This is unlikely, but one could imagine a browser extension that tries to guess what forecasts are relevant to any news article one might be reading and show that to users.
At first glance, it seems to me like that might not be too hard to create an ok version of, which would be used by at least let’s say 100 people. Do you mean that this being used by (say) millions of people is unlikely?
For what it’s worth, I had the same initial impression as you (that making a browser extension wouldn’t be that hard), but came to think more like Ozzie on reflection.
It’s possible we have different definitions of ok.
I have worked with browser extensions before and found them to be a bit of a pain. You often have to do custom work for Safari, Firefox, and Google Chrome. Browsers change the standards, so you have to maintain them and update them in annoying ways at different times.
Perhaps more important, the process of trying to figure out what text is important text of different webpages, and then finding some semantic similarities to match questions, seems tricky to do well enough to be worthwhile. I can imagine a lot of very hacky approaches that would just be annoying most of the time.
I was thinking of something that would be used by, say, 30 to 300 people who are doing important work.
I’ve never really thought about making browser extensions or trying to automate the process of working out what forecasts will be relevant to a given page, so I wouldn’t be especially surprised if it was indeed quite hard.
But maybe the following approach could be relatively easy:
Have the browser extension only check for things like the tags, categories, or keywords a site explicitly includes in the data for a page, like how academic papers list keywords to make them more findable.
I really don’t know how this stuff works, including things like whether lots of sites use a similar format for this such that it’d be easy for the extension to read it.
If that wouldn’t be easy, then maybe this could start as working only for the handful of sites/​formats where this would be easy, such as—I assume—academic papers
Then have the browser extension simply show top ranked forecasts for the relevant (combinations of) tags, categories, or keywords
Rather than trying to show the subset of those forecasts which are especially relevant to the article at hand
E.g., if I’m reading a paper related to one aspect of the history of the Russia nuclear weapons arsenal, the extension might simply show me forecasts about nuclear weapons in general, or maybe about Russian nuclear weapons in general
Or the extension could show a top ranked forecast for each of the relevant tags, categories, or keywords, with a button under each to allow you to see more forecasts relevant to that keyword
(This is just spitballing from someone who knows very little about how any of this works, so I wouldn’t be surprised if implementing ideas like those would be very hard or not very useful anyway.)
At first glance, it seems to me like that might not be too hard to create an ok version of, which would be used by at least let’s say 100 people. Do you mean that this being used by (say) millions of people is unlikely?
Also, I think somewhat related ideas were proposed and discussed in the post Incentivizing forecasting via social media. (Though I’ve only read the summary.)
For what it’s worth, I had the same initial impression as you (that making a browser extension wouldn’t be that hard), but came to think more like Ozzie on reflection.
It’s possible we have different definitions of ok.
I have worked with browser extensions before and found them to be a bit of a pain. You often have to do custom work for Safari, Firefox, and Google Chrome. Browsers change the standards, so you have to maintain them and update them in annoying ways at different times.
Perhaps more important, the process of trying to figure out what text is important text of different webpages, and then finding some semantic similarities to match questions, seems tricky to do well enough to be worthwhile. I can imagine a lot of very hacky approaches that would just be annoying most of the time.
I was thinking of something that would be used by, say, 30 to 300 people who are doing important work.
I’ve never really thought about making browser extensions or trying to automate the process of working out what forecasts will be relevant to a given page, so I wouldn’t be especially surprised if it was indeed quite hard.
But maybe the following approach could be relatively easy:
Have the browser extension only check for things like the tags, categories, or keywords a site explicitly includes in the data for a page, like how academic papers list keywords to make them more findable.
I really don’t know how this stuff works, including things like whether lots of sites use a similar format for this such that it’d be easy for the extension to read it.
If that wouldn’t be easy, then maybe this could start as working only for the handful of sites/​formats where this would be easy, such as—I assume—academic papers
Then have the browser extension simply show top ranked forecasts for the relevant (combinations of) tags, categories, or keywords
Rather than trying to show the subset of those forecasts which are especially relevant to the article at hand
E.g., if I’m reading a paper related to one aspect of the history of the Russia nuclear weapons arsenal, the extension might simply show me forecasts about nuclear weapons in general, or maybe about Russian nuclear weapons in general
Or the extension could show a top ranked forecast for each of the relevant tags, categories, or keywords, with a button under each to allow you to see more forecasts relevant to that keyword
(This is just spitballing from someone who knows very little about how any of this works, so I wouldn’t be surprised if implementing ideas like those would be very hard or not very useful anyway.)