dysiode thanks for the detailed explanation. I believe the problem you're addressing is the same discussed in this thread - https://kagifeedback.org/d/4502-opt-in-collaborative-shared-result-ranking ; I'llmerge them unless I misunderstood something (?)
I like this explicit approach of shareable lists. This is a topic we're going to spend some time analysing and defining next steps. We'll keep the main thread updated.
This is a pivotal feature and Bluesky is showing/leading the way. Their approach is a good proof that users are sensible to ranking algorithms & will use them.
This helps Kagi provide more value & personnalisation without associating itself with an idea or another.
Getting the ball rolling on "Custom Lenses SDK" early on will add another dimension to Kagi, I guess a good part of the userbase is tech literate but building a strong dev community & open source ranking algorithms must take a little time !
It's interesting to see the comparison with the Bluesky collaborative ranking tools. This is a critical feature of decentralized social media: block lists, promote lists, etc. Same with adblockers... domain reputation is everything. uBlock Origin is only as good as its block lists. It would be amazing if Kagi to published these kinds of lists.
It would be something like a Private Tracker: it's a database curated by and for a defined, trusted population. In this case, people who care enough about search quality to pay for Kagi. Kagi would need to be very thoughtful about how to receive trusted feedback from users without reducing privacy. Privacy Pass might be a valuable foundation for this.
Lets see if I can state this as clear user stories...
As a Kagi Subscriber, I want Kagi's high-quality signals about publisher quality & relevance available on other platforms. When Kagi thinks a domain is used for spam or other malfeasance, I want my adblocker to block that domain. When Kagi thinks a user on a decentralized social media platform is fake account, spammer, or otherwise fraudulent, I want that user to be automatically blocked from my feeds on that platform.
As a Kagi Subscriber and decentralized social media user, I am willing to share my personal moderation decisions pseudonymously with Kagi in return for access to blocker feeds/moderation lists. I don't want Kagi to build a database of my social media behavior explicitly tied to my account, but I'm comfortable sending a stream of block/unfollow/report events tied to a stable ID that is not otherwise linked to my Kagi account. I expect that Kagi will use this identifying information exclusively to prevent abuse of the reporting service.
As a Kagi Subscriber, I want the ability to search social media at scale, including the long tail of decentralized social media instances. I expect these searches to be ranked based on high-quality signals. Users who share links to low-reputation domains or are frequently blocked by other Kagi users should be downranked.
pbronez I'm not a decentralised social media user but I do observe what goes on on them, and frankly the community blocklists seem to be a total shitfest, if you'll excuse my language. I've seen competing blocklists, rivalry between blocklist maintainers, said warring maintainers adding each other and anyone who reacts with their rival to blocklists, etc.
The amount of people out of Kagi's 40k users who are going to not only have an account on decentralised social media, actively curate a blocklist (e.g. I'm on X and don't block anyone, even accounts I find annoying/distasteful), AND are willing to connect their SM to Kagi I'm willing to bet is less than 100. Remember the 90/9/1 rule?
Simply put, I don't want my search results influenced by a tiny group of users of (not to be offensive) completely unknown biases, community drama, mental stability, etc. Of course one can then say that I don't have to enable this feature, but then why build it? I'm sure many people feel the same way, Kagi is billed as an unbiased, you-first search engine after all. I'd rather see dev time that would be spent on this going towards improving search in other ways.
Feature: The ability for users to flag a search result as AI slop, or clickbait.
Outcome: Improve search result quality.
User Side:
Users could click the three dot menu next to a search result and instead of hiding a result, they could flag a search result as AI slop/Clickbait. Flagged websites would then be hidden from a user's search results in perpetuity.
Function:
To offload workload from Kagi devs, flagged sites could sent to a community review portal. Individual members could review a site and, if a site meets a certain threshold, it gets filtered into admin/dev portal for final processing. An incentive system could be utilized to attract and retain community volunteers, like flair, badges or discounts.
Complications:
The most privacy-focused way that this could work is that just the page is displayed for review. The best way that this could work is to display the flagged site plus the search query, though this could be used for possible user identification.
Notes:
A resolution system should also be added by site owners who wish to dispute an AI Slop/Clickbait label. The dispute process could function the same as the takedown process (Community > Dev > Action)
It would be nice to be able to turn this off, and/or to be able to set exceptions.
