How Ttek2 Works
Ttek2 is an automated technology topic hub and search engine. It tracks what is happening across tech — hardware, gaming, AI, Linux, homelab and more — by continuously aggregating public sources and organizing them into topic hubs with live activity signals. This page explains, plainly, where the data comes from, how the signals are computed, and where automated AI is and is not involved.
Who runs it
Ttek2 is built and maintained by Mark Janssen. Questions, corrections, or takedown requests: contact@ttek2.com.
What the data is
Ttek2 does not host original reporting. It aggregates public source material:
- RSS feeds, sitemaps, and web pages from an allowlist of technology publishers, fetched by our own crawler, which honors robots directives and a published crawl policy (see /about/crawler).
- Community discussions from Reddit and Hacker News.
- Reference data from Wikipedia.
Each topic hub links back to the original sources. Ttek2 stores source metadata and short excerpts, not full copies.
How the signals are computed (deterministic)
- Activity score reflects how much recent, relevant source material a topic is attracting — mention volume and recency across tracked sources. It is a relative attention signal, not a quality or endorsement score.
- Momentum (rising / stable / falling / new) compares a topic's recent activity against its own short-term baseline (roughly the last several hours).
- Sources and Signals counts show how many distinct outlets and discussions back a topic — a transparency measure of how well-evidenced it is.
- Last updated reflects when a topic's underlying data last changed, not when you loaded the page.
Search ranking and topic classification are deterministic: full-text relevance (BM25) plus rule-based topic matching. There is no AI in the ranking of search results.
Where AI is used (and its limits)
A large language model (currently OpenAI's gpt-5.4-mini) generates the per-topic Pulse — a short summary, themes, a sentiment read, and a signal-strength label — from the public source material already gathered. These are automated summaries of public sources, not editorial judgments, endorsements, or fact-checks. AI is not used to rank search results or to decide what is true. Summaries can be incomplete or wrong; always follow the linked sources.
Severity / alert labels
Each topic carries a signal-strength label — info, notable, or alert — derived automatically from the strength and convergence of its signals. It indicates how much attention a topic is drawing, not that an event is confirmed, official, or endorsed by Ttek2.
Freshness and updates
Crawling and topic refresh run on an automated schedule, so hubs update through the day. The "last updated" timestamp on each topic reflects the most recent change to that topic's data.
Accuracy, prices, and corrections
Because Ttek2 is automated, summaries, classifications, and any extracted prices are best-effort from public sources and may be inaccurate, out of date, or region-specific. Nothing here is advice. To report an error or request removal, email contact@ttek2.com.