How to Fact-Check a Tech History Claim Before Sharing It

Tech history is full of widely repeated claims that don’t hold up under direct verification — several examples are covered directly on this blog. This is the practical checklist for checking a claim yourself before repeating it.

Step 1: identify the claim’s specific, checkable components

Break a claim into its checkable parts: a date, a named person, a specific quote, a causal relationship. “X invented Y in year Z” contains at least three separate things to verify independently — the identity of the actual inventor, the specific date, and whether “invented” is even the accurate verb for what actually happened.

Step 2: search for the claim’s earliest known source

Search the specific quote or claim text, sorted by date
  where the search tool allows it, to find the earliest
  known appearance

If a widely-repeated quote’s earliest findable appearance is itself an unsourced, uncited later retelling — as with the Bill Gates 640K quote, whose earliest print appearance cited no source at all — that’s a significant red flag about the claim’s actual reliability.

Step 3: check whether the person or organization involved has directly addressed the claim

Search "[name] denies" or "[name] response" alongside
  the specific claim

Direct denials or corrections from the actual person or organization involved — as with Gates’s own repeated, on-record denials of the 640K quote — are strong, directly relevant evidence worth actively searching for, not just relying on the popular version.

Step 4: distinguish a primary source from a retelling of a retelling

Is this a court record / SEC filing / original press release,
  or a blog post citing a book citing a documentary
  citing an interview?

Following a claim’s citation chain back as far as it goes often reveals several layers of retelling between the current popular version and any actual primary source — each layer is an opportunity for embellishment or error to creep in.

Step 5: check for a simpler, more mundane explanation before accepting a dramatic one

Dramatic, tidy narratives (“one bad game caused an entire industry’s collapse,” as with the 1983 crash’s popular E.T.-centric version) are more memorable and more shareable than the messier, multi-causal reality — when a historical claim sounds unusually clean and dramatic, that’s itself a reason for extra scrutiny, not less.

Step 6: check whether “invented” is actually the right word

Does the claim describe a genuine first invention, or a
  popularization, commercialization, or specific
  refinement of something that already existed?

The Napster and Apple/Altair cases are both examples where “invented” was the popularly used word for something more accurately described as “popularized” or “made more accessible” — a real, meaningful, but different claim than being genuinely first.

Step 7: be willing to represent genuine, unresolved nuance

Not every question has a clean, single-sentence answer — the “who invented email” question genuinely requires distinguishing between different, both-real technical achievements rather than picking one oversimplified answer, and an honest fact-check sometimes concludes “it’s more complicated than the popular version, here’s why” rather than offering a replacement one-liner.

Step 8: use the specialized verification resources that already exist

Much of this work has been systematized by people who do it professionally. For quotations specifically, Quote Investigator publishes traceable research into famous attributions — including several canonical tech misquotes — showing the earliest located appearance and how the attribution mutated. Google Books date-restricted search can establish when a phrase first appears in print, which is devastating to quotes supposedly uttered decades before their earliest occurrence. The Computer History Museum’s oral history collection preserves hundreds of long-form firsthand interviews with the actual participants — invaluable, with the standing caveat that an oral history is a memory recorded decades later, strong evidence for what someone believes and recalls, weaker evidence for dates and sequences than contemporaneous documents. And Wikipedia is genuinely useful here when used correctly: as an annotated index of citations to follow and check, never as the terminal source itself.

Step 9: watch for citogenesis in particular

One failure mode deserves its own name, and has one: citogenesis, circular sourcing through Wikipedia. An uncited claim gets added to a Wikipedia article; a journalist on deadline repeats it; the resulting news story then gets added to the article as its citation, and the claim is now “sourced” — by itself. Documented cases of exactly this loop exist, some surviving for years. The defense is mechanical rather than clever: when a citation looks like it might post-date the claim it supports, check the Wikipedia article’s edit history and compare dates. If the “source” was published after the claim appeared in the article, you may be looking at the loop — and at that point the claim has zero independent support until you find something that predates it.

Why this discipline matters for tech history specifically

Tech history sits at an unusual intersection of engineering fact and popular mythology, repeated constantly in casual conversation, marketing copy, and conference talks without much scrutiny — a discipline of actually checking claims against primary sources, rather than repeating whatever version is most commonly heard, is what separates accurate tech history from folklore that happens to involve computers.

Related:

Sources: