Choose channels and dates
Set one source collection and publication interval so domain counts answer a reproducible comparison question.
Workflow · Domain comparison
Compare external domains appearing in selected Telegram channels with logical-post counts, first and last dates, source breakdowns, and evidence links.
Research fit
Researchers examining information sources, recurring websites, or promotion patterns who need deterministic domain counts and post-level evidence rather than URL anecdotes or inferred commercial relationships.
Set one source collection and publication interval so domain counts answer a reproducible comparison question.
Use extracted web links with normalized domains and distinct logical-post counting so repeated URLs inside an album do not inflate mentions.
Order domains by the number of posts containing them and keep first-seen and last-seen timestamps.
Review which selected channels contributed the most distinct posts for each domain.
Read cited posts and surrounding language before describing promotion, endorsement, partnership, or campaign behavior.
Practical module
The comparison separates deterministic link frequency from the human judgment required to interpret why a website was shared.
Analysis contract
Postgres groups normalized MessageLink domains, counts distinct logical posts, finds earliest and latest publication timestamps, and calculates per-source contributions within the selected scope.
Selected indexed Telegram channels and optional publication dates. The calculation covers extracted MessageEntity URL and text-URL links, not every textual domain reference.
A deterministic link-analysis recipe with frequency, timing, source contribution, evidence, and interpretation guardrails.
For [selected channels] from [start] to [end], rank external domains by distinct logical posts. Show total post mentions, first and last seen, top contributing channels with counts, and a representative cited post. Treat sharing as observed behavior, not proof of endorsement, ownership, traffic, or coordination.
COUNT(DISTINCT logical_post) grouped by normalized domain, plus MIN/MAX publication time and per-source distinct-post counts.
# Cross-channel domain comparison recipe A deterministic link-analysis recipe with frequency, timing, source contribution, evidence, and interpretation guardrails. ## Calculation - Mode: domain_stats - Formula: COUNT(DISTINCT logical_post) grouped by normalized domain, plus MIN/MAX publication time and per-source distinct-post counts. ## Query For [selected channels] from [start] to [end], rank external domains by distinct logical posts. Show total post mentions, first and last seen, top contributing channels with counts, and a representative cited post. Treat sharing as observed behavior, not proof of endorsement, ownership, traffic, or coordination. ## Inputs - **Sources:** Selected channels whose extracted message links are eligible. - **Period:** Publication dates and timezone applied to every source. - **Counting unit:** Distinct logical post per normalized domain. - **Review question:** The market, narrative, or sourcing decision the link pattern will inform. ## Expected output - Ranked normalized domains - Distinct-post mention counts - First and last appearance dates - Top source contribution - Representative evidence links ## Coverage checks - [ ] Do not interpret counts as clicks - [ ] Do not infer endorsement from a link - [ ] Deduplicate albums - [ ] State extraction and corpus boundaries - [ ] Open representative posts before labeling promotion
Boundaries
Founding research pilot
Bring one recurring research job and a small set of public sources. We will assess fit personally before offering a pilot.
30 founding researcher places · rolling admission · personal reply within 2 business days
Analytics cluster
Continue researching