5 ms·

This is a pretty innovative feature. Here's the source code for a sample "Hacker News" Goggle. Essentially, it will prioritize domains popular on Hacker News.

This is a pretty innovative feature. Here's the source code for a sample "Hacker News" Goggle. Essentially, it will prioritize domains popular on Hacker News. I could even see a browser plugin that lets you add or remove domains as you visit them. https://github.com/brave/goggles-quickstart/blob/main/goggles/hacker_news.goggle https://github.com/brave/goggles-quickstart/blob/main/goggle... And here's the language syntax. https://github.com/brave/goggles-quickstart/blob/main/goggles/quickstart.goggle https://github.com/brave/goggles-quickstart/blob/main/goggle...

lostlogin 4y ago
Assuming I’m understanding Goggles correctly, I’d like to be able to have more than one search bar. An unfiltered, and a couple of custom ones.
- benjaminjackman 4y ago
  maybe combine them with user defined !bang commands like on duckduckgo, !hn search_term (searches hn sites list from above).
- sp332 4y ago
  You can use keywords to select search engines in both Chrome and Firefox. If you configure "custom1" as a keyword, then just type custom1 followed by a space into the address bar it will activate that search engine and send the rest of your query to that URL.
  - ghostly_s 4y ago
    This wouldn't let you use multiple "flavors" of Brave though unless they expose different URLs for 'Goggles'; sounds like you just are allowed to enable one as a sitewide setting.
    - sp332 4y ago
      They do have different URLs for different goggles. Do a search for the word "test" with a Goggle on, then grab the URL from the address bar and paste that into your new bookmark. Replace "test" with "%s", add a keyword, and you're good to go. Repeat with other goggles.
irthomasthomas 4y ago
And I was wondering how long it would take for the hn version to arrive. Very cool.
lowwave 4y ago
> sample "Hacker News" Goggle wonder how they came up with that list? That list it self is a goldmine on it's own.
- culi 4y ago
  Just run a web scraper to keep tally of domains posted to HN and their upvotes?
- Barrin92 4y ago
  you can search pretty much all of HN here, that's probably a pretty good start for a list. https://hn.algolia.com/ https://hn.algolia.com/
- vorticalbox 4y ago
  Maybe used api[0] to get The top stories and then just parsed all the domains. [0] https://github.com/HackerNews/API https://github.com/HackerNews/API
mintplant 4y ago
The syntax makes me wonder if the implementation is related to their high-performance filter rule evaluation library, adblock-rust [0]. Whatever is doing the dynamic reranking would need to evaluate lots of rules and churn through links fast. [0] https://github.com/brave/adblock-rust https://github.com/brave/adblock-rust
nico 4y ago
This sounds like a killer feature. Would be great to complement with social media as we’ll. For example, make the domains in posts within my networks, more relevant. That way, through controlling/curating my networks, I get to decide what’s more relevant to me in search results. Would be awesome to add stackoverflow in there too :)
- sassy_quat 4y ago
  Man, with tech evangelists like you, Brave is going receive an anti-trust case for making a killer app.
- SmellTheGlove 4y ago
  > Would be great to complement with social media as we’ll. For example, make the domains in posts within my networks, more relevant. The cynic in me can’t help but wonder if this is just dialing up an echo chamber if we include social media proximity as a filter.
  - s3tz 4y ago
    Would help to have an easy toggle between the different ones.
  - jmathai 4y ago
    Promoting ways to do curate your own echo chamber beats other companies doing it for you based on ad revenue.
  - keenmaster 4y ago
