Skip to content
    All resources

    From a website to reusable knowledge

    A practical introduction to website mirrors, source versions and retrieval with Nexus.

    Start with a question

    Choose a task before collecting a website. For example: which products does an organisation offer, and which pages support that description? A narrow question makes it easier to decide what to collect and whether the result is useful.

    In Nexus, work within the intended organisation and add the website you want to examine. Check the domain and any subsite boundary. Documentation, blogs and product applications may live on separate domains and need their own sources. Nexus access is currently restricted to approved accounts.

    Capture, then inspect

    A Mirror records a capture of a website. Keep the capture date and original page addresses alongside the saved material so a later reader can identify which version was used.

    Inspect the capture status before drawing conclusions. A saved record can be an error response, a resource or an incomplete page. A completed capture does not guarantee that every relevant page was reachable. Spot-check important pages against the original source and record gaps.

    Retrieve with evidence

    When a source is searchable, use a concrete question or a few distinctive terms. Inspect the returned passage in its surrounding page, then retain the source address and version with any finding.

    Capture and search indexing are separate stages. Check search availability rather than assuming that a running or completed capture is immediately searchable. Retrieved text is evidence to evaluate; it does not by itself establish that an interpretation is correct.

    Make the result reusable

    A useful research note contains the finding, the supporting passage, its page address, the capture date and any uncertainty. When the source changes, compare versions before updating the finding.

    For application access, Nexus provides organisation API documentation and a request playground in the workspace. Public API publishing is a separate owner-controlled action. Review the intended audience and included sources before enabling it.