 9y ago
Identifying "quality" content algorithmically seems like a hard problem to me. If Facebook wanted to play the long game by promoting high quality content in the feed, how would you propose they identify it in order to promote it?
- blacksite_ 9y ago
  if content.is_quality(): content.promote() else: content.demote() Four lines, Facebook...
- ridgeguy 9y ago
  Your question is excellent and gets to the heart of the matter. How could Facebook, in a neutral way, cull crap and promote quality? I wish I had robust answers. All I have at the moment is opinions. • Algorithmic methods can't deliver near term solutions. Algorithms embody the biases of their human coders. I doubt we currently know how to algorithmically evaluate inputs for quality in the Facebook or "fake news" context. We need more time and effort to learn how to do this. • Promoting quality currently requires humans, who are expensive. If Facebook really prioritizes feed quality, we'll see serious hiring. If we don't, then evidence of commitment is lacking. • I believe (no proof to offer, mainly intuition) quality can eventually be algorithmically determined, but it's a hard problem. It will take years, lots of smart people and $ to figure out how to make it work at scale. • I don't think Facebook's announced path, which is to solicit those who are arguably more susceptible to "fake news" to be arbitrators of quality, will succeed. Facebook, like others, is caught between maximizing 90 day quarterly profits and becoming a persistent human resource. I don't envy Zuckerberg at all. I hope he can do the right thing. We'll see.

rdruxn 9y ago

Except their moves and announcements in recent weeks suggest that they're in fact prioritizing higher quality content e.g. local news (https://www.axios.com/facebook-to-prioritize-local-news-in-news-feed-1517258545-f4b20fdf-6f6d-4511-abec-d5bf334ebca1.html https://www.axios.com/facebook-to-prioritize-local-news-in-n...)

bonniemuffin 9y ago

Identifying "quality" content algorithmically seems like a hard problem to me. If Facebook wanted to play the long game by promoting high quality content in the feed, how would you propose they identify it in order to promote it?

blacksite_ 9y ago
if content.is_quality(): content.promote() else: content.demote() Four lines, Facebook...
ridgeguy 9y ago
Your question is excellent and gets to the heart of the matter. How could Facebook, in a neutral way, cull crap and promote quality? I wish I had robust answers. All I have at the moment is opinions. • Algorithmic methods can't deliver near term solutions. Algorithms embody the biases of their human coders. I doubt we currently know how to algorithmically evaluate inputs for quality in the Facebook or "fake news" context. We need more time and effort to learn how to do this. • Promoting quality currently requires humans, who are expensive. If Facebook really prioritizes feed quality, we'll see serious hiring. If we don't, then evidence of commitment is lacking. • I believe (no proof to offer, mainly intuition) quality can eventually be algorithmically determined, but it's a hard problem. It will take years, lots of smart people and $ to figure out how to make it work at scale. • I don't think Facebook's announced path, which is to solicit those who are arguably more susceptible to "fake news" to be arbitrators of quality, will succeed. Facebook, like others, is caught between maximizing 90 day quarterly profits and becoming a persistent human resource. I don't envy Zuckerberg at all. I hope he can do the right thing. We'll see.