    Our brains are not well-equipped to down-rank bad information sufficiently once we are exposed to it. By then, it’s often too late. We do have powers of planning and selective exposure though, so we can say “I think these newspapers are authoritative and these others aren’t, so that’s what I’m going to subscribe to” much easier than “I’m going to selectively ignore this trash information that I see in google results or my Facebook wall.” Tools like this will allow reasonable people to curate the most authoritative sources on their side of the ideological spectrum. So even a right-leaning lens will tend to exclude the most extreme sources, and this will ease pressure on Fox such that it can present more center-right content without losing views. Same with the left-leaning lenses and MSNBC. If this idea proliferates (of selective exposure in a digital environment, where algorithms don’t entirely choose what you see), it can help deflate the network effects of extremism and populism in a virtuous cycle.
danskeren 4y ago
This is one of the first things I did when I implemented my own Bing proxy, although I did the site boosting/downranking/exclusions and rewrote URLs (e.g. www.reddit.com > old.reddit.com) myself since I didn't have any users, but the plans was always to give users the ability to control this. I later learned that it's actually against Microsoft's Bing Search API use and display requirements (https://docs.microsoft.com/en-us/bing/search-apis/bing-web-search/use-display-requirements#restrictions https://docs.microsoft.com/en-us/bing/search-apis/bing-web-s...), which states that you're not allowed to: > Modify the results content (other than to reformat them in a way that does not violate any other requirement), unless required by law or agreed to by Microsoft. > Reorder, including by omission, the results displayed in an answer when an order or ranking is provided, unless required by law or agreed to by Microsoft. > Display content that was not included within any part of a response in a way that would lead a user to believe that content is part of the response. And there are many other restrictions imposed upon you. I'd argue the reason you see almost zero innovation in this space (the only thing I can think of is !bangs) is because non-Google/Bing(/Brave) Search Engines have very little control over the received data, and with the little control we do have then there are many restrictions that prevents us from making these 'innovative' features. Based on my pessimistic reading then we're not allowed under any circumstances to merge API results from multiple search engines (e.g. Bing + Yandex + Google), and for every site we want to exclude/boost/derank/rewrite sites/urls then we need explicit permission from a Microsoft representative (which would prevent us from implementing this kind of feature). We're also not allowed to display third-party ads on any pages that displays part of the API results (which is funny because I feel it's near impossible to join Microsoft's Advertising network even though my site in my opinion have a good purpose and provide a high quality UI/UX experience - https://ask.moe https://ask.moe).. I've later switched to Google's Programmable Search Engine and even though my site has been approved by Google as a non-profit, as well as approved for AdSense, then they will only display ads in the search results if I turn off revenue-share. This industry desperately needs legislation that makes it possible for everybody to obtain access to Google/Microsoft's crawled data, and this access should not be limited to expensive APIs that make it impossible for anybody besides Google/Microsoft to do any sort of innovation.
- nico 4y ago
niutech 4y ago
This is not so innovative, you've been able to do so for many years using Google Custom Search (now Programmable Search Engine) - e.g. many years ago I made an open source projects search: https://cse.google.com/cse?cx=008818486088109706548:xnogqdiabr0 https://cse.google.com/cse?cx=008818486088109706548:xnogqdia...
- dubswithus 4y ago
  It looks like is targeted towards website owners and not the general public. https://developers.google.com/custom-search https://developers.google.com/custom-search

