Berian Reed

What I do

I run BR Digital, a technical SEO consultancy, and have done since 2016. The work is mostly the unglamorous end of the discipline: crawl budgets, index bloat, structured data, log files, and working out why a page that should rank does not. Clients have included Auto Trader, MoneySuperMarket, Matchesfashion and Megabus, which is to say large sites with correspondingly large ways to go wrong.

I have taught the in-house SEO masterclass at BrightonSEO, spoken there on automating SEO across large websites, and I judge the Global Search Awards. I mention this only because a site about computer science written by an anonymous domain is worth less than one where you can check who wrote it and what they know.

That background is the reason this site exists in the shape it does. Most reference material about computer science is either a textbook or a thin page assembled to catch a search query. I wanted something in between: pages that answer the question you actually typed, with a citation for anything that could be wrong and a visible date on anything that goes stale.

Where the degree fits

Manchester is a reasonable place to have studied this subject. The university built the Small-Scale Experimental Machine, the first computer to run a stored program, and Alan Turing worked there from 1948 until his death. Some of that history is on this site because I first encountered it as an undergraduate walking past the plaques.

A degree from 2005 is not a licence to pronounce on the state of the field in 2026, and I do not treat it as one. It means I can read a paper without being fooled by it, and I know which claims need a source. Everything on this site that could be wrong carries a citation so you can check it without taking my word for anything.

How these pages are made

Worth stating plainly, because it affects how much you should trust what you read here. I write and edit the pages. 44 of them carry my name.

Two automated jobs run against the site every week. One re-checks pages whose facts have a shelf life, ranking positions and prize statuses and the like, and proposes corrections. The other reads Search Console for the questions people actually typed to reach a page, and drafts FAQ entries only where the page already answers them. Neither writes anything from its own knowledge. The FAQ job must quote a sentence that already exists on the page, and the script checks that quote against the page before it saves anything. Both jobs open a pull request. Nothing reaches the site until I have read it and merged it.

The notes for AI agents page describes the same arrangement from the machine side, along with the markdown and open data surfaces.

Corrections

If something here is wrong, I would rather know. Every page shows the date its facts were last checked and lists its sources. Reach me on LinkedIn.