Skip to content

Internet research - Category Home

Internet research - Topics

Trial access to a specialized database is long enough to build a workflow you cannot keep Getting institutional library access from a laptop at home Oral history projects like the Internet History Program at CHM recorded engineers decades after the events and memory drift is a known problem Fred Turner's From Counterculture to Cyberculture argues the internet's culture came from the New Communalists rather than the military Studying virality without amplifying the post: the ethics of misinformation research Sampling bias worsens when a platform API only returns the most engaged accounts The bitcoin and blockchain histories already show the same hagiography problems the internet history field has long fought Search operators most people never use: site: and filetype: and exact phrases Measuring which sites are blocked from inside a country without getting caught Trend reports built from venture funding data miss whatever is popular without money behind it Aaron Swartz pushed for open access to research and his prosecution became a turning point for the open access movement Fake reviews and astroturfed accounts poison any study that trusts platform metrics Whose archive counts as evidence - corporate papers stay closed while hobbyist mailing list archives remain open Google Trends spikes for a keyword last weeks while the underlying behavior lasts years Why the same citation resolves to three different metadata records across bibliographic services Finding primary sources instead of the articles that summarize them Amazon Mechanical Turk samples skew young and educated in ways that quietly bias results Open access mandates changed who pays for publishing without changing who reads Query interfaces that cap results at a few hundred rows hide whether a gap is real The outline question is really a knowledge organisation question and library classification schemes answer it differently Where Wizards Stay Up Late was written by journalists and mixes archival material with interviews that historians treat with caution Janet Abbate's Inventing the Internet is the standard scholarly correction to popular origin stories Building a personal archive of sources because links rot within a few years Self-selected online polls predict elections badly and keep getting cited anyway Link rot destroys a large share of the citations in web-based studies within a decade Participants recruited through Facebook groups represent that group and little else Pre-registering an online experiment does not stop researchers from mining subgroups afterward A paywalled statistics database bills per download and that number changes how research gets done Scraping public posts is legal in some countries and a privacy violation in others The claim that people check their phones 96 times a day has weak evidence behind it Screen time dashboards count a video playing in a background tab as active attention Wayback snapshots are not neutral copies - archives get edited and pages vanish