lostlogin 4y ago

Assuming I’m understanding Goggles correctly, I’d like to be able to have more than one search bar. An unfiltered, and a couple of custom ones.

benjaminjackman 4y ago
maybe combine them with user defined !bang commands like on duckduckgo, !hn search_term (searches hn sites list from above).
sp332 4y ago
You can use keywords to select search engines in both Chrome and Firefox. If you configure "custom1" as a keyword, then just type custom1 followed by a space into the address bar it will activate that search engine and send the rest of your query to that URL.
- ghostly_s 4y ago
  This wouldn't let you use multiple "flavors" of Brave though unless they expose different URLs for 'Goggles'; sounds like you just are allowed to enable one as a sitewide setting.
  - sp332 4y ago
    They do have different URLs for different goggles. Do a search for the word "test" with a Goggle on, then grab the URL from the address bar and paste that into your new bookmark. Replace "test" with "%s", add a keyword, and you're good to go. Repeat with other goggles.

irthomasthomas 4y ago

And I was wondering how long it would take for the hn version to arrive. Very cool.

lowwave 4y ago

> sample "Hacker News" Goggle wonder how they came up with that list? That list it self is a goldmine on it's own.

culi 4y ago
Just run a web scraper to keep tally of domains posted to HN and their upvotes?
Barrin92 4y ago
you can search pretty much all of HN here, that's probably a pretty good start for a list. https://hn.algolia.com/ https://hn.algolia.com/
vorticalbox 4y ago
Maybe used api[0] to get The top stories and then just parsed all the domains. [0] https://github.com/HackerNews/API https://github.com/HackerNews/API

mintplant 4y ago

The syntax makes me wonder if the implementation is related to their high-performance filter rule evaluation library, adblock-rust [0]. Whatever is doing the dynamic reranking would need to evaluate lots of rules and churn through links fast. [0] https://github.com/brave/adblock-rust https://github.com/brave/adblock-rust

nico 4y ago

This sounds like a killer feature. Would be great to complement with social media as we’ll. For example, make the domains in posts within my networks, more relevant. That way, through controlling/curating my networks, I get to decide what’s more relevant to me in search results. Would be awesome to add stackoverflow in there too :)

sassy_quat 4y ago
Man, with tech evangelists like you, Brave is going receive an anti-trust case for making a killer app.
SmellTheGlove 4y ago
> Would be great to complement with social media as we’ll. For example, make the domains in posts within my networks, more relevant. The cynic in me can’t help but wonder if this is just dialing up an echo chamber if we include social media proximity as a filter.
- s3tz 4y ago
  Would help to have an easy toggle between the different ones.
- jmathai 4y ago
  Promoting ways to do curate your own echo chamber beats other companies doing it for you based on ad revenue.
- keenmaster 4y ago
  Our brains are not well-equipped to down-rank bad information sufficiently once we are exposed to it. By then, it’s often too late. We do have powers of planning and selective exposure though, so we can say “I think these newspapers are authoritative and these others aren’t, so that’s what I’m going to subscribe to” much easier than “I’m going to selectively ignore this trash information that I see in google results or my Facebook wall.” Tools like this will allow reasonable people to curate the most authoritative sources on their side of the ideological spectrum. So even a right-leaning lens will tend to exclude the most extreme sources, and this will ease pressure on Fox such that it can present more center-right content without losing views. Same with the left-leaning lenses and MSNBC. If this idea proliferates (of selective exposure in a digital environment, where algorithms don’t entirely choose what you see), it can help deflate the network effects of extremism and populism in a virtuous cycle.

danskeren 4y ago

This is one of the first things I did when I implemented my own Bing proxy, although I did the site boosting/downranking/exclusions and rewrote URLs (e.g. www.reddit.com > old.reddit.com) myself since I didn't have any users, but the plans was always to give users the ability to control this. I later learned that it's actually against Microsoft's Bing Search API use and display requirements (https://docs.microsoft.com/en-us/bing/search-apis/bing-web-search/use-display-requirements#restrictions https://docs.microsoft.com/en-us/bing/search-apis/bing-web-s...), which states that you're not allowed to: > Modify the results content (other than to reformat them in a way that does not violate any other requirement), unless required by law or agreed to by Microsoft. > Reorder, including by omission, the results displayed in an answer when an order or ranking is provided, unless required by law or agreed to by Microsoft. > Display content that was not included within any part of a response in a way that would lead a user to believe that content is part of the response. And there are many other restrictions imposed upon you. I'd argue the reason you see almost zero innovation in this space (the only thing I can think of is !bangs) is because non-Google/Bing(/Brave) Search Engines have very little control over the received data, and with the little control we do have then there are many restrictions that prevents us from making these 'innovative' features. Based on my pessimistic reading then we're not allowed under any circumstances to merge API results from multiple search engines (e.g. Bing + Yandex + Google), and for every site we want to exclude/boost/derank/rewrite sites/urls then we need explicit permission from a Microsoft representative (which would prevent us from implementing this kind of feature). We're also not allowed to display third-party ads on any pages that displays part of the API results (which is funny because I feel it's near impossible to join Microsoft's Advertising network even though my site in my opinion have a good purpose and provide a high quality UI/UX experience - https://ask.moe https://ask.moe).. I've later switched to Google's Programmable Search Engine and even though my site has been approved by Google as a non-profit, as well as approved for AdSense, then they will only display ads in the search results if I turn off revenue-share. This industry desperately needs legislation that makes it possible for everybody to obtain access to Google/Microsoft's crawled data, and this access should not be limited to expensive APIs that make it impossible for anybody besides Google/Microsoft to do any sort of innovation.

nico 4y ago

niutech 4y ago

This is not so innovative, you've been able to do so for many years using Google Custom Search (now Programmable Search Engine) - e.g. many years ago I made an open source projects search: https://cse.google.com/cse?cx=008818486088109706548:xnogqdiabr0 https://cse.google.com/cse?cx=008818486088109706548:xnogqdia...

dubswithus 4y ago
It looks like is targeted towards website owners and not the general public. https://developers.google.com/custom-search https://developers.google.com/custom-search