JeliaWhat it is
A catalogue of the technical books I have collected over a working life in infrastructure: 544 titles at the last count. .NET and C#, C and C++, security, Linux, networking, databases, DevOps, Python, PowerShell, assembly and a good deal more, down to a handful on retro computing and IRC.
You can browse by category, narrow by series or publisher, or search for a title. It is private. The catalogue sits behind its own sign-in, because it is a personal reference collection rather than a public one.
Who it is for
Me, first. Twenty-odd years in this trade leaves you with a lot of books, spread across drives and folders, half of them named by whatever the download called them. When I need the definitive guide to something at short notice, I want to find it in seconds, not remember which machine it was on.
It is also a working example of a problem almost every organisation has: a valuable collection of documents that nobody can find because nobody ever catalogued it.
What I built, and how
- Gathered everything in one place. Files collected over many years, brought into a single collection.
- Separated books from everything else. A folder of PDFs is never only books. Manuals, scans, paperwork and fragments were moved out, so the shelf holds books.
- Identified each book from its contents. A file name is not a title. The first attempt classified from file names and filed far too much in the wrong place, so it was rebuilt to open each book and check it against a public book database for its real title, author and publisher.
- Classified it. Every title sits in one category, with series and publisher as further filters, so the collection can be browsed as well as searched. Tidying the last stragglers is ongoing.
- Put a front end on it. A single page with filters down one side and the books grouped by series, quick to load and easy to scan.
- Kept it private and looked after. It runs on the same estate as everything else here, behind its own sign-in, with the same monitoring and backups.
What it shows I can do
Cataloguing sounds dull until you need something and cannot find it. The skills here are the ones that matter in any records or knowledge project:
- Turning a messy pile of files into structured data with consistent names and categories.
- Deciding on a classification people can actually use, and applying it consistently.
- Checking records against an authoritative source instead of trusting file names.
- Noticing when the first approach is wrong, and rebuilding it rather than patching around it.
- Putting a simple, fast interface on the result, with access limited to the people who should have it.
For a client, the same work applies to policy libraries, technical documentation, vendor contracts, old project folders, or the shared drive everyone is afraid to tidy.
Questions people ask
Can I browse it?
The catalogue is private, so it is not open to the public. This page describes how it was built.
Why not just use a folder and the search box?
Because search only finds what is named well, and most files are not. A catalogue knows what each item actually is, not what someone called it.
How do you keep it current?
New books go through the same identification and classification as the rest. The count on the page is whatever the catalogue holds today, not a figure typed in by hand.
Could you do this for our documents?
Yes. The first step is an inventory: what you have, where it lives, and what people most often fail to find. The structure follows from that.