Open Portal

Free internet search engine built completely from scratch.

  • 4Bpages
  • 33Mdomains
  • 150languages

Free access to the information is a basic human right. In practice, most of what people find on the internet today passes through a few search engines owned by a few corporations. The goal of Open Portal is to challenge the status quo and provide an alternative search engine.

What makes Open Portal different

  • Free.

    Open Portal is free to use, with no account required.

  • Independent.

    Our search engine built from scratch and it serves data from our own index.

  • Private.

    We don't collect your data for advertising.

FAQ

Where does the data come from?

The index is built from the Common Crawl corpus, a free and widely used repository of web crawl data. While crawling is a fundamental part of a search engine, we decided to focus on the data processing pipeline and search quality first.

What is the status of the project?

What you see is a proof of concept. The index is not yet updated regularly, and search quality is still a work in progress.

How does the search work?

The engine uses tantivy, a full-text search library written in Rust. For each query, the top 1,000 candidates are retrieved using a text relevance signal (BM25) and then reranked with a host importance signal (PageRank), following Craswell et al. (2005). PageRank values come from the Common Crawl web graph.

Why can't I find my website?

The index only contains pages that are in the Common Crawl corpus. Common Crawl doesn't cover the whole web, so a number of sites and pages are missing.

Why are the results different from Google?

Currently, our ranking is based on only two signals: text relevance and host importance. For comparison, modern search engines also incorporate semantic search and rely on years of user data. Expect search quality improve over time.

What data do you collect?

There are no accounts, ads or third-party trackers. The web server keeps standard access logs (IP address, query, time) to monitor traffic and performance.

Is there an API?

Not yet. If you need one, let us know what you'd use it for.

Do you have any plans to monetize the project?

There are currently no plans to monetize the project. To cover hardware costs, we are actively looking for sponsors.

Acknowledgements

Open Portal would not be possible without Common Crawl, which provides data for the index.

Contact

Hi, I'm Evgeny. I run Open Portal entirely on my own. If you have an idea, or just want to say hi, reach out to evgeny@openportal.org.