The subject is technology and the industry built on it: the
companies, the products, the research and the theories, the rules made
about all of it, and the working lives of the people inside it.
A great deal of what is published about technology is paid for by
somebody with something to sell into it: a product, a platform, a
course, a forecast. That is not a conspiracy and it is not a measurement
either; it is an incentive, and it shapes what gets said and what never
gets checked.
So the work here is done the other way round. Every story is traced
to the document it comes from, not to another outlet’s account of it.
Every source carries a grade for how much weight it will take and the
date it was read. And every story says what its evidence does not tell
you.
The standard
Sources are sorted into five grades. A grade 1 source is a body with
the power to compel an answer and nothing to sell you: a regulator, a
court, a statistics office. A grade 5 source is somebody describing the
thing they sell, or themselves.
The grade travels with the source, beside the story, not in a note at
the back, and it decides what the source may be used for: a stated
amount or fact at the top of the scale, a direction only in the middle,
and a claim to be quoted as a claim at the bottom.
The standard is set out in full
here, including what happens when nothing meets it.
What is published
News: what changed in
technology each day, from the document it changed in: the filing, the
ruling, the release, the paper. Every story shows its sources and their
grades, and a dated correction sits above anything that turns out to be
wrong. The brief is the day’s news in a minute,
and the newsletter sums up each week.
Reports: one question at a
time, with the evidence gathered for it and a plain statement of what
could not be established. The first of them are the
findings: single questions settled in public, with the working
shown. Some are explainers: how a number you meet in the news is made,
and what it counts. There is no opinion section, because there is no
opinion here.
The research: the evidence
the reports rest on, one project at a time. The first, Computing Careers, is on
computing work and the labour market around it, and produced a book.
The Fine Print: for
members, every Friday, the week’s most repeated technology claims, each
checked against the document it came from. It is the one thing here that
only members get.
The datasets: what the
research produced, free under CC BY 4.0, with the grade of every source
beside it.
What this is not
It is not a vendor, an analyst firm, a recruiter or a consultancy,
and it sells nothing whose value depends on the news coming out a
particular way.
Nobody commissioned or sponsored any of this. There are no ads, no
affiliate links, no referral codes and no paid placements. This site
runs no analytics, and the only cookie it sets is the one you ask for by
signing in, which is why there is no cookie banner. The privacy section says exactly what is kept, and it
is short.
It is not neutral about method, which is the one thing worth being
partisan about. A claim you cannot check is worth nothing, whoever is
making it, and that applies to everything published here.
How to argue with it
Every story shows its sources and their grades, so the reasoning is
what you disagree with rather than a conclusion you are asked to
trust.
If something here is wrong, send it. That
has already happened, on a headline growth projection from the United
States Bureau of Labor Statistics: a reader noticed that two figures
printed on the same page could not both be true, and was right. The page
was not taken down. The
correction sits at the address that carried the error, written above
the figure it replaced.