is the selling point that they just give you access to a unified access in where you get to write each request toward the particular api provider's api spec? if so why not just sign up with the vendor directly?

I have my own infra setup and started with 25k seed domains including major known startups, YC companies and things like that. I collected some of the data via scraping and then merged it with 3rd party enrichment sources. Once the crawling per domain is finished, it is put on a 30 day schedule to repeat the crawl just to keep things fresh.

If you are on desktop and click on a result entry, you'll see expanded data for each domain. In reality that's just a small portion of the total data I collect.

Anyway, really looking forward to hearing your thoughts!

It's mostly for navigational search. Say you're looking for a tool that does X, this engine will match your query against all the domains in the database to find you the closest matching domain that solves for that pain point you have.

It's like a discovery engine to help you find tools, products, services, or domains that cater to (or market themselves as a solution for) the problem you have.

Curious to hear if you have any feedback on the concept as is or if you can see any potential changes that would make it work better?

I first thought about this approach back in 2011 when I was in college. But never did anything with the idea because I did not have the resources at the time to build something like this.

As of late however, it has become rather clear that the current distribution channels for businesses, both organic and paid, are getting squeezed.

- Google doesnt send you traffic anymore

- Search results are infested with SEO optimized content that no one can compete with

- LLMs cite the same garbage

- Distribution via advertising is getting more expensive each day

- Social channel are all about clickbait and maximize following which encourages quantity over quality

It's clear that the relationship between product-sellers and potential-buyers is completely out of whack today. Most of the time they can't even find each other.

That's why I decided to crawl a small index of 25k websites and try this out.

The idea is to profile a domain as a whole (single entry) and match the user's query to find the best fitting product/service that solves their pain point. This ignores any SEO optimized content, spam, black hat garbage, and instead focuses on finding the best match making the distribution playing field more equal.

Is this a good idea? I still don't know. That's why I'm looking for feedback. I could easily improve the search result quality by like 5-10x but I'm trying to ascertain first if this approach is even helpful to people.

Curious to hear if you have any feedback for me?

I recently launched this experimental search engine and would love the community's feedback!

Traditional search engines index every document (webpage) and then match your query to their database of indexed pages.

This has worked out great over the last 30 years or so as this process is pretty good for informational search. However, as of late, SEO-optimized content is dominating search results. Even LLM citations (I looked at over 1M of them) are dominated by said SEO-optimized content.

So I decided to try something new: Instead of having each webpage get its own index entry, the domain as a whole gets a single entry.

The advantages: SEO-optimized content no longer dominates the results and smaller/better fit solutions tend to find their way to the top of the search results.

The disadvantanges: This type of search engine does not work well for informational search intent.

What I'm looking for right now is a little bit of feedback to know if this approach has any merit. I'd appreciate it immensely if you would spare a moment to run a few queries and tell what you think!

Link to site: https://galefox.com/

Comments URL: https://news.ycombinator.com/item?id=49167063

Points: 4

# Comments: 6

Last month however I decided to go back to the idea and give it a shot. Right now I'm in the process of scraping and building a huge index. The technical challenges have been plenty. But I should be ready to publish an alpha version by end of month or so.

Back in the day when I was like 15 and DC++ was still a thing, I used to browse people's shared folders. One day I came across a file called "the paradox of false positive". It was a 1 pager that described how a machine which is 99.9% accurate at identifying terrorists would be completely useless due to this false positive base rate fallacy you're describing.

It really stuck with me throughout the years. It's kind o remarkable how even a 99.9% accurate heuristic is insufficient at scale.

Which begs the question: lets assume the intentions are pure (which we know they're not but lets be generous), what other options are there when 99.9% heuristic is not good enough? how do you design systems when they're guaranteed to fail as they scale up?

edit: and what do you know, I just saw this as I scrolled down on HN https://news.ycombinator.com/item?id=48816959

else echo "You're full of shit!";

// output

> You're full of shit!

Today's PHP is better than it has ever been. Are there some things that are rough around the edges? yes ofc. But there is no language that doesn't have that. It's all trade-offs.

Last week I switched from Nginx+FPM to FrankenPHP and my god even the deployment experience got 10x better.

Safe to say that if you haven't tried the language, give it a shot. Within a few days you'll know if it's a good fit for you or not!

You'd think the founders would Google for naming conflict before choosing a name.

I'm dying to see S-1 filing for Anthropic or OpenAI. I don't actually think inference is as cheap as people say if you consider the total cost (hardware, energy, capex, etc)

Most people in the BiP these days barely know how to deploy a database or host something using nginx. it's all vercel, supabase, aws, clerk, yada yada. Cost aside, I think that people are addicted to complexity.

Comments URL: https://news.ycombinator.com/item?id=47605636

Points: 2

# Comments: 0

Also a lot of this "money" is in cloud compute and credits not cash so...