This could also be implemented as community filters, like curators on steam but for raising/lowering domains
I'll merge this thread with another on the same subject. We're currently focused on shipping some important projects, but I hope we can share a few ideas on how to accomplish this soon
carl
I really like your idea and agree that these communal rankings could get really messy and political for Kagi. It should stay simple and have features that enhance rather than replace like Kagi's philosophy states. I feel like these communal ranking features end up replacing once search and creates a sphere of influence.
marcel001122 Couldn't agree more with your point and touches upon what I am trying to say in the above paragraph. Foundational features like this are very dangerous and like someone said before in this thread, it creates a pandoras box.
The super lenses idea I believe is the best solution/compromise to this OP's requests, from what I've read.
Enabling the subscription to rankings created by users using the super lenses idea, rather than a communal automatic ranking system, would avoid associating kagi with certain opinions as well as alleviate the risk of the default Kagi search being taken over by bots or certain groups pushing their website cause they managed to get the highest communal ranks.
I thought in the past of doing exactly that for my own search engine (before discovering Kagi), and I quickly realized that the time spent on a website is not proportional to its quality, in many cases it is actually the opposite.
Plus, it would worsen the SEO problem and just make websites optimize for you to spend more time there.
A more manual but less flawed approach would be community rating, literally just rating the results manually, but even that is a very flawed approach, since it could allow mass-downvoting or upvoting potentially instrumented, and also, nobody wants a big popup saying "please rank as a good result if you liked this".
In reality allowing "community" based ranking will probably never work because it will end up being instrumented.
Vlad I would not have a downvote/upvote system for search results. I don't need to do that on other search engines, so why would I need to do it on Kagi?
I think this is a cool idea, but just because I spend a short amount of time on a webpage doesn't necessarily mean I didn't find what I was looking for.
I haven't read through this entire thread, so this may already have been brought up. I'm not sure what would be a good solution to this except for just not opting in to the feature.
Anybody remember Web of Trust (WoT), that got fucked up?
I would like to have something similar here. Users can rate a search result with some metrics (content, advertising, … or something like this) and optionally add a comment.
When there are at least X reviews for a domain, the shield icon for all search results to this domain (the one on the right side of results) gets colored according to the reviews from red to green.
When there are X reviews for a specific result, there could be some icon in the shield (a checkmark?) and the review results for this specific page can be shown as color.
With this, we won’t interfere with the ranking, but users can decide to not click on a result, check the detailed ratings and rank the domain for themselves.
Why domain and page specific? I think there are sites where some content is great, some is bad. For example: I searched for a free video edition software a while ago. I found a list that compared different options and this list was really helpful. But the list was on the website of one of these options that isn’t really free, also other content on this site wasn’t that great. So I would rate this specific list up, but other parts of the website down.
Another example could be help pages of companies or Reddit. Not quality of content differs from topic to topic. So the overall score is based on all reviews as it reflects the overall quality of the website. But specific pages can be great or bad.
The idea tempts me. However, I think it's a dangerous temptation. I don't like the idea of that data being collected or tracked, even if opt-in. There's the possibility likelihood of the pool being soured by bad actors, which I don't think is avoidable (RoxyRoxyRoxy makes a good point).
Something like this should be in the realm of a third-party group/extension instead. Kagi should rather focus on other features.
I'm of the opinion that implementing this would deteriorate Kagi.
These suggestions strike me as not only counter to the commitment of minimizing user data collection, but also as solutions in search of a problem. It is not convincing to me that integrating other users' preferences and behaviors into my search will result in better results. Tracking time spent on results is invasive and involves either solving or ignoring the edge cases mentioned earlier, and other users' votes on results are just not relevant to me, nor would I care to involve them at all even if they were.
The discussion on this thread has convinced me to remove my upvote. I wish we could leverage the experience of other users more directly to improve results, but on balance it seems unlikely to work out that well. And even if Kagi could find a way, it would likely require a disproportionate amount of effort compared to the value added.
I do still believe that dismissing the idea entirely is unnecessarily cynical. Kagi is better positioned to make this work than others. But it isn’t easy or sure and focus is needed at this stage for certain!
