Scope

Scope refers to the area within which something applies or is visible – for example a variable in program code, a permission of an app, or the agreed extent of a project. Whoever defines the scope draws a clear line between "included" and "not included".

Scope is English and means roughly “area” or “reach”. It always refers to a boundary: within this boundary something applies, outside it does not. In technology the word appears in three very different contexts. In programming it describes the places in the program’s text where a particular name may be used. For apps and online services it describes what a program is allowed to access, for example only your email address and not your photos. And in project management it describes what an assignment covers and what is explicitly not included. The common pattern is always the same: someone deliberately draws a line.

Why narrow boundaries limit damage

The most important reason for scopes is security. When you sign in to a third-party app with your Google account, a window appears with a list. It states what the app is allowed to see. These exact items are scopes. A note-taking app that only requests your name can, even in the event of a hacking attack, only lose your name. An app with full access to your inbox, on the other hand, would be a disaster.

Experts call this principle “least privilege”, meaning as few rights as possible. It also applies to AI systems. An assistant that can independently write emails and delete files is more dangerous than one that is only allowed to read. Many discussions about safe AI agents are, at their core, discussions about the right scope.

In programming, on the other hand, it is about order. Hundreds of people often work on a large project. Without boundaries, names would constantly overwrite each other. Narrow scopes ensure that a change in one place doesn’t break something in ten other places.

How the boundary is drawn

In program code, a scope usually arises through curly braces or through a function, that is, a self-contained step in the program. Everything created within these braces exists only there. As soon as the section has finished running, it disappears again. A distinction is made between local names, which only apply within a small section, and global names, which are visible throughout the entire program. Experienced developers avoid global names because no one can keep track of who is changing them anymore.

An image for this: a scope is like a room in a shared flat. What’s in your room belongs to you and bothers no one. What’s in the kitchen concerns everyone. And whoever puts everything in the kitchen creates chaos.

With online services this works differently on a technical level. There, scopes are passed as short text keys, such as “read” for reading or “write” for writing. The service then issues a digital credential in which exactly these keys are entered. Every subsequent request is checked against this credential. If the matching entry is missing, the request is rejected.

Scope in projects, contracts, and news

Outside of code, you mainly encounter this term in project management. There, the scope is stated in a document describing the extent of the assignment. When a client keeps adding new wishes during the work, this is called “scope creep”, meaning gradual expansion. This is one of the most common reasons why software projects finish late and end up too expensive.

In business news, scope also comes up in the context of audits and regulation. It might then be said that a business division does “not fall within the scope” of an investigation. In the EU’s law on artificial intelligence, too, the central question is which systems even fall within it. A common misconception is to confuse scope with objective. The objective states what something is meant to achieve. The scope merely states how far the boundary extends.

Subscribe free. Unsubscribe the second it sucks.

High-signal news across AI, business, UX, and tech. Every morning.